idnits 2.17.1 draft-ietf-appsawg-rfc5451bis-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 RFC5451, 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 (July 11, 2013) is 3941 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 498, but not defined == Missing Reference: 'RFC4408' is mentioned on line 700, but not defined ** Obsolete undefined reference: RFC 4408 (Obsoleted by RFC 7208) -- 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 5451 (Obsoleted by RFC 7001) -- Obsolete informational reference (is this intentional?): RFC 4408 (ref. 'SPF') (Obsoleted by RFC 7208) Summary: 1 error (**), 0 flaws (~~), 3 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Individual submission M. Kucherawy 3 Internet-Draft July 11, 2013 4 Obsoletes: 5451, 6577 5 (if approved) 6 Intended status: Standards Track 7 Expires: January 12, 2014 9 Message Header Field for Indicating Message Authentication Status 10 draft-ietf-appsawg-rfc5451bis-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 19 users, or 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 January 12, 2014. 38 Copyright Notice 40 Copyright (c) 2013 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 . . . . . . . . . . . . . . . . . . . . . . 5 58 1.3. Processing Scope . . . . . . . . . . . . . . . . . . . . . 6 59 1.4. Requirements . . . . . . . . . . . . . . . . . . . . . . . 6 60 1.5. Definitions . . . . . . . . . . . . . . . . . . . . . . . 6 61 1.5.1. Key Words . . . . . . . . . . . . . . . . . . . . . . 7 62 1.5.2. Security . . . . . . . . . . . . . . . . . . . . . . . 7 63 1.5.3. Email Architecture . . . . . . . . . . . . . . . . . . 7 64 1.5.4. Other Terms . . . . . . . . . . . . . . . . . . . . . 8 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. Authentication Identifier Field . . . . . . . . . . . . . 12 70 2.4. Version Tokens . . . . . . . . . . . . . . . . . . . . . . 13 71 2.5. Defined Methods and Result Values . . . . . . . . . . . . 14 72 2.5.1. DKIM and DomainKeys . . . . . . . . . . . . . . . . . 14 73 2.5.2. SPF and Sender-ID . . . . . . . . . . . . . . . . . . 15 74 2.5.3. "iprev" . . . . . . . . . . . . . . . . . . . . . . . 16 75 2.5.4. SMTP AUTH . . . . . . . . . . . . . . . . . . . . . . 17 76 2.5.5. Other Registered Codes . . . . . . . . . . . . . . . . 17 77 2.5.6. Extension Methods . . . . . . . . . . . . . . . . . . 18 78 2.5.7. Extension Result Codes . . . . . . . . . . . . . . . . 18 79 3. The "iprev" Authentication Method . . . . . . . . . . . . . . 19 80 4. Adding the Header Field to a Message . . . . . . . . . . . . . 20 81 4.1. Header Field Position and Interpretation . . . . . . . . . 21 82 4.2. Local Policy Enforcement . . . . . . . . . . . . . . . . . 22 83 5. Removing Existing Header Fields . . . . . . . . . . . . . . . 23 84 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 24 85 6.1. The Authentication-Results Header Field . . . . . . . . . 24 86 6.2. Email Authentication Method Name Registry . . . . . . . . 24 87 6.3. Email Authentication Result Names Registry . . . . . . . . 25 88 7. Implementation Status . . . . . . . . . . . . . . . . . . . . 25 89 7.1. Google Mail . . . . . . . . . . . . . . . . . . . . . . . 26 90 7.2. Yahoo! Mail . . . . . . . . . . . . . . . . . . . . . . . 26 91 7.3. Hotmail . . . . . . . . . . . . . . . . . . . . . . . . . 27 92 7.4. Courier MTA . . . . . . . . . . . . . . . . . . . . . . . 27 93 7.5. sid-milter . . . . . . . . . . . . . . . . . . . . . . . . 27 94 7.6. opendkim . . . . . . . . . . . . . . . . . . . . . . . . . 28 95 7.7. opendmarc . . . . . . . . . . . . . . . . . . . . . . . . 28 96 7.8. authres . . . . . . . . . . . . . . . . . . . . . . . . . 29 98 8. Security Considerations . . . . . . . . . . . . . . . . . . . 29 99 8.1. Forged Header Fields . . . . . . . . . . . . . . . . . . . 29 100 8.2. Misleading Results . . . . . . . . . . . . . . . . . . . . 31 101 8.3. Header Field Position . . . . . . . . . . . . . . . . . . 31 102 8.4. Reverse IP Query Denial-of-Service Attacks . . . . . . . . 32 103 8.5. Mitigation of Backscatter . . . . . . . . . . . . . . . . 32 104 8.6. Internal MTA Lists . . . . . . . . . . . . . . . . . . . . 32 105 8.7. Attacks against Authentication Methods . . . . . . . . . . 32 106 8.8. Intentionally Malformed Header Fields . . . . . . . . . . 32 107 8.9. Compromised Internal Hosts . . . . . . . . . . . . . . . . 33 108 8.10. Encapsulated Instances . . . . . . . . . . . . . . . . . . 33 109 8.11. Reverse Mapping . . . . . . . . . . . . . . . . . . . . . 33 110 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 33 111 9.1. Normative References . . . . . . . . . . . . . . . . . . . 33 112 9.2. Informative References . . . . . . . . . . . . . . . . . . 34 113 Appendix A. Acknowledgements . . . . . . . . . . . . . . . . . . 36 114 Appendix B. Legacy MUAs . . . . . . . . . . . . . . . . . . . . . 36 115 Appendix C. Authentication-Results Examples . . . . . . . . . . . 36 116 C.1. Trivial Case; Header Field Not Present . . . . . . . . . . 