idnits 2.17.1 draft-ietf-appsawg-rfc7001bis-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document obsoletes RFC7001, but the abstract doesn't seem to mention this, which it should. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 3, 2015) is 3242 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'CFWS' is mentioned on line 501, but not defined == Missing Reference: 'RFC7208' is mentioned on line 806, but not defined -- Obsolete informational reference (is this intentional?): RFC 4870 (ref. 'DOMAINKEYS') (Obsoleted by RFC 4871) -- Obsolete informational reference (is this intentional?): RFC 5226 (ref. 'IANA-CONSIDERATIONS') (Obsoleted by RFC 8126) -- Obsolete informational reference (is this intentional?): RFC 3501 (ref. 'IMAP') (Obsoleted by RFC 9051) -- Obsolete informational reference (is this intentional?): RFC 7410 (ref. 'PTYPES-REGISTRY') (Obsoleted by RFC 7601) -- Obsolete informational reference (is this intentional?): RFC 5451 (Obsoleted by RFC 7001) -- Obsolete informational reference (is this intentional?): RFC 6577 (Obsoleted by RFC 7001) -- Obsolete informational reference (is this intentional?): RFC 7001 (Obsoleted by RFC 7601) Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 9 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Individual submission M. Kucherawy 3 Internet-Draft June 3, 2015 4 Obsoletes: 7001, 7410 5 (if approved) 6 Intended status: Standards Track 7 Expires: December 5, 2015 9 Message Header Field for Indicating Message Authentication Status 10 draft-ietf-appsawg-rfc7001bis-10 12 Abstract 14 This document specifies a message header field called Authentication- 15 Results for use with electronic mail messages to indicate the results 16 of message authentication efforts. Any receiver-side software, such 17 as mail filters or Mail User Agents (MUAs), can use this header field 18 to relay that information in a convenient and meaningful way to users 19 or to make sorting and filtering decisions. 21 Status of This Memo 23 This Internet-Draft is submitted in full conformance with the 24 provisions of BCP 78 and BCP 79. 26 Internet-Drafts are working documents of the Internet Engineering 27 Task Force (IETF). Note that other groups may also distribute 28 working documents as Internet-Drafts. The list of current Internet- 29 Drafts is at http://datatracker.ietf.org/drafts/current/. 31 Internet-Drafts are draft documents valid for a maximum of six months 32 and may be updated, replaced, or obsoleted by other documents at any 33 time. It is inappropriate to use Internet-Drafts as reference 34 material or to cite them other than as "work in progress." 36 This Internet-Draft will expire on December 5, 2015. 38 Copyright Notice 40 Copyright (c) 2015 IETF Trust and the persons identified as the 41 document authors. All rights reserved. 43 This document is subject to BCP 78 and the IETF Trust's Legal 44 Provisions Relating to IETF Documents 45 (http://trustee.ietf.org/license-info) in effect on the date of 46 publication of this document. Please review these documents 47 carefully, as they describe your rights and restrictions with respect 48 to this document. Code Components extracted from this document must 49 include Simplified BSD License text as described in Section 4.e of 50 the Trust Legal Provisions and are provided without warranty as 51 described in the Simplified BSD License. 53 Table of Contents 55 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 56 1.1. Purpose . . . . . . . . . . . . . . . . . . . . . . . . . 5 57 1.2. Trust Boundary . . . . . . . . . . . . . . . . . . . . . . 6 58 1.3. Processing Scope . . . . . . . . . . . . . . . . . . . . . 6 59 1.4. Requirements . . . . . . . . . . . . . . . . . . . . . . . 7 60 1.5. Definitions . . . . . . . . . . . . . . . . . . . . . . . 7 61 1.5.1. Key Words . . . . . . . . . . . . . . . . . . . . . . 7 62 1.5.2. Security . . . . . . . . . . . . . . . . . . . . . . . 7 63 1.5.3. Email Architecture . . . . . . . . . . . . . . . . . . 8 64 1.5.4. Other Terms . . . . . . . . . . . . . . . . . . . . . 9 65 1.6. Trust Environment . . . . . . . . . . . . . . . . . . . . 9 66 2. Definition and Format of the Header Field . . . . . . . . . . 9 67 2.1. General Description . . . . . . . . . . . . . . . . . . . 9 68 2.2. Formal Definition . . . . . . . . . . . . . . . . . . . . 10 69 2.3. Property Types (ptypes) and Properties . . . . . . . . . . 12 70 2.4. The "policy" ptype . . . . . . . . . . . . . . . . . . . . 13 71 2.5. Authentication Identifier Field . . . . . . . . . . . . . 14 72 2.6. Version Tokens . . . . . . . . . . . . . . . . . . . . . . 15 73 2.7. Defined Methods and Result Values . . . . . . . . . . . . 16 74 2.7.1. DKIM and DomainKeys . . . . . . . . . . . . . . . . . 16 75 2.7.2. SPF and Sender ID . . . . . . . . . . . . . . . . . . 17 76 2.7.3. "iprev" . . . . . . . . . . . . . . . . . . . . . . . 19 77 2.7.4. SMTP AUTH . . . . . . . . . . . . . . . . . . . . . . 19 78 2.7.5. Other Registered Codes . . . . . . . . . . . . . . . . 21 79 2.7.6. Extension Methods . . . . . . . . . . . . . . . . . . 21 80 2.7.7. Extension Result Codes . . . . . . . . . . . . . . . . 22 81 3. The "iprev" Authentication Method . . . . . . . . . . . . . . 22 82 4. Adding the Header Field to a Message . . . . . . . . . . . . . 23 83 4.1. Header Field Position and Interpretation . . . . . . . . . 25 84 4.2. Local Policy Enforcement . . . . . . . . . . . . . . . . . 26 85 5. Removing Existing Header Fields . . . . . . . . . . . . . . . 26 86 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 27 87 6.1. The Authentication-Results Header Field . . . . . . . . . 27 88 6.2. "Email Authentication Methods" Registry Description . . . 28 89 6.3. "Email Authentication Methods" Registry Update . . . . . . 29 90 6.4. "Email Authentication Property Types" Registry . . . . . . 30 91 6.5. "Email Authentication Result Names" Description . . . . . 31 92 6.6. "Email Authentication Result Names" Update . . . . . . . . 32 93 6.7. SMTP Enhanced Status Codes . . . . . . . . . . . . . . . . 33 94 7. Security Considerations . . . . . . . . . . . . . . . . . . . 33 95 7.1. Forged Header Fields . . . . . . . . . . . . . . . . . . . 33 96 7.2. Misleading Results . . . . . . . . . . . . . . . . . . . . 35 97 7.3. Header Field Position . . . . . . . . . . . . . . . . . . 35 98 7.4. Reverse IP Query Denial-of-Service Attacks . . . . . . . . 35 99 7.5. Mitigation of Backscatter . . . . . . . . . . . . . . . . 35 100 7.6. Internal MTA Lists . . . . . . . . . . . . . . . . . . . . 36 101 7.7. Attacks against Authentication Methods . . . . . . . . . . 36 102 7.8. Intentionally Malformed Header Fields . . . . . . . . . . 36 103 7.9. Compromised Internal Hosts . . . . . . . . . . . . . . . . 36 104 7.10. Encapsulated Instances . . . . . . . . . . . . . . . . . . 36 105 7.11. Reverse Mapping . . . . . . . . . . . . . . . . . . . . . 37 106 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 37 107 8.1. Normative References . . . . . . . . . . . . . . . . . . . 37 108 8.2. Informative References . . . . . . . . . . . . . . . . . . 38 109 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 40 110 Appendix B. Legacy MUAs . . . . . . . . . . . . . . . . . . . . . 40 111 Appendix C. Authentication-Results Examples . . . . . . . . . . . 40 112 C.1. Trivial Case; Header Field Not Present . . . . . . . . . . 41 113 C.2. Nearly Trivial Case; Service Provided, but No 114 Authentication Done . . . . . . . . . . . . . . . . . . . 41 115 C.3. Service Provided, Authentication Done . . . . . . . . . . 42 116 C.4. Service Provided, Several Authentications Done, Single 117 MTA . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 118 C.5. Service Provided, Several Authentications Done, 119 Different MTAs . . . . . . . . . . . . . . . . . . . . . . 44 120 C.6. Service Provided, Multi-Tiered Authentication Done . . . . 46 121 C.7. Comment-Heavy Example . . . . . . . . . . . . . . . . . . 47 122 Appendix D. Operational Considerations about Message 123 Authentication . . . . . . . . . . . . . . . . . . . 48 124 Appendix E. Change Since RFC7001 . . . . . . . . . . . . . . . . 49 126 1. Introduction 128 This document describes a header field called Authentication-Results 129 for electronic mail messages that presents the results of a message 130 authentication effort in a machine-readable format. The intent of 131 the header field is to create a place to collect such data when 132 message authentication mechanisms are in use so that a Mail User 133 Agent (MUA) and downstream filters can make filtering decisions 134 and/or provide a recommendation to the user as to the validity of the 135 message's origin and possibly the safety and integrity of its 136 content. 138 This document revises the original definition found in [RFC5451] 139 based upon various authentication protocols in current use and 140 incorporates errata logged since the publication of the original 141 specification. 143 End users are not expected to be direct consumers of this header 144 field. This header field is intended for consumption by programs 145 that will then use such data or render it in a human-usable form. 147 This document specifies the format of this header field and discusses 148 the implications of its presence or absence. However, it does not 149 discuss how the data contained in the header field ought to be used, 150 such as what filtering decisions are appropriate or how an MUA might 151 render those results, as these are local policy and/or user interface 152 design questions that are not appropriate for this document. 154 At the time of publication of this document, the following are 155 published email authentication methods: 157 o Author Domain Signing Practices ([ADSP]) (Historic) 159 o SMTP Service Extension for Authentication ([AUTH]) 161 o DomainKeys Identified Mail Signatures ([DKIM]) 163 o Domain-based Message Authentication, Reporting and Conformance 164 ([DMARC]) 166 o Sender Policy Framework ([SPF]) 168 o reverse IP address name validation ("iprev", defined in Section 3) 170 o Require-Recipient-Valid-Since Header Field and SMTP Service 171 Extension ([RRVS]) 173 o S/MIME Signature Verification ([SMIME-REG]) 175 o Vouch By Reference ([VBR]) 177 o DomainKeys ([DOMAINKEYS]) (Historic) 179 o Sender ID ([SENDERID]) (Experimental) 181 There exist registries for tokens used within this header field that 182 refer to the specifications listed above. Section 6 describes the 183 registries and their contents, and specifies the process by which 184 entries are added or updated. It also updates the existing contents 185 to match the current states of these specifications. 187 This specification is not intended to be restricted to domain-based 188 authentication schemes, but the existing schemes in that family have 189 proven to be a good starting point for implementations. The goal is 190 to give current and future authentication schemes a common framework 191 within which to deliver their results to downstream agents and 192 discourage the creation of unique header fields for each. 194 Although SPF defined a header field called "Received-SPF" and the 195 historic DomainKeys defined one called "DomainKey-Status" for this 196 purpose, those header fields are specific to the conveyance of their 197 respective results only and thus are insufficient to satisfy the 198 requirements enumerated below. In addition, many SPF implementations 199 have adopted the header field specified here at least as an option, 200 and DomainKeys has been obsoleted by DKIM. 202 1.1. Purpose 204 The header field defined in this document is expected to serve 205 several purposes: 207 1. Convey the results of various message authentication checks, 208 which are applied by upstream filters and Mail Transfer Agents 209 (MTAs) and then passed to MUAs and downstream filters within the 210 same "trust domain". Such agents might wish to render those 211 results to end users or to use those data to apply more or less 212 stringent content checks based on authentication results; 214 2. Provide a common location within a message for this data; 216 3. Create an extensible framework for reporting new authentication 217 methods as they emerge. 219 In particular, the mere presence of this header field does not mean 220 its contents are valid. Rather, the header field is reporting 221 assertions made by one or more authentication schemes (supposedly) 222 applied somewhere upstream. For an MUA or downstream filter to treat 223 the assertions as actually valid, there must be an assessment of the 224 trust relationship among such agents, the validating MTA, and the 225 mechanism for conveying the information. 227 1.2. Trust Boundary 229 This document makes several references to the "trust boundary" of an 230 administrative management domain (ADMD). Given the diversity among 231 existing mail environments, a precise definition of this term isn't 232 possible. 234 Simply put, a transfer from the producer of the header field to the 235 consumer must occur within a context that permits the consumer to 236 treat assertions by the producer as being reliable and accurate 237 (trustworthy). How this trust is obtained is outside the scope of 238 this document. It is entirely a local matter. 