37 117 C.2. Nearly Trivial Case; Service Provided, But No 118 Authentication Done . . . . . . . . . . . . . . . . . . . 37 119 C.3. Service Provided, Authentication Done . . . . . . . . . . 38 120 C.4. Service Provided, Several Authentications Done, Single 121 MTA . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 122 C.5. Service Provided, Several Authentications Done, 123 Different MTAs . . . . . . . . . . . . . . . . . . . . . . 40 124 C.6. Service Provided, Multi-Tiered Authentication Done . . . . 42 125 C.7. Comment-Heavy Example . . . . . . . . . . . . . . . . . . 43 126 Appendix D. Operational Considerations about Message 127 Authentication . . . . . . . . . . . . . . . . . . . 44 128 Appendix E. Changes since RFC5451 . . . . . . . . . . . . . . . . 45 130 1. Introduction 132 This document describes a header field called Authentication-Results 133 for electronic mail messages that presents the results of a message 134 authentication effort in a machine-readable format. The intent of 135 the header field is to create a place to collect such data when 136 message authentication mechanisms are in use so that a Mail User 137 Agent (MUA) and downstream filters can make filtering decisions 138 and/or provide a recommendation to the user as to the validity of the 139 message's origin and possibly the safety and integrity of its 140 content. 142 This document revises the original definition found in [RFC5451] 143 based upon various authentication protocols in current use and 144 incorporates errata logged since the publication of the original 145 specification. 147 End users are not expected to be direct consumers of this header 148 field. This header field is intended for consumption by programs 149 that will then use such data or render it in a human-usable form. 151 This document specifies the format of this header field and discusses 152 the implications of its presence or absence. However, it does not 153 discuss how the data contained in the header field ought to be used, 154 such as what filtering decisions are appropriate, or how an MUA might 155 render those results) as these are local policy and/or user interface 156 design questions that are not appropriate for this document. 158 At the time of publication of this document, the following are 159 published, domain-level email authentication methods in common use: 161 o Author Domain Signing Practices ([ADSP]) 163 o SMTP Service Extension for Authentication ([AUTH]) 165 o DomainKeys Identified Mail Signatures ([DKIM]) 167 o Sender Policy Framework ([SPF]) 169 o Vouch-By-Reference ([VBR]) 171 o reverse IP address name validation ("iprev", defined in Section 3) 173 In addition, the following are non-standard methods recognized by 174 this specification that are no longer common: 176 o DomainKeys ([DOMAINKEYS]) (Historic) 177 o Sender ID ([SENDERID]) (Experimental) 179 This specification is not intended to be restricted to domain-based 180 authentication schemes, but the existing schemes in that family have 181 proven to be a good starting point for implementations. The goal is 182 to give current and future authentication schemes a common framework 183 within which to deliver their results to downstream agents and 184 discourage the creation of unique header fields for each. 186 Although SPF defined a header field called "Received-SPF" and the 187 historic DomainKeys defined one called "DomainKey-Status" for this 188 purpose, those header fields are specific to the conveyance of their 189 respective results only and thus are insufficient to satisfy the 190 requirements enumerated below. In addition, many SPF implementations 191 have adopted the header field specified here at least as an option, 192 and DomainKeys has been obsoleted by DKIM. 194 1.1. Purpose 196 The header field defined in this document is expected to serve 197 several purposes: 199 1. Convey the results of various message authentication checks, 200 which are applied by upstream filters and Mail Transfer Agents 201 (MTAs) and then passed to MUAs and downstream filters within the 202 same "trust domain". Such agents might wish to render those 203 results to end users or to use those data to apply more or less 204 stringent content checks based on authentication results; 206 2. Provide a common location within a message for this data; 208 3. Create an extensible framework for reporting new authentication 209 methods as they emerge. 211 In particular, the mere presence of this header field does not mean 212 its contents are valid. Rather, the header field is reporting 213 assertions made by one or more authentication schemes (supposedly) 214 applied somewhere upstream. For an MUA or downstream filter to treat 215 the assertions as actually valid, there must be an assessment of the 216 trust relationship among such agents, the validating MTA, and the 217 mechanism for conveying the information. 219 1.2. Trust Boundary 221 This document makes several references to the "trust boundary" of an 222 administrative management domain (ADMD). Given the diversity among 223 existing mail environments, a precise definition of this term isn't 224 possible. 226 Simply put, a transfer from the producer of the header field to the 227 consumer must occur within a context that permits the consumer to 228 treat assertions by the producer as being reliable and accurate 229 (trustworthy). How this trust is obtained is outside the scope of 230 this document. It is entirely a local matter. 232 Thus, this document defines a "trust boundary" as the delineation 233 between "external" and "internal" entities. Services that are 234 internal -- within the trust boundary -- are provided by the ADMD's 235 infrastructure for its users. Those that are external are outside of 236 the authority of the ADMD. By this definition, hosts that are within 237 a trust boundary are subject to the ADMD's authority and policies, 238 independent of their physical placement or their physical operation. 239 For example, a host within a trust boundary might actually be 240 operated by a remote service provider and reside physically within 241 its data center. 243 It is possible for a message to be evaluated inside a trust boundary, 244 but then depart and re-enter the trust boundary. An example might be 245 a forwarded message such as a message/rfc822 attachment (see 246 Multipurpose Internet Mail Extensions [MIME]), or one that is part of 247 a multipart/digest. The details reported by this field cannot be 248 trusted in that case. Thus, this field found within one of those 249 media types is typically ignored. 251 1.3. Processing Scope 253 The content of this header field is meant to convey to message 254 consumers that authentication work on the message was already done 255 within its trust boundary, and those results are being presented. It 256 is not intended to provide message parameters to consumers so that 257 they can perform authentication protocols on their own. 259 1.4. Requirements 261 This document establishes no new requirements on existing protocols 262 or servers. 264 In particular, this document establishes no requirement on MTAs to 265 reject or filter arriving messages that do not pass authentication 266 checks. The data conveyed by the specified header field's contents 267 are for the information of MUAs and filters and are to be used at 268 their discretion. 270 1.5. Definitions 272 This section defines various terms used throughout this document. 274 1.5.1. Key Words 276 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 277 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 278 document are to be interpreted as described in [KEYWORDS]. 280 1.5.2. Security 282 Guidelines for Writing RFC Text on Security Considerations 283 ([SECURITY]) discusses authentication and authorization and the 284 conflation of the two concepts. The use of those terms within the 285 context of recent message security work has given rise to slightly 286 different definitions, and this document reflects those current 287 usages, as follows: 289 o "Authorization" is the establishment of permission to use a 290 resource or represent an identity. In this context, authorization 291 indicates that a message from a particular ADMD arrived via a 292 route the ADMD has explicitly approved. 294 o "Authentication" is the assertion of validity of a piece of data 295 about a message (such as the sender's identity) or the message in 296 its entirety. 298 As examples: SPF and Sender-ID are authorization mechanisms in that 299 they express a result that shows whether or not the ADMD that 300 apparently sent the message has explicitly authorized the connecting 301 Simple Mail Transfer Protocol ([SMTP]) client to relay messages on 302 its behalf, but they do not actually validate any other property of 303 the message itself. By contrast, DKIM is agnostic as to the routing 304 of a message but uses cryptographic signatures to authenticate 305 agents, assign (some) responsibility for the message (which implies 306 authorization), and ensure that the listed portions of the message 307 were not modified in transit. Since the signatures are not tied to 308 SMTP connections, they can be added by either the ADMD of origin, 309 intermediate ADMDs (such as a mailing list server), other handling 310 agents, or any combination. 312 Rather than create a separate header field for each class of 313 solution, this proposal groups them both into a single header field. 315 1.5.3. Email Architecture 317 o A "border MTA" is an MTA that acts as a gateway between the 318 general Internet and the users within an organizational boundary. 319 (See also Section 1.2.) 321 o A "delivery MTA" (or Mail Delivery Agent or MDA) is an MTA that 322 actually enacts delivery of a message to a user's inbox or other 323 final delivery. 325 o An "intermediate MTA" is any MTA that is not a delivery MTA and is 326 also not the first MTA to handle the message. 