240 Thus, this document defines a "trust boundary" as the delineation 241 between "external" and "internal" entities. Services that are 242 internal -- within the trust boundary -- are provided by the ADMD's 243 infrastructure for its users. Those that are external are outside of 244 the authority of the ADMD. By this definition, hosts that are within 245 a trust boundary are subject to the ADMD's authority and policies, 246 independent of their physical placement or their physical operation. 247 For example, a host within a trust boundary might actually be 248 operated by a remote service provider and reside physically within 249 its data center. 251 It is possible for a message to be evaluated inside a trust boundary 252 but then depart and re-enter the trust boundary. An example might be 253 a forwarded message such as a message/rfc822 attachment (see 254 Multipurpose Internet Mail Extensions [MIME]) or one that is part of 255 a multipart/digest. The details reported by this field cannot be 256 trusted in that case. Thus, this field found within one of those 257 media types is typically ignored. 259 1.3. Processing Scope 261 The content of this header field is meant to convey to message 262 consumers that authentication work on the message was already done 263 within its trust boundary, and those results are being presented. It 264 is not intended to provide message parameters to consumers so that 265 they can perform authentication protocols on their own. 267 1.4. Requirements 269 This document establishes no new requirements on existing protocols 270 or servers. 272 In particular, this document establishes no requirement on MTAs to 273 reject or filter arriving messages that do not pass authentication 274 checks. The data conveyed by the specified header field's contents 275 are for the information of MUAs and filters and are to be used at 276 their discretion. 278 1.5. Definitions 280 This section defines various terms used throughout this document. 282 1.5.1. Key Words 284 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 285 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 286 document are to be interpreted as described in [KEYWORDS]. 288 1.5.2. Security 290 "Guidelines for Writing RFC Text on Security Considerations" 291 ([SECURITY]) discusses authentication and authorization and the 292 conflation of the two concepts. The use of those terms within the 293 context of recent message security work has given rise to slightly 294 different definitions, and this document reflects those current 295 usages, as follows: 297 o "Authorization" is the establishment of permission to use a 298 resource or represent an identity. In this context, authorization 299 indicates that a message from a particular ADMD arrived via a 300 route the ADMD has explicitly approved. 302 o "Authentication" is the assertion of validity of a piece of data 303 about a message (such as the sender's identity) or the message in 304 its entirety. 306 As examples: SPF and Sender ID are authorization mechanisms in that 307 they express a result that shows whether or not the ADMD that 308 apparently sent the message has explicitly authorized the connecting 309 Simple Mail Transfer Protocol ([SMTP]) client to relay messages on 310 its behalf, but they do not actually validate any other property of 311 the message itself. By contrast, DKIM is agnostic as to the routing 312 of a message but uses cryptographic signatures to authenticate 313 agents, assign (some) responsibility for the message (which implies 314 authorization), and ensure that the listed portions of the message 315 were not modified in transit. Since the signatures are not tied to 316 SMTP connections, they can be added by either the ADMD of origin, 317 intermediate ADMDs (such as a mailing list server), other handling 318 agents, or any combination. 320 Rather than create a separate header field for each class of 321 solution, this proposal groups them both into a single header field. 323 1.5.3. Email Architecture 325 o A "border MTA" is an MTA that acts as a gateway between the 326 general Internet and the users within an organizational boundary. 327 (See also Section 1.2.) 329 o A "delivery MTA" (or Mail Delivery Agent or MDA) is an MTA that 330 actually enacts delivery of a message to a user's inbox or other 331 final delivery. 333 o An "intermediate MTA" is any MTA that is not a delivery MTA and is 334 also not the first MTA to handle the message. 336 The following diagram illustrates the flow of mail among these 337 defined components. See Internet Mail Architecture [EMAIL-ARCH] for 338 further discussion on general email system architecture, which 339 includes detailed descriptions of these components, and Appendix D of 340 this document for discussion about the common aspects of email 341 authentication in current environments. 343 +-----+ +-----+ +------------+ 344 | MUA |-->| MSA |-->| Border MTA | 345 +-----+ +-----+ +------------+ 346 | 347 | 348 V 349 +----------+ 350 | Internet | 351 +----------+ 352 | 353 | 354 V 355 +-----+ +-----+ +------------------+ +------------+ 356 | MUA |<--| MDA |<--| Intermediate MTA |<--| Border MTA | 357 +-----+ +-----+ +------------------+ +------------+ 359 Generally, it is assumed that the work of applying message 360 authentication schemes takes place at a border MTA or a delivery MTA. 361 This specification is written with that assumption in mind. However, 362 there are some sites at which the entire mail infrastructure consists 363 of a single host. In such cases, such terms as "border MTA" and 364 "delivery MTA" might well apply to the same machine or even the very 365 same agent. It is also possible that some message authentication 366 tests could take place on an intermediate MTA. Although this 367 document doesn't specifically describe such cases, they are not meant 368 to be excluded. 370 1.5.4. Other Terms 372 In this document, the term "producer" refers to any component that 373 adds this header field to messages it is handling, and "consumer" 374 refers to any component that identifies, extracts, and parses the 375 header field to use as part of a handling decision. 377 1.6. Trust Environment 379 This header field permits one or more message validation mechanisms 380 to communicate output to one or more separate assessment mechanisms. 381 These mechanisms operate within a unified trust boundary that defines 382 an Administrative Management Domain (ADMD). An ADMD contains one or 383 more entities that perform validation and generate the header field 384 and one or more that consume it for some type of assessment. The 385 field often contains no integrity or validation mechanism of its own, 386 so its presence must be trusted implicitly. Hence, valid use of the 387 header field requires removing any occurrences of it that are present 388 when the message enters the ADMD. This ensures that later 389 occurrences have been added within the trust boundary of the ADMD. 391 The authserv-id token defined in Section 2.2 can be used to reference 392 an entire ADMD or a specific validation engine within an ADMD. 393 Although the labeling scheme is left as an operational choice, some 394 guidance for selecting a token is provided in later sections of this 395 document. 397 2. Definition and Format of the Header Field 399 This section gives a general overview of the format of the header 400 field being defined and then provides more formal specification. 402 2.1. General Description 404 The header field specified here is called Authentication-Results. It 405 is a Structured Header Field as defined in Internet Message Format 406 ([MAIL]), and thus all of the related definitions in that document 407 apply. 409 This header field is added at the top of the message as it transits 410 MTAs that do authentication checks, so some idea of how far away the 411 checks were done can be inferred. It is therefore considered to be a 412 trace field as defined in [MAIL], and thus all of the related 413 definitions in that document apply. 415 The value of the header field (after removing comments) consists of 416 an authentication identifier, an optional version, and then a series 417 of statements and supporting data. The statements are of the form 418 "method=result" and indicate which authentication method(s) were 419 applied and their respective results. For each such statement, the 420 supporting data can include a "reason" string and one or more 421 "property=value" statements indicating which message properties were 422 evaluated to reach that conclusion. 424 The header field can appear more than once in a single message, more 425 than one result can be represented in a single header field, or a 426 combination of these can be applied. 428 2.2. Formal Definition 430 Formally, the header field is specified as follows using Augmented 431 Backus-Naur Form ([ABNF]): 433 authres-header = "Authentication-Results:" [CFWS] authserv-id 434 [ CFWS authres-version ] 435 ( no-result / 1*resinfo ) [CFWS] CRLF 437 authserv-id = value 438 ; see below for a description of this element 440 authres-version = 1*DIGIT [CFWS] 441 ; indicates which version of this specification is in use; 442 ; this specification is version "1", and the absence of a 443 ; version implies this version of the specification 445 no-result = [CFWS] ";" [CFWS] "none" 446 ; the special case of "none" is used to indicate that no 447 ; message authentication was performed 449 resinfo = [CFWS] ";" methodspec [ CFWS reasonspec ] 450 *( CFWS propspec ) 452 methodspec = [CFWS] method [CFWS] "=" [CFWS] result 453 ; indicates which authentication method was evaluated 454 ; and what its output was 456 reasonspec = "reason" [CFWS] "=" [CFWS] value 457 ; a free-form comment on the reason the given result 458 ; was returned 460 propspec = ptype [CFWS] "." [CFWS] property [CFWS] "=" pvalue 461 ; an indication of which properties of the message 462 ; were evaluated by the authentication scheme being 463 ; applied to yield the reported result 465 method = Keyword [ [CFWS] "/" [CFWS] method-version ] 466 ; a method indicates which method's result is 467 ; represented by "result", and is one of the methods 468 ; explicitly defined as valid in this document 469 ; or is an extension method as defined below 471 method-version = 1*DIGIT [CFWS] 472 ; indicates which version of the method specification is 473 ; in use, corresponding to the matching entry in the IANA 474 ; "Email Authentication Methods" registry; a value of "1" 475 ; is assumed if this version string is absent 477 result = Keyword 478 ; indicates the results of the attempt to authenticate 479 ; the message; see below for details 481 ptype = Keyword 482 ; indicates whether the property being evaluated was 483 ; a parameter to an [SMTP] command, was a value taken 484 ; from a message header field, was some property of 485 ; the message body, or was some other property evaluated by 486 ; the receiving MTA; expected to be one of the "property 487 ; types" explicitly defined as valid, or an extension 488 ; ptype, as defined below 490 property = special-smtp-verb / Keyword 491 ; indicates more specifically than "ptype" what the 492 ; source of the evaluated property is; the exact meaning 493 ; is specific to the method whose result is being reported 494 ; and is defined more clearly below 496 special-smtp-verb = "mailfrom" / "rcptto" 497 ; special cases of [SMTP] commands that are made up 498 ; of multiple words 500 pvalue = [CFWS] ( value / [ [ local-part ] "@" ] domain-name ) 501 [CFWS] 502 ; the value extracted from the message property defined 503 ; by the "ptype.property" construction 505 "local-part" is defined in Section 3.4.1 of [MAIL], and "CFWS" is 506 defined in Section 3.2.2 of [MAIL]. 508 "Keyword" is defined in Section 4.1.2 of [SMTP]. 510 The "value" is as defined in Section 5.1 of [MIME]. 512 The "domain-name" is as defined in Section 3.5 of [DKIM]. 514 The "Keyword" used in "result" above is further constrained by the 515 necessity of being enumerated in Section 2.7. 517 See Section 2.5 for a description of the authserv-id element. 519 If the value portion of a "pvalue" construction identifies something 520 intended to be an e-mail identity, then it MUST use the right hand 521 portion of that ABNF definition. 523 The list of commands eligible for use with the "smtp" ptype can be 524 found in Section 4.1 of [SMTP]. 526 The "propspec" may be omitted if, for example, the method was unable 527 to extract any properties to do its evaluation yet has a result to 528 report. 530 Where an SMTP command name is being reported as a "property", the 531 agent generating the header field represents that command by 532 converting it to lowercase and dropping any spaces (e.g., "MAIL FROM" 533 becomes "mailfrom", "RCPT TO" becomes "rcptto", etc.). 535 A "ptype" value of "policy" indicates a policy decision about the 536 message not specific to a property of the message that could be 537 extracted. See Section 2.4 for details. 539 Examples of complete messages using this header field can be found in 540 Appendix C. 542 2.3. Property Types (ptypes) and Properties 544 The "ptype" in the ABNF above indicates the general type of property 545 being described by the result being reported, upon which the reported 546 result was based. Coupled with the "property", which is more 547 specific, they indicate from which particular part of the message the 548 reported data were extracted. 550 Combinations of ptypes and properties are registered and described in 551 the "Email Authentication Methods" registry, coupled with the 552 authentication methods with which they are used. This is further 553 described in Section 6. 555 Legal values of "ptype" are as defined in the IANA "Email 556 Authentication Property Types" registry, created by 557 [PTYPES-REGISTRY]. The initial values and what they typically 558 indicate are as follows, copied from [RFC7001]: 560 body: Information that was extracted from the body of the message. 561 This might be an arbitrary string of bytes, a hash of a string of 562 bytes, a Uniform Resource Identifier, or some other content of 563 interest. The "property" is an indication of where within the 564 message body the extracted content was found, and can indicate an 565 offset, identify a MIME part, etc. 567 header: Indicates information that was extracted from the header of 568 the message. This might be the value of a header field or some 569 portion of a header field. The "property" gives a more precise 570 indication of the place in the header from which the extraction 571 took place. 573 policy: A local policy mechanism was applied that augments or 574 overrides the result returned by the authentication mechanism. 575 (See Section 2.4.) 577 smtp: Indicates information that was extracted from an SMTP command 578 that was used to relay the message. The "property" indicates 579 which SMTP command included the extracted content as a parameter. 581 Results reported using unknown ptypes MUST NOT be used in making 582 handling decisions. They can be safely ignored by consumers. 584 Entries in the "Email Authentication Methods" registry can define 585 properties that deviate from these definitions when appropriate. 586 Such deviations need to be clear in the registry and/or in the 587 defining document. See Section 2.7.1 for an example. 589 2.4. The "policy" ptype 591 A special ptype value of "policy" is also defined. This ptype is 592 provided to indicate that some local policy mechanism was applied 593 that augments or even replaces (i.e., overrides) the result returned 594 by the authentication mechanism. The property and value in this case 595 identify the local policy that was applied and the result it 596 returned. 598 For example, a DKIM signature is not required to include the Subject 599 header field in the set of fields that are signed. An ADMD receiving 600 such a message might decide that such a signature is unacceptable, 601 even if it passes, because the content of the Subject header field 602 could be altered post-signing without invalidating the signature. 603 Such an ADMD could replace the DKIM "pass" result with a "policy" 604 result and then also include the following in the corresponding 605 Authentication-Result field: 607 ... dkim=fail policy.dkim-rules=unsigned-subject ... 609 In this case, the property is "dkim-rules", indicating some local 610 check by that name took place and that check returned a result of 611 "unsigned-subject". These are arbitrary names selected by (and 612 presumably used within) the ADMD making use of them, so they are not 613 normally registered with IANA or otherwise specified apart from 614 setting syntax restrictions that allow for easy parsing within the 615 rest of the header field. 617 This ptype existed in the original specification for this header 618 field, but without a complete description or example of intended use. 619 As a result, it has not seen any practical use to date that matches 620 its intended purpose. These added details are provided to guide 621 implementers toward proper use. 623 2.5. Authentication Identifier Field 625 Every Authentication-Results header field has an authentication 626 service identifier field (authserv-id above). Specifically, this is 627 any string intended to identify the authentication service within the 628 ADMD that conducted authentication checks on the message. This 629 identifier is intended to be machine-readable and not necessarily 630 meaningful to users. 632 Since agents consuming this field will use this identifier to 633 determine whether its contents are of interest (and are safe to use), 634 the uniqueness of the identifier MUST be guaranteed by the ADMD that 635 generates it and MUST pertain to that ADMD. MUAs or downstream 636 filters SHOULD use this identifier to determine whether or not the 637 data contained in an Authentication-Results header field ought to be 638 used or ignored. 640 For simplicity and scalability, the authentication service identifier 641 SHOULD be a common token used throughout the ADMD. Common practice 642 is to use the DNS domain name used by or within that ADMD, sometimes 643 called the "organizational domain", but this is not strictly 644 necessary. 646 For tracing and debugging purposes, the authentication identifier can 647 instead be the specific hostname of the MTA performing the 648 authentication check whose result is being reported. Moreover, some 649 implementations define a substructure to the identifier; these are 650 outside of the scope of this specification. 652 Note, however, that using a local, relative identifier like a flat 653 hostname, rather than a hierarchical and globally unique ADMD 654 identifier like a DNS domain name, makes configuration more difficult 655 for large sites. The hierarchical identifier permits aggregating 656 related, trusted systems together under a single, parent identifier, 657 which in turn permits assessing the trust relationship with a single 658 reference. The alternative is a flat namespace requiring 659 individually listing each trusted system. Since consumers will use 660 the identifier to determine whether to use the contents of the header 661 field: 663 o Changes to the identifier impose a large, centralized 664 administrative burden. 666 o Ongoing administrative changes require constantly updating this 667 centralized table, making it difficult to ensure that an MUA or 668 downstream filter will have access to accurate information for 669 assessing the usability of the header field's content. In 670 particular, consumers of the header field will need to know not 671 only the current identifier(s) in use but previous ones as well to 672 account for delivery latency or later re-assessment of the header 673 field's contents. 675 Examples of valid authentication identifiers are "example.com", 676 "mail.example.org", "ms1.newyork.example.com", and "example-auth". 678 2.6. Version Tokens 680 The grammar above provides for the optional inclusion of versions on 681 both the header field itself (attached to the authserv-id token) and 682 on each of the methods being reported. The method version refers to 683 the method itself, which is specified in the documents describing 684 those methods, while the authserv-id version refers to this document 685 and thus the syntax of this header field. 687 The purpose of including these is to avoid misinterpretation of the 688 results. That is, if a parser finds a version after an authserv-id 689 that it does not explicitly know, it can immediately discontinue 690 trying to parse since what follows might not be in an expected 691 format. For a method version, the parser SHOULD ignore a method 692 result if the version is not supported in case the semantics of the 693 result have a different meaning than what is expected. For example, 694 if a hypothetical DKIM version 2 yielded a "pass" result for 695 different reasons than version 1 does, a consumer of this field might 696 not want to use the altered semantics. Allowing versions in the 697 syntax is a way to indicate this and let the consumer of the header 698 field decide. 700 2.7. Defined Methods and Result Values 702 Each individual authentication method returns one of a set of 703 specific result values. The subsections below provide references to 704 the documents defining the authentication methods specifically 705 supported by this document, and their corresponding result values. 706 Verifiers SHOULD use these values as described below. New methods 707 not specified in this document, but intended to be supported by the 708 header field defined here, MUST include a similar result table either 709 in their defining documents or in supplementary ones. 711 2.7.1. DKIM and DomainKeys 713 DKIM is represented by the "dkim" method and is defined in [DKIM]. 714 DomainKeys is defined in [DOMAINKEYS] and is represented by the 715 "domainkeys" method. 717 Section 3.8 of [DOMAINKEYS] enumerates some possible results of a 718 DomainKeys evaluation. Those results are not used when generating 719 this header field; rather, the results returned are listed below. 721 A signature is "acceptable to the ADMD" if it passes local policy 722 checks (or there are no specific local policy checks). For example, 723 an ADMD policy might require that the signature(s) on the message be 724 added using the DNS domain present in the From header field of the 725 message, thus making third-party signatures unacceptable even if they 726 verify. 728 Both DKIM and DomainKeys use the same result set, as follows: 730 none: The message was not signed. 732 pass: The message was signed, the signature or signatures were 733 acceptable to the ADMD, and the signature(s) passed verification 734 tests. 736 fail: The message was signed and the signature or signatures were 737 acceptable to the ADMD, but they failed the verification test(s). 739 policy: The message was signed, but some aspect of the signature or 740 signatures was not acceptable to the ADMD. 742 neutral: The message was signed, but the signature or signatures 743 contained syntax errors or were not otherwise able to be 744 processed. This result is also used for other failures not 745 covered elsewhere in this list. 747 temperror: The message could not be verified due to some error that 748 is likely transient in nature, such as a temporary inability to 749 retrieve a public key. A later attempt may produce a final 750 result. 752 permerror: The message could not be verified due to some error that 753 is unrecoverable, such as a required header field being absent. A 754 later attempt is unlikely to produce a final result. 756 DKIM results are reported using a ptype of "header". The property, 757 however, represents one of the tags found in the DKIM-Signature 758 header field rather than a distinct header field. For example, the 759 ptype-property combination "header.d" refers to the content of the 760 "d" (signing domain) tag from within the signature header field, and 761 not a distinct header field called "d". 763 The ability to report different DKIM results for a multiply-signed 764 message is described in [RFC6008]. 766 [DKIM] advises that if a message fails verification, it is to be 767 treated as an unsigned message. A report of "fail" here permits the 768 receiver of the report to decide how to handle the failure. A report 769 of "neutral" or "none" preempts that choice, ensuring the message 770 will be treated as if it had not been signed. 772 Section 3.1 of [DOMAINKEYS] describes a process by which the sending 773 address of the message is determined. DomainKeys results are thus 774 reported along with the signing domain name, the sending address of 775 the message, and the name of the header field from which the latter 776 was extracted. This means that a DomainKeys result includes a ptype- 777 property combination of "header.d", plus one of "header.from" and 778 "header.sender". The sending address extracted from the header is 779 included with any [MAIL]-style comments removed; moreover, the local- 780 part of the address and the "@" character are removed if it has not 781 been authenticated in some way. 783 2.7.2. SPF and Sender ID 785 SPF and Sender ID use the "spf" and "sender-id" method names, 786 respectively. The result values for SPF are defined in Section 2.6 787 of [SPF], and those definitions are included here by reference: 789 +-----------+--------------------------------+ 790 | Code | Meaning | 791 +-----------+--------------------------------+ 792 | none | [RFC7208], Section 2.6.1 | 793 +-----------+--------------------------------+ 794 | pass | [RFC7208], Section 2.6.3 | 795 +-----------+--------------------------------+ 796 | fail | [RFC7208], Section 2.6.4 | 797 +-----------+--------------------------------+ 798 | softfail | [RFC7208], Section 2.6.5 | 799 +-----------+--------------------------------+ 800 | policy | [this RFC], Section 2.4 | 801 +-----------+--------------------------------+ 802 | neutral | [RFC7208], Section 2.6.2 | 803 +-----------+--------------------------------+ 804 | temperror | [RFC7208], Section 2.6.6 | 805 +-----------+--------------------------------+ 806 | permerror | [RFC7208], Section 2.6.7 | 807 +-----------+--------------------------------+ 809 These result codes are used in the context of this specification to 810 reflect the result returned by the component conducting SPF 811 evaluation. 813 For SPF, the ptype used is "smtp", and the property is either 814 "mailfrom" or "helo", since those values are the ones SPF can 815 evaluate. (If the SMTP client issued the EHLO command instead of 816 HELO, the property used is "helo".) 818 The "sender-id" method is described in [SENDERID]. For this method, 819 the ptype used is "header" and the property will be the name of the 820 header field from which the Purported Responsible address (see [PRA]) 821 was extracted, namely one of "Resent-Sender", "Resent-From", 822 "Sender", or "From". 824 The results for Sender ID are listed and described in Section 4.2 of 825 [SENDERID], but for the purposes of this specification, the SPF 826 definitions enumerated above are used instead. Also, [SENDERID] 827 specifies result codes that use mixed case, but they are typically 828 used all lowercase in this context. 830 For both methods, an additional result of "policy" is defined, which 831 means the client was authorized to inject or relay mail on behalf of 832 the sender's DNS domain according to the authentication method's 833 algorithm, but local policy dictates that the result is unacceptable. 834 For example, "policy" might be used if SPF returns a "pass" result, 835 but a local policy check matches the sending DNS domain to one found 836 in an explicit list of unacceptable DNS domains (e.g., spammers). 838 If the retrieved sender policies used to evaluate SPF and Sender ID 839 do not contain explicit provisions for authenticating the local-part 840 (see Section 3.4.1 of [MAIL]) of an address, the "pvalue" reported 841 along with results for these mechanisms SHOULD NOT include the local- 842 part or the following "@" character. 844 2.7.3. "iprev" 846 The result values used by the "iprev" method, defined in Section 3, 847 are as follows: 849 pass: The DNS evaluation succeeded, i.e., the "reverse" and 850 "forward" lookup results were returned and were in agreement. 852 fail: The DNS evaluation failed. In particular, the "reverse" and 853 "forward" lookups each produced results, but they were not in 854 agreement, or the "forward" query completed but produced no 855 result, e.g., a DNS RCODE of 3, commonly known as NXDOMAIN, or an 856 RCODE of 0 (NOERROR) in a reply containing no answers, was 857 returned. 859 temperror: The DNS evaluation could not be completed due to some 860 error that is likely transient in nature, such as a temporary DNS 861 error, e.g., a DNS RCODE of 2, commonly known as SERVFAIL, or 862 other error condition resulted. A later attempt may produce a 863 final result. 865 permerror: The DNS evaluation could not be completed because no PTR 866 data are published for the connecting IP address, e.g., a DNS 867 RCODE of 3, commonly known as NXDOMAIN, or an RCODE of 0 (NOERROR) 868 in a reply containing no answers, was returned. This prevented 869 completion of the evaluation. A later attempt is unlikely to 870 produce a final result. 872 There is no "none" for this method since any TCP connection 873 delivering email has an IP address associated with it, so some kind 874 of evaluation will always be possible. 876 The result is reported using a ptype of "policy" (as this is not part 877 of any established protocol) and a property of "iprev". 879 For discussion of the format of DNS replies, see "Domain Names - 880 Implementation and Specification" ([DNS]). 882 2.7.4. SMTP AUTH 884 SMTP AUTH (defined in [AUTH]) is represented by the "auth" method. 885 Its result values are as follows: 887 none: SMTP authentication was not attempted. 889 pass: The SMTP client authenticated to the server reporting the 890 result using the protocol described in [AUTH]. 892 fail: The SMTP client attempted to authenticate to the server using 893 the protocol described in [AUTH] but was not successful (such as 894 providing a valid identity but an incorrect password). 896 temperror: The SMTP client attempted to authenticate using the 897 protocol described in [AUTH] but was not able to complete the 898 attempt due to some error that is likely transient in nature, such 899 as a temporary directory service lookup error. A later attempt 900 may produce a final result. 902 permerror: The SMTP client attempted to authenticate using the 903 protocol described in [AUTH] but was not able to complete the 904 attempt due to some error that is likely not transient in nature, 905 such as a permanent directory service lookup error. A later 906 attempt is not likely to produce a final result. 908 The result of AUTH is reported using a ptype of "smtp" and a property 909 of either: 911 o "auth", in which case the value is the authorization identity 912 generated by the exchange initiated by the AUTH command; or 914 o "mailfrom", in which case the value is the mailbox identified by 915 the AUTH parameter used with the MAIL FROM command. 917 If both identities are available, both can be reported. For example, 918 consider this command issued by a client that has completed session 919 authentication with the AUTH command resulting in an authorized 920 identity of "client@c.example": 922 MAIL FROM: AUTH= 924 This could result in a resinfo construction like so: 926 ; auth=pass smtp.auth=client@c.example smtp.mailfrom=bob@b.example 928 Note that in all cases other than "pass", the message was sent by an 929 unauthenticated client. All non-"pass" cases SHOULD thus be treated 930 as equivalent with respect to this method. 932 2.7.5. Other Registered Codes 934 Result codes were also registered in other RFCs as follows: 936 o Vouch By Reference (in [AR-VBR], represented by "vbr"); 938 o Authorized Third-Party Signatures (in [ATPS], represented by 939 "dkim-atps"); 941 o Author Domain Signing Practices (in [ADSP], represented by "dkim- 942 adsp"); 944 o Require-Recipient-Valid-Since (in [RRVS], represented by "rrvs"); 946 o S/MIME (in [SMIME-REG], represented by "smime"). 948 2.7.6. Extension Methods 950 Additional authentication method identifiers (extension methods) may 951 be defined in the future by later revisions or extensions to this 952 specification. These method identifiers are registered with the 953 Internet Assigned Numbers Authority (IANA) and, preferably, published 954 in an RFC. See Section 6 for further details. 956 Extension methods can be defined for the following reasons: 958 1. To allow additional information from new authentication systems 959 to be communicated to MUAs or downstream filters. The names of 960 such identifiers ought to reflect the name of the method being 961 defined but ought not be needlessly long. 963 2. To allow the creation of "sub-identifiers" that indicate 964 different levels of authentication and differentiate between 965 their relative strengths, e.g., "auth1-weak" and "auth1-strong". 967 Authentication method implementers are encouraged to provide adequate 968 information, via message header field comments if necessary, to allow 969 an MUA developer to understand or relay ancillary details of 970 authentication results. For example, if it might be of interest to 971 relay what data was used to perform an evaluation, such information 972 could be relayed as a comment in the header field, such as: 974 Authentication-Results: example.com; 975 foo=pass bar.baz=blob (2 of 3 tests OK) 977 Experimental method identifiers MUST only be used within ADMDs that 978 have explicitly consented to use them. These method identifiers and 979 the parameters associated with them are not documented in RFCs. 981 Therefore, they are subject to change at any time and not suitable 982 for production use. Any MTA, MUA, or downstream filter intended for 983 production use SHOULD ignore or delete any Authentication-Results 984 header field that includes an experimental (unknown) method 985 identifier. 987 2.7.7. Extension Result Codes 989 Additional result codes (extension results) might be defined in the 990 future by later revisions or extensions to this specification. 991 Result codes MUST be registered with the Internet Assigned Numbers 992 Authority (IANA) and preferably published in an RFC. See Section 6 993 for further details. 995 Experimental results MUST only be used within ADMDs that have 996 explicitly consented to use them. These results and the parameters 997 associated with them are not formally documented. Therefore, they 998 are subject to change at any time and not suitable for production 999 use. Any MTA, MUA, or downstream filter intended for production use 1000 SHOULD ignore or delete any Authentication-Results header field that 1001 includes an extension result. 1003 3. The "iprev" Authentication Method 1005 This section defines an additional authentication method called 1006 "iprev". 1008 "iprev" is an attempt to verify that a client appears to be valid 1009 based on some DNS queries, which is to say that the IP address is 1010 explicitly associated with a domain name. Upon receiving a session 1011 initiation of some kind from a client, the IP address of the client 1012 peer is queried for matching names (i.e., a number-to-name 1013 translation, also known as a "reverse lookup" or a "PTR" record 1014 query). Once that result is acquired, a lookup of each of the names 1015 (i.e., a name-to-number translation, or an "A" or "AAAA" record 1016 query) thus retrieved is done. The response to this second check 1017 will typically result in at least one mapping back to the client's IP 1018 address. 1020 Expressed as an algorithm: If the client peer's IP address is I, the 1021 list of names to which I maps (after a "PTR" query) is the set N, and 1022 the union of IP addresses to which each member of N maps (after 1023 corresponding "A" and "AAAA" queries) is L, then this test is 1024 successful if I is an element of L. 1026 Often an MTA receiving a connection that fails this test will simply 1027 reject the connection using the enhanced status code defined in 1028 [AUTH-ESC]. If an operator instead wishes to make this information 1029 available to downstream agents as a factor in handling decisions, it 1030 records a result in accordance with Section 2.7.3. 1032 The response to a PTR query could contain multiple names. To prevent 1033 heavy DNS loads, agents performing these queries MUST be implemented 1034 such that the number of names evaluated by generation of 1035 corresponding A or AAAA queries is limited so as not to be unduly 1036 taxing to the DNS infrastructure, though it MAY be configurable by an 1037 administrator. As an example, Section 4.6.4 of [SPF] chose a limit 1038 of 10 for its implementation of this algorithm. 1040 "DNS Extensions to Support IP Version 6" ([DNS-IP6]) discusses the 1041 query formats for the IPv6 case. 1043 There is some contention regarding the wisdom and reliability of this 1044 test. For example, in some regions, it can be difficult for this 1045 test ever to pass because the practice of arranging to match the 1046 forward and reverse DNS is infrequently observed. Therefore, the 1047 precise implementation details of how a verifier performs an "iprev" 1048 test are not specified here. The verifier MAY report a successful or 1049 failed "iprev" test at its discretion having done some kind of check 1050 of the validity of the connection's identity using DNS. It is 1051 incumbent upon an agent making use of the reported "iprev" result to 1052 understand what exactly that particular verifier is attempting to 1053 report. 1055 Extensive discussion of reverse DNS mapping and its implications can 1056 be found in "Considerations for the use of DNS Reverse Mapping" 1057 ([DNSOP-REVERSE]). In particular, it recommends that applications 1058 avoid using this test as a means of authentication or security. Its 1059 presence in this document is not an endorsement but is merely 1060 acknowledgment that the method remains common and provides the means 1061 to relay the results of that test. 1063 4. Adding the Header Field to a Message 1065 This specification makes no attempt to evaluate the relative 1066 strengths of various message authentication methods that may become 1067 available. The methods listed are an order-independent set; their 1068 sequence does not indicate relative strength or importance of one 1069 method over another. Instead, the MUA or downstream filter consuming 1070 this header field is to interpret the result of each method based on 1071 its own knowledge of what that method evaluates. 1073 Each "method" MUST refer to an authentication method declared in the 1074 IANA registry or an extension method as described in Section 2.7.6, 1075 and each "result" MUST refer to a result code declared in the IANA 1076 registry or an extension result code as defined in Section 2.7.7. 1078 See Section 6 for further information about the registered methods 1079 and result codes. 1081 An MTA compliant with this specification adds this header field 1082 (after performing one or more message authentication tests) to 1083 indicate which MTA or ADMD performed the test, which test got 1084 applied, and what the result was. If an MTA applies more than one 1085 such test, it adds this header field either once per test or once 1086 indicating all of the results. An MTA MUST NOT add a result to an 1087 existing header field. 1089 An MTA MAY add this header field containing only the authentication 1090 identifier portion and the "none" token (see Section 2.2) to indicate 1091 explicitly that no message authentication schemes were applied prior 1092 to delivery of this message. 1094 An MTA adding this header field has to take steps to identify it as 1095 legitimate to the MUAs or downstream filters that will ultimately 1096 consume its content. One process to do so is described in Section 5. 1097 Further measures may be necessary in some environments. Some 1098 possible solutions are enumerated in Section 7.1. This document does 1099 not mandate any specific solution to this issue as each environment 1100 has its own facilities and limitations. 1102 Most known message authentication methods focus on a particular 1103 identifier to evaluate. SPF and Sender ID differ in that they can 1104 yield a result based on more than one identifier; specifically, SPF 1105 can evaluate the RFC5321.HELO parameter or the RFC5321.MailFrom 1106 parameter, and Sender ID can evaluate the RFC5321.MailFrom parameter 1107 or the Purported Responsible Address (PRA) identity. When generating 1108 this field to report those results, only the parameter that yielded 1109 the result is included. 1111 For MTAs that add this header field, adding header fields in order 1112 (at the top), per Section 3.6 of [MAIL], is particularly important. 1113 Moreover, this header field SHOULD be inserted above any other trace 1114 header fields such MTAs might prepend. This placement allows easy 1115 detection of header fields that can be trusted. 1117 End users making direct use of this header field might inadvertently 1118 trust information that has not been properly vetted. If, for 1119 example, a basic SPF result were to be relayed that claims an 1120 authenticated addr-spec, the local-part of that addr-spec has 1121 actually not been authenticated. Thus, an MTA adding this header 1122 field SHOULD NOT include any data that has not been authenticated by 1123 the method(s) being applied. Moreover, MUAs SHOULD NOT render to 1124 users such information if it is presented by a method known not to 1125 authenticate it. 1127 4.1. Header Field Position and Interpretation 1129 In order to ensure non-ambiguous results and avoid the impact of 1130 false header fields, MUAs and downstream filters SHOULD NOT interpret 1131 this header field unless specifically configured to do so by the user 1132 or administrator. That is, this interpretation should not be "on by 1133 default". Naturally then, users or administrators ought not activate 1134 such a feature unless they are certain the header field will be 1135 validly added by an agent within the ADMD that accepts the mail that 1136 is ultimately read by the MUA, and instances of the header field 1137 appearing to originate within the ADMD but are actually added by 1138 foreign MTAs will be removed before delivery. 1140 Furthermore, MUAs and downstream filters SHOULD NOT interpret this 1141 header field unless the authentication service identifier it bears 1142 appears to be one used within its own ADMD as configured by the user 1143 or administrator. 1145 MUAs and downstream filters MUST ignore any result reported using a 1146 "result" not specified in the IANA "Result Code" registry or a 1147 "ptype" not listed in the corresponding registry for such values as 1148 defined in Section 6. Moreover, such agents MUST ignore a result 1149 indicated for any "method" they do not specifically support. 1151 An MUA SHOULD NOT reveal these results to end users, absent careful 1152 human factors design considerations and testing, for the presentation 1153 of trust-related materials. For example, an attacker could register 1154 examp1e.com (note the digit "one") and send signed mail to intended 1155 victims; a verifier would detect that the signature was valid and 1156 report a "pass" even though it's clear the DNS domain name was 1157 intended to mislead. See Section 7.2 for further discussion. 1159 As stated in Section 2.1, this header field MUST be treated as though 1160 it were a trace header field as defined in Section 3.6.7 of [MAIL] 1161 and hence MUST NOT be reordered and MUST be prepended to the message, 1162 so that there is generally some indication upon delivery of where in 1163 the chain of handling MTAs the message authentication was done. 1165 Note that there are a few message handlers that are only capable of 1166 appending new header fields to a message. Strictly speaking, these 1167 handlers are not compliant with this specification. They can still 1168 add the header field to carry authentication details, but any signal 1169 about where in the handling chain the work was done may be lost. 1170 Consumers SHOULD be designed such that this can be tolerated, 1171 especially from a producer known to have this limitation. 1173 MUAs SHOULD ignore instances of this header field discovered within 1174 message/rfc822 MIME attachments. 1176 Further discussion of these topics can be found in Section 7 below. 1178 4.2. Local Policy Enforcement 1180 Some sites have a local policy that considers any particular 1181 authentication policy's non-recoverable failure results (typically 1182 "fail" or similar) as justification for rejecting the message. In 1183 such cases, the border MTA SHOULD issue an SMTP rejection response to 1184 the message, rather than adding this header field and allowing the 1185 message to proceed toward delivery. This is more desirable than 1186 allowing the message to reach an internal host's MTA or spam filter, 1187 thus possibly generating a local rejection such as a Delivery Status 1188 Notification (DSN) [DSN] to a forged originator. Such generated 1189 rejections are colloquially known as "backscatter". 1191 The same MAY also be done for local policy decisions overriding the 1192 results of the authentication methods (e.g., the "policy" result 1193 codes described in Section 2.7). 1195 Such rejections at the SMTP protocol level are not possible if local 1196 policy is enforced at the MUA and not the MTA. 1198 5. Removing Existing Header Fields 1200 For security reasons, any MTA conforming to this specification MUST 1201 delete any discovered instance of this header field that claims, by 1202 virtue of its authentication service identifier, to have been added 1203 within its trust boundary but that did not come directly from another 1204 trusted MTA. For example, an MTA for example.com receiving a message 1205 MUST delete or otherwise obscure any instance of this header field 1206 bearing an authentication service identifier indicating that the 1207 header field was added within example.com prior to adding its own 1208 header fields. This could mean each MTA will have to be equipped 1209 with a list of internal MTAs known to be compliant (and hence 1210 trustworthy). 1212 For simplicity and maximum security, a border MTA could remove all 1213 instances of this header field on mail crossing into its trust 1214 boundary. However, this may conflict with the desire to access 1215 authentication results performed by trusted external service 1216 providers. It may also invalidate signed messages whose signatures 1217 cover external instances of this header field. A more robust border 1218 MTA could allow a specific list of authenticating MTAs whose 1219 information is to be admitted, removing the header field originating 1220 from all others. 1222 As stated in Section 1.2, a formal definition of "trust boundary" is 1223 deliberately not made here. It is entirely possible that a border 1224 MTA for example.com will explicitly trust authentication results 1225 asserted by upstream host example.net even though they exist in 1226 completely disjoint administrative boundaries. In that case, the 1227 border MTA MAY elect not to delete those results; moreover, the 1228 upstream host doing some authentication work could apply a signing 1229 technology such as [DKIM] on its own results to assure downstream 1230 hosts of their authenticity. An example of this is provided in 1231 Appendix C. 1233 Similarly, in the case of messages signed using [DKIM] or other 1234 message-signing methods that sign header fields, this removal action 1235 could invalidate one or more signatures on the message if they 1236 covered the header field to be removed. This behavior can be 1237 desirable since there's little value in validating the signature on a 1238 message with forged header fields. However, signing agents MAY 1239 therefore elect to omit these header fields from signing to avoid 1240 this situation. 1242 An MTA SHOULD remove any instance of this header field bearing a 1243 version (express or implied) that it does not support. However, an 1244 MTA MUST remove such a header field if the [SMTP] connection relaying 1245 the message is not from a trusted internal MTA. This means the MTA 1246 needs to be able to understand versions of this header field at least 1247 as late as the ones understood by the MUAs or other consumers within 1248 its ADMD. 1250 6. IANA Considerations 1252 IANA has registered the defined header field and created tables as 1253 described below. These registry actions were originally defined by 1254 [RFC5451] and updated by [RFC6577] and [RFC7001]. The created 1255 registries are being further updated here to increase their 1256 completeness. 1258 6.1. The Authentication-Results Header Field 1260 [RFC5451] added the Authentication-Results header field to the IANA 1261 "Permanent Message Header Field Names" registry, per the procedure 1262 found in [IANA-HEADERS]. That entry is to be updated to reference 1263 this document. The following is the registration template: 1265 Header field name: Authentication-Results 1266 Applicable protocol: mail ([MAIL]) 1267 Status: Standard 1268 Author/Change controller: IETF 1269 Specification document(s): [this RFC] 1270 Related information: none 1272 6.2. "Email Authentication Methods" Registry Description 1274 Names of message authentication methods supported by this 1275 specification are to be registered with IANA, with the exception of 1276 experimental names as described in Section 2.7.6. Along with each 1277 method is recorded the properties that accompany the method's result. 1279 The "Email Authentication Parameters" group, and within it the "Email 1280 Authentication Methods" registry, were created by [RFC5451] for this 1281 purpose. [RFC6577] added a "status" field for each entry. [RFC7001] 1282 amended the rules governing that registry, and also added a "version" 1283 field to the registry. 1285 The reference for that registry shall be updated to reference this 1286 document. 1288 New entries are assigned only for values that have received Expert 1289 Review, per [IANA-CONSIDERATIONS]. The designated expert shall be 1290 appointed by the IESG. The designated expert has discretion to 1291 request that a publication be referenced if a clear, concise 1292 definition of the authentication method cannot be provided such that 1293 interoperability is assured. Registrations should otherwise be 1294 permitted. The designated expert can also handle requests to mark 1295 any current registration as "deprecated". 1297 No two entries can have the same combination of method, ptype, and 1298 property. 1300 An entry in this registry contains the following: 1302 Method: the name of the method; 1304 Definition: a reference to the document that created this entry, if 1305 any (see below); 1307 ptype: a "ptype" value appropriate for use with that method; 1309 property: a "property" value matching that "ptype" also appropriate 1310 for use with that method; 1312 Value: a brief description of the value to be supplied with that 1313 method/ptype/property tuple; 1315 Status: the status of this entry, which is either: 1317 active: The entry is in current use. 1319 deprecated: The entry is no longer in current use. 1321 Version: a version number associated with the method (preferably 1322 starting at "1"). 1324 The "Definition" field will typically refer to a permanent document, 1325 or at least some descriptive text, where additional information about 1326 the entry being added can be found. This might in turn reference the 1327 document where the method is defined so that all of the semantics 1328 around creating or interpreting an Authentication-Results header 1329 field using this method, ptype, and property can be understood. 1331 6.3. "Email Authentication Methods" Registry Update 1333 The following changes are to be made to this registry upon approval 1334 of this document: 1336 1. The "Defined" field shall be renamed to "Definition", to be 1337 consistent with the other registries in this group. 1339 2. The current entry for the "auth" method shall have its "property" 1340 field changed to "mailfrom", and its "Definition" field changed 1341 to this document. 1343 3. The entry for the "dkim" method, "header" ptype and "b" property 1344 shall now reference [RFC6008] as its defining document, and the 1345 reference shall be removed from the description. 1347 4. All other "dkim", "domainkeys", "iprev", "sender-id", and "spf" 1348 method entries shall have their "Definition" fields changed to 1349 this document, as this document contains a complete description 1350 of the registry and these corresponding values. 1352 5. All "smime" entries have their "Definition" fields changed to 1353 [SMIME-REG]. 1355 6. The "value" field of the "smime" entry using property "smime- 1356 part" shall be changed to read: "The MIME body part reference 1357 that contains the S/MIME signature. See Section 3.2.1 of RFC7281 1358 for full syntax." 1360 7. The following entry is to be added: 1362 Method: auth 1364 Definition: [this document] 1366 ptype: smtp 1368 property: auth 1370 Value: identity confirmed by the AUTH command 1372 Status: active 1374 Version: 1 1376 8. The values of the "domainkeys" entries for ptype "header" are 1377 updated as follows: 1379 from: contents of the [MAIL] From: header field, after removing 1380 comments, and removing the local-part and following "@" if not 1381 authenticated 1383 sender: contents of the [MAIL] Sender: header field, after 1384 removing comments, and removing the local-part and following 1385 "@" if not authenticated 1387 9. All entries for "dkim-adsp" and "domainkeys" shall have their 1388 Status values changed to "deprecated", reflecting the fact that 1389 the corresponding specifications now have Historical status. 1390 Their "Definition" fields shall also be modified to include a 1391 reference to this document. 1393 6.4. "Email Authentication Property Types" Registry 1395 [PTYPES-REGISTRY] created the Email Authentication Property Types 1396 registry. 1398 Entries in this registry are subject to the Expert Review rules as 1399 described in [IANA-CONSIDERATIONS]. Each entry in the registry 1400 requires the following values: 1402 ptype: The name of the ptype being registered, which must fit within 1403 the ABNF described in Section 2.2. 1405 Definition: An optional reference to a defining specification. 1407 Description: A brief description of what sort of information this 1408 "ptype" is meant to cover. 1410 For new entries, the Designated Expert needs to assure that the 1411 description provided for the new entry adequately describes the 1412 intended use. An example would be helpful to include in the entry's 1413 defining document, if any, although entries in the "Email 1414 Authentication Methods" registry or the "Email Authentication Result 1415 Names" registry might also serve as examples of intended use. 1417 As this is a complete re-statement of the definition and rules for 1418 this registry, IANA shall update this registry to show Section 2.3 of 1419 this document as the current definitions for the "body", "header", 1420 "policy" and "smtp" entries of that registry. References to 1421 [RFC7001] shall be removed. 1423 6.5. "Email Authentication Result Names" Description 1425 Names of message authentication result codes supported by this 1426 specification must be registered with IANA, with the exception of 1427 experimental codes as described in Section 2.7.7. A registry was 1428 created by [RFC5451] for this purpose. [RFC6577] added the "status" 1429 column, and [RFC7001] updated the rules governing that registry. 1431 New entries are assigned only for values that have received Expert 1432 Review, per [IANA-CONSIDERATIONS]. The designated expert shall be 1433 appointed by the IESG. The designated expert has discretion to 1434 request that a publication be referenced if a clear, concise 1435 definition of the authentication result cannot be provided such that 1436 interoperability is assured. Registrations should otherwise be 1437 permitted. The designated expert can also handle requests to mark 1438 any current registration as "deprecated". 1440 No two entries can have the same combination of method and code. 1442 An entry in this registry contains the following: 1444 Auth Method: an authentication method for which results are being 1445 returned using the header field defined in this document; 1447 Code: a result code that can be returned for this authentication 1448 method; 1450 Specification: either free form text explaining the meaning of this 1451 method-code combination, or a reference to such a definition. 1453 Status: the status of this entry, which is either: 1455 active: The entry is in current use. 1457 deprecated: The entry is no longer in current use. 1459 6.6. "Email Authentication Result Names" Update 1461 This document includes a complete description of the registry, 1462 obsoleting [RFC7001]. Accordingly, the following changes are to be 1463 made to this registry on publication of this document: 1465 o The "Defined" field shall be removed. 1467 o The "Meaning" field shall be renamed to "Specification", as 1468 described above. 1470 o The "Auth Method" field shall appear before the "Code" field. 1472 o For easier searching, the table shall be arranged such that it is 1473 sorted first by Auth Method, then by Code within each Auth Method 1474 grouping. 1476 o All entries for the "dkim", "domainkeys", "spf", "sender-id", 1477 "auth", and "iprev" methods shall have their "Specification" 1478 fields changed to refer to this document, as follows: 1480 dkim: [this document] Section 2.7.1 1482 domainkeys: [this document] Section 2.7.1 1484 spf: [this document] Section 2.7.2 1486 sender-id: [this document] Section 2.7.2 1488 auth: [this document] Section 2.7.4 1490 iprev: [this document] Section 2.7.3 1492 o All entries for "dkim-adsp" that are missing an explicit reference 1493 to a defining document shall have [ADSP] added to their 1494 "Specification" fields. 1496 o All entries for "dmarc" shall have their "Specification" fields 1497 changed to reference Secton 11.2 of [DMARC]. 1499 o All entries for "dkim-adsp" and "domainkeys" shall have their 1500 Status values changed to "deprecated", reflecting the fact that 1501 the corresponding specifications now have Historical status. 1502 Their "Specification" fields shall also be modified to include a 1503 reference to this document. 1505 6.7. SMTP Enhanced Status Codes 1507 The entry for X.7.25 in the Enumerated Status Codes sub-registry of 1508 the SMTP Enhanced Status Codes registry is to be updated to refer to 1509 this document instead of [RFC7001]. 1511 7. Security Considerations 1513 The following security considerations apply when adding or processing 1514 the Authentication-Results header field: 1516 7.1. Forged Header Fields 1518 An MUA or filter that accesses a mailbox whose messages are handled 1519 by a non-conformant MTA, and understands Authentication-Results 1520 header fields, could potentially make false conclusions based on 1521 forged header fields. A malicious user or agent could forge a header 1522 field using the DNS domain of a receiving ADMD as the authserv-id 1523 token in the value of the header field and, with the rest of the 1524 value, claim that the message was properly authenticated. The non- 1525 conformant MTA would fail to strip the forged header field, and the 1526 MUA could inappropriately trust it. 1528 For this reason, it is best not to have processing of the 1529 Authentication-Results header field enabled by default; instead, it 1530 should be ignored, at least for the purposes of enacting filtering 1531 decisions, unless specifically enabled by the user or administrator 1532 after verifying that the border MTA is compliant. It is acceptable 1533 to have an MUA aware of this specification but have an explicit list 1534 of hostnames whose Authentication-Results header fields are 1535 trustworthy; however, this list should initially be empty. 1537 Proposed alternative solutions to this problem were made some time 1538 ago and are listed below. To date, they have not been developed due 1539 to lack of demand but are documented here should the information be 1540 useful at some point in the future: 1542 1. Possibly the simplest is a digital signature protecting the 1543 header field, such as using [DKIM], that can be verified by an 1544 MUA by using a posted public key. Although one of the main 1545 purposes of this document is to relieve the burden of doing 1546 message authentication work at the MUA, this only requires that 1547 the MUA learn a single authentication scheme even if a number of 1548 them are in use at the border MTA. Note that [DKIM] requires 1549 that the From header field be signed, although in this 1550 application, the signing agent (a trusted MTA) likely cannot 1551 authenticate that value, so the fact that it is signed should be 1552 ignored. Where the authserv-id is the ADMD's domain name, the 1553 authserv-id matching this valid internal signature's "d=" DKIM 1554 value is sufficient. 1556 2. Another would be a means to interrogate the MTA that added the 1557 header field to see if it is actually providing any message 1558 authentication services and saw the message in question, but this 1559 isn't especially palatable given the work required to craft and 1560 implement such a scheme. 1562 3. Yet another might be a method to interrogate the internal MTAs 1563 that apparently handled the message (based on Received header 1564 fields) to determine whether any of them conform to Section 5 of 1565 this memo. This, too, has potentially high barriers to entry. 1567 4. Extensions to [IMAP], [SMTP], and [POP3] could be defined to 1568 allow an MUA or filtering agent to acquire the authserv-id in use 1569 within an ADMD, thus allowing it to identify which 1570 Authentication-Results header fields it can trust. 1572 5. On the presumption that internal MTAs are fully compliant with 1573 Section 3.6 of [MAIL] and the compliant internal MTAs are using 1574 their own hostnames or the ADMD's DNS domain name as the 1575 authserv-id token, the header field proposed here should always 1576 appear above a Received header added by a trusted MTA. This can 1577 be used as a test for header field validity. 1579 Support for some of these is being considered for future work. 1581 In any case, a mechanism needs to exist for an MUA or filter to 1582 verify that the host that appears to have added the header field (a) 1583 actually did so and (b) is legitimately adding that header field for 1584 this delivery. Given the variety of messaging environments deployed 1585 today, consensus appears to be that specifying a particular mechanism 1586 for doing so is not appropriate for this document. 1588 Mitigation of the forged header field attack can also be accomplished 1589 by moving the authentication results data into meta-data associated 1590 with the message. In particular, an [SMTP] extension could be 1591 established to communicate authentication results from the border MTA 1592 to intermediate and delivery MTAs; the latter of these could arrange 1593 to store the authentication results as meta-data retrieved and 1594 rendered along with the message by an [IMAP] client aware of a 1595 similar extension in that protocol. The delivery MTA would be told 1596 to trust data via this extension only from MTAs it trusts, and border 1597 MTAs would not accept data via this extension from any source. There 1598 is no vector in such an arrangement for forgery of authentication 1599 data by an outside agent. 1601 7.2. Misleading Results 1603 Until some form of service for querying the reputation of a sending 1604 agent is widely deployed, the existence of this header field 1605 indicating a "pass" does not render the message trustworthy. It is 1606 possible for an arriving piece of spam or other undesirable mail to 1607 pass checks by several of the methods enumerated above (e.g., a piece 1608 of spam signed using [DKIM] by the originator of the spam, which 1609 might be a spammer or a compromised system). In particular, this 1610 issue is not resolved by forged header field removal discussed above. 1612 Hence, MUAs and downstream filters must take some care with use of 1613 this header even after possibly malicious headers are scrubbed. 1615 7.3. Header Field Position 1617 Despite the requirements of [MAIL], header fields can sometimes be 1618 reordered en route by intermediate MTAs. The goal of requiring 1619 header field addition only at the top of a message is an 1620 acknowledgment that some MTAs do reorder header fields, but most do 1621 not. Thus, in the general case, there will be some indication of 1622 which MTAs (if any) handled the message after the addition of the 1623 header field defined here. 1625 7.4. Reverse IP Query Denial-of-Service Attacks 1627 Section 4.6.4 of [SPF] describes a DNS-based denial-of-service attack 1628 for verifiers that attempt DNS-based identity verification of 1629 arriving client connections. A verifier wishing to do this check and 1630 report this information needs to take care not to go to unbounded 1631 lengths to resolve "A" and "PTR" queries. MUAs or other filters 1632 making use of an "iprev" result specified by this document need to be 1633 aware of the algorithm used by the verifier reporting the result and, 1634 especially, its limitations. 1636 7.5. Mitigation of Backscatter 1638 Failing to follow the instructions of Section 4.2 can result in a 1639 denial-of-service attack caused by the generation of [DSN] messages 1640 (or equivalent) to addresses that did not send the messages being 1641 rejected. 1643 7.6. Internal MTA Lists 1645 Section 5 describes a procedure for scrubbing header fields that may 1646 contain forged authentication results about a message. A compliant 1647 installation will have to include, at each MTA, a list of other MTAs 1648 known to be compliant and trustworthy. Failing to keep this list 1649 current as internal infrastructure changes may expose an ADMD to 1650 attack. 1652 7.7. Attacks against Authentication Methods 1654 If an attack becomes known against an authentication method, clearly 1655 then the agent verifying that method can be fooled into thinking an 1656 inauthentic message is authentic, and thus the value of this header 1657 field can be misleading. It follows that any attack against the 1658 authentication methods supported by this document is also a security 1659 consideration here. 1661 7.8. Intentionally Malformed Header Fields 1663 It is possible for an attacker to add an Authentication-Results 1664 header field that is extraordinarily large or otherwise malformed in 1665 an attempt to discover or exploit weaknesses in header field parsing 1666 code. Implementers must thoroughly verify all such header fields 1667 received from MTAs and be robust against intentionally as well as 1668 unintentionally malformed header fields. 1670 7.9. Compromised Internal Hosts 1672 An internal MUA or MTA that has been compromised could generate mail 1673 with a forged From header field and a forged Authentication-Results 1674 header field that endorses it. Although it is clearly a larger 1675 concern to have compromised internal machines than it is to prove the 1676 value of this header field, this risk can be mitigated by arranging 1677 that internal MTAs will remove this header field if it claims to have 1678 been added by a trusted border MTA (as described above), yet the 1679 [SMTP] connection is not coming from an internal machine known to be 1680 running an authorized MTA. However, in such a configuration, 1681 legitimate MTAs will have to add this header field when legitimate 1682 internal-only messages are generated. This is also covered in 1683 Section 5. 1685 7.10. Encapsulated Instances 1687 MIME messages can contain attachments of type "message/rfc822", which 1688 contain other messages. Such an encapsulated message can also 1689 contain an Authentication-Results header field. Although the 1690 processing of these is outside of the intended scope of this document 1691 (see Section 1.3), some early guidance to MUA developers is 1692 appropriate here. 1694 Since MTAs are unlikely to strip Authentication-Results header fields 1695 after mailbox delivery, MUAs are advised in Section 4.1 to ignore 1696 such instances within MIME attachments. Moreover, when extracting a 1697 message digest to separate mail store messages or other media, such 1698 header fields should be removed so that they will never be 1699 interpreted improperly by MUAs that might later consume them. 1701 7.11. Reverse Mapping 1703 Although Section 3 of this memo includes explicit support for the 1704 "iprev" method, its value as an authentication mechanism is limited. 1705 Implementers of both this proposal and agents that use the data it 1706 relays are encouraged to become familiar with the issues raised by 1707 [DNSOP-REVERSE] when deciding whether or not to include support for 1708 "iprev". 1710 8. References 1712 8.1. Normative References 1714 [ABNF] Crocker, D. and P. Overell, "Augmented BNF for 1715 Syntax Specifications: ABNF", STD 68, 1716 RFC 5234, January 2008. 1718 [IANA-HEADERS] Klyne, G., Nottingham, M., and J. Mogul, 1719 "Registration Procedures for Message Header 1720 Fields", BCP 90, RFC 3864, September 2004. 1722 [KEYWORDS] Bradner, S., "Key words for use in RFCs to 1723 Indicate Requirement Levels", BCP 14, 1724 RFC 2119. 1726 [MAIL] Resnick, P., Ed., "Internet Message Format", 1727 RFC 5322, October 2008. 1729 [MIME] Freed, N. and N. Borenstein, "Multipurpose 1730 Internet Mail Extensions (MIME) Part One: 1731 Format of Internet Message Bodies", RFC 2045, 1732 November 1996. 1734 [SMTP] Klensin, J., "Simple Mail Transfer Protocol", 1735 RFC 5321, October 2008. 1737 8.2. Informative References 1739 [ADSP] Allman, E., Fenton, J., Delany, M., and J. 1740 Levine, "DomainKeys Identified Mail (DKIM) 1741 Author Domain Signing Practices (ADSP)", 1742 RFC 5617, August 2009. 1744 [AR-VBR] Kucherawy, M., "Authentication-Results 1745 Registration for Vouch by Reference Results", 1746 RFC 6212, April 2011. 1748 [ATPS] Kucherawy, M., "DomainKeys Identified Mail 1749 (DKIM) Authorized Third-Party Signatures", 1750 RFC 6541, February 2012. 1752 [AUTH] Siemborski, R. and A. Melnikov, "SMTP Service 1753 Extension for Authentication", RFC 4954, 1754 July 2007. 1756 [AUTH-ESC] Kucherawy, M., "Email Authentication Status 1757 Codes", RFC 7372, September 2014. 1759 [DKIM] Crocker, D., Hansen, T., and M. Kucherawy, 1760 "DomainKeys Identified Mail (DKIM) 1761 Signatures", STD 76, RFC 6376, September 2011. 1763 [DMARC] Kucherawy, M., Ed. and E. Zwicky, Ed., 1764 "Domain-based Message Authentication, 1765 Reporting and Conformance (DMARC)", RFC 7489, 1766 March 2015. 1768 [DNS] Mockapetris, P., "Domain names - 1769 Implementation and Specification", STD 13, 1770 RFC 1035, November 1987. 1772 [DNS-IP6] Thomson, S., Huitema, C., Ksinant, V., and M. 1773 Souissi, "DNS Extensions to Support IP Version 1774 6", RFC 3596, October 2003. 1776 [DNSOP-REVERSE] Senie, D. and A. Sullivan, "Considerations for 1777 the use of DNS Reverse Mapping", Work 1778 in Progress, March 2008. 1780 [DOMAINKEYS] Delany, M., "Domain-Based Email Authentication 1781 Using Public Keys Advertised in the DNS 1782 (DomainKeys)", RFC 4870, May 2007. 1784 [DSN] Moore, K. and G. Vaudreuil, "An Extensible 1785 Message Format for Delivery Status 1786 Notifications", RFC 3464, January 2003. 1788 [EMAIL-ARCH] Crocker, D., "Internet Mail Architecture", 1789 RFC 5598, July 2009. 1791 [IANA-CONSIDERATIONS] Narten, T. and H. Alvestrand, "Guidelines for 1792 Writing an IANA Considerations Section in 1793 RFCs", BCP 26, RFC 5226, May 2008. 1795 [IMAP] Crispin, M., "INTERNET MESSAGE ACCESS PROTOCOL 1796 - VERSION 4rev1", RFC 3501, March 2003. 1798 [POP3] Myers, J. and M. Rose, "Post Office Protocol - 1799 Version 3", STD 53, RFC 1939, May 1996. 1801 [PRA] Lyon, J., "Purported Responsible Address in 1802 E-Mail Messages", RFC 4407, April 2006. 1804 [PTYPES-REGISTRY] Kucherawy, M., "A Property Types Registry for 1805 the Authentication-Results Header Field", 1806 RFC 7410, December 2014. 1808 [RFC5451] Kucherawy, M., "Message Header Field for 1809 Indicating Message Authentication Status", 1810 RFC 5451, April 2009. 1812 [RFC6008] Kucherawy, M., "Authentication-Results 1813 Registration for Differentiating among 1814 Cryptographic Results", RFC 6008, 1815 September 2010. 1817 [RFC6577] Kucherawy, M., "Authentication-Results 1818 Registration Update for Sender Policy 1819 Framework (SPF) Results", RFC 6577, 1820 March 2012. 1822 [RFC7001] Kucherawy, M., "Message Header Field for 1823 Indicating Message Authentication Status", 1824 RFC 7001, September 2013. 1826 [RRVS] Mills, W. and M. Kucherawy, "The Require- 1827 Recipient-Valid-Since Header Field and SMTP 1828 Service Extension", RFC 7293, July 2014. 1830 [SECURITY] Rescorla, E. and B. Korver, "Guidelines for 1831 Writing RFC Text on Security Considerations", 1832 BCP 72, RFC 3552, July 2003. 1834 [SENDERID] Lyon, J. and M. Wong, "Sender ID: 1835 Authenticating E-Mail", RFC 4406, April 2006. 1837 [SMIME-REG] Melnikov, A., "Authentication-Results 1838 Registration for S/MIME Signature 1839 Verification", RFC 7281, June 2014. 1841 [SPF] Kitterman, S., "Sender Policy Framework (SPF) 1842 for Authorizing Use of Domains in E-Mail, 1843 Version 1", RFC 7208, April 2014. 1845 [VBR] Hoffman, P., Levine, J., and A. Hathcock, 1846 "Vouch By Reference", RFC 5518, April 2009. 1848 Appendix A. Acknowledgments 1850 The author wishes to acknowledge the following individuals for their 1851 review and constructive criticism of this document: Stephane 1852 Bortzmeyer, Scott Kitterman, John Levine, Tom Petch, and Pete 1853 Resnick. 1855 Appendix B. Legacy MUAs 1857 Implementers of this protocol should be aware that many MUAs are 1858 unlikely to be retrofitted to support the new header field and its 1859 semantics. In the interests of convenience and quicker adoption, a 1860 delivery MTA might want to consider adding things that are processed 1861 by existing MUAs in addition to the Authentication-Results header 1862 field. One suggestion is to include a Priority header field, on 1863 messages that don't already have such a header field, containing a 1864 value that reflects the strength of the authentication that was 1865 accomplished, e.g., "low" for weak or no authentication, "normal" or 1866 "high" for good or strong authentication. 1868 Some modern MUAs can already filter based on the content of this 1869 header field. However, there is keen interest in having MUAs make 1870 some kind of graphical representation of this header field's meaning 1871 to end users. Until this capability is added, other interim means of 1872 conveying authentication results may be necessary while this proposal 1873 and its successors are adopted. 1875 Appendix C. Authentication-Results Examples 1877 This section presents some examples of the use of this header field 1878 to indicate authentication results. 1880 C.1. Trivial Case; Header Field Not Present 1882 The trivial case: 1884 Received: from mail-router.example.com 1885 (mail-router.example.com [192.0.2.1]) 1886 by server.example.org (8.11.6/8.11.6) 1887 with ESMTP id g1G0r1kA003489; 1888 Fri, Feb 15 2002 17:19:07 -0800 1889 From: sender@example.com 1890 Date: Fri, Feb 15 2002 16:54:30 -0800 1891 To: receiver@example.org 1892 Message-Id: <12345.abc@example.com> 1893 Subject: here's a sample 1895 Hello! Goodbye! 1897 Example 1: Trivial Case 1899 The Authentication-Results header field is completely absent. The 1900 MUA may make no conclusion about the validity of the message. This 1901 could be the case because the message authentication services were 1902 not available at the time of delivery, or no service is provided, or 1903 the MTA is not in compliance with this specification. 1905 C.2. Nearly Trivial Case; Service Provided, but No Authentication Done 1907 A message that was delivered by an MTA that conforms to this 1908 specification but provides no actual message authentication service: 1910 Authentication-Results: example.org 1; none 1911 Received: from mail-router.example.com 1912 (mail-router.example.com [192.0.2.1]) 1913 by server.example.org (8.11.6/8.11.6) 1914 with ESMTP id g1G0r1kA003489; 1915 Fri, Feb 15 2002 17:19:07 -0800 1916 From: sender@example.com 1917 Date: Fri, Feb 15 2002 16:54:30 -0800 1918 To: receiver@example.org 1919 Message-Id: <12345.abc@example.com> 1920 Subject: here's a sample 1922 Hello! Goodbye! 1924 Example 2: Header Present but No Authentication Done 1926 The Authentication-Results header field is present, showing that the 1927 delivering MTA conforms to this specification. It used its DNS 1928 domain name as the authserv-id. The presence of "none" (and the 1929 absence of any method and result tokens) indicates that no message 1930 authentication was done. The version number of the specification to 1931 which the field's content conforms is explicitly provided. 1933 C.3. Service Provided, Authentication Done 1935 A message that was delivered by an MTA that conforms to this 1936 specification and applied some message authentication: 1938 Authentication-Results: example.com; 1939 spf=pass smtp.mailfrom=example.net 1940 Received: from dialup-1-2-3-4.example.net 1941 (dialup-1-2-3-4.example.net [192.0.2.200]) 1942 by mail-router.example.com (8.11.6/8.11.6) 1943 with ESMTP id g1G0r1kA003489; 1944 Fri, Feb 15 2002 17:19:07 -0800 1945 From: sender@example.net 1946 Date: Fri, Feb 15 2002 16:54:30 -0800 1947 To: receiver@example.com 1948 Message-Id: <12345.abc@example.net> 1949 Subject: here's a sample 1951 Hello! Goodbye! 1953 Example 3: Header Reporting Results 1955 The Authentication-Results header field is present, indicating that 1956 the border MTA conforms to this specification. The authserv-id is 1957 once again the DNS domain name. Furthermore, the message was 1958 authenticated by that MTA via the method specified in [SPF]. Note 1959 that since that method cannot authenticate the local-part, it has 1960 been omitted from the result's value. The MUA could extract and 1961 relay this extra information if desired. 1963 C.4. Service Provided, Several Authentications Done, Single MTA 1965 A message that was relayed inbound via a single MTA that conforms to 1966 this specification and applied three different message authentication 1967 checks: 1969 Authentication-Results: example.com; 1970 auth=pass (cram-md5) smtp.auth=sender@example.net; 1971 spf=pass smtp.mailfrom=example.net 1972 Authentication-Results: example.com; 1973 sender-id=pass header.from=example.net 1974 Received: from dialup-1-2-3-4.example.net (8.11.6/8.11.6) 1975 (dialup-1-2-3-4.example.net [192.0.2.200]) 1976 by mail-router.example.com (8.11.6/8.11.6) 1977 with ESMTPA id g1G0r1kA003489; 1978 Fri, Feb 15 2002 17:19:07 -0800 1979 Date: Fri, Feb 15 2002 16:54:30 -0800 1980 To: receiver@example.com 1981 From: sender@example.net 1982 Message-Id: <12345.abc@example.net> 1983 Subject: here's a sample 1985 Hello! Goodbye! 1987 Example 4: Headers Reporting Results from One MTA 1989 The Authentication-Results header field is present, indicating that 1990 the delivering MTA conforms to this specification. Once again, the 1991 receiving DNS domain name is used as the authserv-id. Furthermore, 1992 the sender authenticated herself/himself to the MTA via a method 1993 specified in [AUTH], and both SPF and Sender ID checks were done and 1994 passed. The MUA could extract and relay this extra information if 1995 desired. 1997 Two Authentication-Results header fields are not required since the 1998 same host did all of the checking. The authenticating agent could 1999 have consolidated all the results into one header field. 2001 This example illustrates a scenario in which a remote user on a 2002 dialup connection (example.net) sends mail to a border MTA 2003 (example.com) using SMTP authentication to prove identity. The 2004 dialup provider has been explicitly authorized to relay mail as 2005 example.com resulting in passes by the SPF and Sender ID checks. 2007 C.5. Service Provided, Several Authentications Done, Different MTAs 2009 A message that was relayed inbound by two different MTAs that conform 2010 to this specification and applied multiple message authentication 2011 checks: 2013 Authentication-Results: example.com; 2014 sender-id=fail header.from=example.com; 2015 dkim=pass (good signature) header.d=example.com 2016 Received: from mail-router.example.com 2017 (mail-router.example.com [192.0.2.1]) 2018 by auth-checker.example.com (8.11.6/8.11.6) 2019 with ESMTP id i7PK0sH7021929; 2020 Fri, Feb 15 2002 17:19:22 -0800 2021 DKIM-Signature: v=1; a=rsa-sha256; s=gatsby; d=example.com; 2022 t=1188964191; c=simple/simple; h=From:Date:To:Subject: 2023 Message-Id:Authentication-Results; 2024 bh=sEuZGD/pSr7ANysbY3jtdaQ3Xv9xPQtS0m70; 2025 b=EToRSuvUfQVP3Bkz ... rTB0t0gYnBVCM= 2026 Authentication-Results: example.com; 2027 auth=pass (cram-md5) smtp.auth=sender@example.com; 2028 spf=fail smtp.mailfrom=example.com 2029 Received: from dialup-1-2-3-4.example.net 2030 (dialup-1-2-3-4.example.net [192.0.2.200]) 2031 by mail-router.example.com (8.11.6/8.11.6) 2032 with ESMTPA id g1G0r1kA003489; 2033 Fri, Feb 15 2002 17:19:07 -0800 2034 From: sender@example.com 2035 Date: Fri, Feb 15 2002 16:54:30 -0800 2036 To: receiver@example.com 2037 Message-Id: <12345.abc@example.com> 2038 Subject: here's a sample 2040 Hello! Goodbye! 2042 Example 5: Headers Reporting Results from Multiple MTAs 2044 The Authentication-Results header field is present, indicating 2045 conformance to this specification. Once again, the authserv-id used 2046 is the recipient's DNS domain name. The header field is present 2047 twice because two different MTAs in the chain of delivery did 2048 authentication tests. The first MTA, mail-router.example.com, 2049 reports that SMTP AUTH and SPF were both used and that the former 2050 passed while the latter failed. In the SMTP AUTH case, additional 2051 information is provided in the comment field, which the MUA can 2052 choose to render if desired. 2054 The second MTA, auth-checker.example.com, reports that it did a 2055 Sender ID test (which failed) and a DKIM test (which passed). Again, 2056 additional data about one of the tests is provided as a comment, 2057 which the MUA may choose to render. Also noteworthy here is the fact 2058 that there is a DKIM signature added by example.com that assured the 2059 integrity of the lower Authentication-Results field. 2061 Since different hosts did the two sets of authentication checks, the 2062 header fields cannot be consolidated in this example. 2064 This example illustrates more typical transmission of mail into 2065 example.com from a user on a dialup connection example.net. The user 2066 appears to be legitimate as he/she had a valid password allowing 2067 authentication at the border MTA using SMTP AUTH. The SPF and Sender 2068 ID tests failed since example.com has not granted example.net 2069 authority to relay mail on its behalf. However, the DKIM test passed 2070 because the sending user had a private key matching one of 2071 example.com's published public keys and used it to sign the message. 2073 C.6. Service Provided, Multi-Tiered Authentication Done 2075 A message that had authentication done at various stages, one of 2076 which was outside the receiving ADMD: 2078 Authentication-Results: example.com; 2079 dkim=pass reason="good signature" 2080 header.i=@mail-router.example.net; 2081 dkim=fail reason="bad signature" 2082 header.i=@newyork.example.com 2083 Received: from mail-router.example.net 2084 (mail-router.example.net [192.0.2.250]) 2085 by chicago.example.com (8.11.6/8.11.6) 2086 for 2087 with ESMTP id i7PK0sH7021929; 2088 Fri, Feb 15 2002 17:19:22 -0800 2089 DKIM-Signature: v=1; a=rsa-sha256; s=furble; 2090 d=mail-router.example.net; t=1188964198; c=relaxed/simple; 2091 h=From:Date:To:Message-Id:Subject:Authentication-Results; 2092 bh=ftA9J6GtX8OpwUECzHnCkRzKw1uk6FNiLfJl5Nmv49E=; 2093 b=oINEO8hgn/gnunsg ... 9n9ODSNFSDij3= 2094 Authentication-Results: example.net; 2095 dkim=pass (good signature) header.i=@newyork.example.com 2096 Received: from smtp.newyork.example.com 2097 (smtp.newyork.example.com [192.0.2.220]) 2098 by mail-router.example.net (8.11.6/8.11.6) 2099 with ESMTP id g1G0r1kA003489; 2100 Fri, Feb 15 2002 17:19:07 -0800 2101 DKIM-Signature: v=1; a=rsa-sha256; s=gatsby; 2102 d=newyork.example.com; 2103 t=1188964191; c=simple/simple; 2104 h=From:Date:To:Message-Id:Subject; 2105 bh=sEu28nfs9fuZGD/pSr7ANysbY3jtdaQ3Xv9xPQtS0m7=; 2106 b=EToRSuvUfQVP3Bkz ... rTB0t0gYnBVCM= 2107 From: sender@newyork.example.com 2108 Date: Fri, Feb 15 2002 16:54:30 -0800 2109 To: meetings@example.net 2110 Message-Id: <12345.abc@newyork.example.com> 2111 Subject: here's a sample 2113 Example 6: Headers Reporting Results from Multiple MTAs in Different 2114 ADMDs 2116 In this example, we see multi-tiered authentication with an extended 2117 trust boundary. 2119 The message was sent from someone at example.com's New York office 2120 (newyork.example.com) to a mailing list managed at an intermediary. 2122 The message was signed at the origin using DKIM. 2124 The message was sent to a mailing list service provider called 2125 example.net, which is used by example.com. There, 2126 meetings@example.net is expanded to a long list of recipients, one of 2127 whom is at the Chicago office. In this example, we will assume that 2128 the trust boundary for chicago.example.com includes the mailing list 2129 server at example.net. 2131 The mailing list server there first authenticated the message and 2132 affixed an Authentication-Results header field indicating such using 2133 its DNS domain name for the authserv-id. It then altered the message 2134 by affixing some footer text to the body, including some 2135 administrivia such as unsubscription instructions. Finally, the 2136 mailing list server affixes a second DKIM signature and begins 2137 distribution of the message. 2139 The border MTA for chicago.example.com explicitly trusts results from 2140 mail-router.example.net, so that header field is not removed. It 2141 performs evaluation of both signatures and determines that the first 2142 (most recent) is a "pass" but, because of the aforementioned 2143 modifications, the second is a "fail". However, the first signature 2144 included the Authentication-Results header added at mail- 2145 router.example.net that validated the second signature. Thus, 2146 indirectly, it can be determined that the authentications claimed by 2147 both signatures are indeed valid. 2149 Note that two styles of presenting meta-data about the result are in 2150 use here. In one case, the "reason=" clause is present, which is 2151 intended for easy extraction by parsers; in the other case, the CFWS 2152 production of the ABNF is used to include such data as a header field 2153 comment. The latter can be harder for parsers to extract given the 2154 varied supported syntaxes of mail header fields. 2156 C.7. Comment-Heavy Example 2158 The formal syntax permits comments within the content in a number of 2159 places. For the sake of illustration, this example is also legal: 2161 Authentication-Results: foo.example.net (foobar) 1 (baz); 2162 dkim (Because I like it) / 1 (One yay) = (wait for it) fail 2163 policy (A dot can go here) . (like that) expired 2164 (this surprised me) = (as I wasn't expecting it) 1362471462 2166 Example 7: A Very Comment-Heavy but Perfectly Legal Example 2168 Appendix D. Operational Considerations about Message Authentication 2170 This protocol is predicated on the idea that authentication (and 2171 presumably in the future, reputation) work is typically done by 2172 border MTAs rather than MUAs or intermediate MTAs; the latter merely 2173 make use of the results determined by the former. Certainly this is 2174 not mandatory for participation in electronic mail or message 2175 authentication, but this protocol and its deployment to date are 2176 based on that model. The assumption satisfies several common ADMD 2177 requirements: 2179 1. Service operators prefer to resolve the handling of problem 2180 messages as close to the border of the ADMD as possible. This 2181 enables, for example, rejection of messages at the SMTP level 2182 rather than generating a DSN internally. Thus, doing any of the 2183 authentication or reputation work exclusively at the MUA or 2184 intermediate MTA renders this desire unattainable. 2186 2. Border MTAs are more likely to have direct access to external 2187 sources of authentication or reputation information since modern 2188 MUAs are more likely to be heavily firewalled. Thus, some MUAs 2189 might not even be able to complete the task of performing 2190 authentication or reputation evaluations without complex proxy 2191 configurations or similar burdens. 2193 3. MUAs rely upon the upstream MTAs within their trust boundaries to 2194 make correct (as much as is possible) evaluations about the 2195 message's envelope, header, and content. Thus, MUAs don't need 2196 to know how to do the work that upstream MTAs do; they only need 2197 the results of that work. 2199 4. Evaluations about the quality of a message, from simple token 2200 matching (e.g., a list of preferred DNS domains) to cryptanalysis 2201 (e.g., public/private key work), are at least a little bit 2202 expensive and thus need to be minimized. To that end, performing 2203 those tests at the border MTA is far preferred to doing that work 2204 at each MUA that handles a message. If an ADMD's environment 2205 adheres to common messaging protocols, a reputation query or an 2206 authentication check performed by a border MTA would return the 2207 same result as the same query performed by an MUA. By contrast, 2208 in an environment where the MUA does the work, a message arriving 2209 for multiple recipients would thus cause authentication or 2210 reputation evaluation to be done more than once for the same 2211 message (i.e., at each MUA), causing needless amplification of 2212 resource use and creating a possible denial-of-service attack 2213 vector. 2215 5. Minimizing change is good. As new authentication and reputation 2216 methods emerge, the list of methods supported by this header 2217 field would presumably be extended. If MUAs simply consume the 2218 contents of this header field rather than actually attempt to do 2219 authentication and/or reputation work, then MUAs only need to 2220 learn to parse this header field once; emergence of new methods 2221 requires only a configuration change at the MUAs and software 2222 changes at the MTAs (which are presumably fewer in number). When 2223 choosing to implement these functions in MTAs vs. MUAs, the 2224 issues of individual flexibility, infrastructure inertia, and 2225 scale of effort must be considered. It is typically easier to 2226 change a single MUA than an MTA because the modification affects 2227 fewer users and can be pursued with less care. However, changing 2228 many MUAs is more effort than changing a smaller number of MTAs. 2230 6. For decisions affecting message delivery and display, assessment 2231 based on authentication and reputation is best performed close to 2232 the time of message transit, as a message makes its journey 2233 toward a user's inbox, not afterwards. DKIM keys and IP address 2234 reputations, etc., can change over time or even become invalid, 2235 and users can take a long time to read a message once delivered. 2236 The value of this work thus degrades, perhaps quickly, once the 2237 delivery process has completed. This seriously diminishes the 2238 value of this work when done other than at MTAs. 2240 Many operational choices are possible within an ADMD, including the 2241 venue for performing authentication and/or reputation assessment. 2242 The current specification does not dictate any of those choices. 2243 Rather, it facilitates those cases in which information produced by 2244 one stage of analysis needs to be transported with the message to the 2245 next stage. 2247 Appendix E. Change Since RFC7001 2249 o Apply RFC7410. 2251 o Update all the RFC4408 references to RFC7208. 2253 o Add section explaining "property" values. (Errata #4201) 2255 o Some minor text reorganization. 2257 o Give registry history, enough that this is now the authoritative 2258 registry definition. 2260 o Add text explaining each of the method-ptype-property tuples 2261 registered by this document. 2263 o Change the meaning of the "Defined" column of the methods registry 2264 to be the place where each entry was created and described; it is 2265 expected that this will then refer to the method's defining 2266 document. Provide IANA with corresponding update instructions. 2268 o Clean up registry structure and content, and replace all RFC7001 2269 references with pointers to this document. 2271 o Add references: [DMARC], [PRA], [RFC6008], [RFC6577], [RRVS], 2272 [SMIME-REG]. 2274 o Add description of values that can be extracted from SMTP AUTH 2275 sessions and an example. 2277 o Much more complete descriptions of reporting DomainKeys results. 2279 o Add more detail about Sender ID. 2281 o Mark all ADSP and DomainKeys entries as deprecated since their 2282 defining documents are as well. 2284 o Rework some text around ignoring unknown ptypes. 2286 o Completely describe the ptypes registry. 2288 o EHLO is mapped to HELO for SPF. 2290 o RFC7208 uses all-lowercase result strings now, so adjust prose 2291 accordingly. 2293 o Update list of supported methods, and mention the registries 2294 immediately below there. 2296 o Mention that when a local-part is removed, the "@" goes with it. 2298 o Refer to RFC7328 in the "iprev" definition. 2300 o Correction to "smime-part" prose. 2302 o Examples that use SMTP AUTH now claim "with ESMTPA" in the 2303 Received fields. 2305 o Minor editorial adjustments. 2307 Author's Address 2309 Murray S. Kucherawy 2310 270 Upland Drive 2311 San Francisco, CA 94127 2312 US 2314 EMail: superuser@gmail.com