328 The following diagram illustrates the flow of mail among these 329 defined components. See Internet Mail Architecture [EMAIL-ARCH] for 330 further discussion on general email system architecture, which 331 includes detailed descriptions of these components, and Appendix D of 332 this document for discussion about the common aspects of email 333 authentication in current environments. 335 +-----+ +-----+ +------------+ 336 | MUA |-->| MSA |-->| Border MTA | 337 +-----+ +-----+ +------------+ 338 | 339 | 340 V 341 +----------+ 342 | Internet | 343 +----------+ 344 | 345 | 346 V 347 +-----+ +-----+ +------------------+ +------------+ 348 | MUA |<--| MDA |<--| Intermediate MTA |<--| Border MTA | 349 +-----+ +-----+ +------------------+ +------------+ 351 Generally, it is assumed that the work of applying message 352 authentication schemes takes place at a border MTA or a delivery MTA. 353 This specification is written with that assumption in mind. However, 354 there are some sites at which the entire mail infrastructure consists 355 of a single host. In such cases, such terms as "border MTA" and 356 "delivery MTA" might well apply to the same machine or even the very 357 same agent. It is also possible that some message authentication 358 tests could take place on an intermediate MTA. Although this 359 document doesn't specifically describe such cases, they are not meant 360 to be excluded. 362 1.5.4. Other Terms 364 In this document, the term "producer" refers to any component that 365 adds this header field to messages it is handling, and "consumer" 366 refers to any component that identifies, extracts, and parses the 367 header field to use as part of a handling decision. 369 1.6. Trust Environment 371 This header field permits one or more message validation mechanisms 372 to communicate output to one or more separate assessment mechanisms. 373 These mechanisms operate within a unified trust boundary that defines 374 an Administrative Management Domain (ADMD). An ADMD contains one or 375 more entities that perform validation and generate the header field, 376 and one or more that consume it for some type of assessment. The 377 field often contains no integrity or validation mechanism of its own, 378 so its presence must be trusted implicitly. Hence, valid use of the 379 header field requires removing any occurrences of it that are present 380 when the message enters the ADMD. This ensures that later 381 occurrences have been added within the trust boundary of the ADMD. 383 The "authserv-id" token defined in Section 2.2 can be used to 384 reference an entire ADMD or a specific validation engine within an 385 ADMD. Although the labeling scheme is left as an operational choice, 386 some guidance for selecting a token is provided in later sections of 387 this document. 389 2. Definition and Format of the Header Field 391 This section gives a general overview of the format of the header 392 field being defined, and then provides more formal specification. 394 2.1. General Description 396 The header field specified here is called "Authentication-Results". 397 It is a Structured Header Field as defined in Internet Message Format 398 ([MAIL]) and thus all of the related definitions in that document 399 apply. 401 This header field is added at the top of the message as it transits 402 MTAs that do authentication checks, so some idea of how far away the 403 checks were done can be inferred. It is therefore considered to be a 404 Trace Field as defined in [MAIL], and thus all of the related 405 definitions in that document apply. 407 The value of the header field (after removing comments) consists of 408 an authentication identifier, an optional version, and then a series 409 of statements and supporting data. The statements are of the form 410 "method=result", and indicate which authentication method(s) were 411 applied and their respective results. For each such statement, the 412 supporting data can include a "reason" string, and one or more 413 "property=value" statements indicating which message properties were 414 evaluated to reach that conclusion. 416 The header field can appear more than once in a single message, or 417 more than one result can be represented in a single header field, or 418 a combination of these can be applied. 420 2.2. Formal Definition 422 Formally, the header field is specified as follows using Augmented 423 Backus-Naur Form ([ABNF]): 425 authres-header = "Authentication-Results:" [CFWS] authserv-id 426 [ CFWS authres-version ] 427 ( no-result / 1*resinfo ) [CFWS] CRLF 429 authserv-id = value 430 ; see below for a description of this element 432 authres-version = 1*DIGIT [CFWS] 433 ; indicates which version of this specification is in use; 434 ; this specification is version "1", and the absence of a 435 ; version implies this version of the specification 437 no-result = [CFWS] ";" [CFWS] "none" 438 ; the special case of "none" is used to indicate that no 439 ; message authentication was performed 441 resinfo = [CFWS] ";" methodspec [ CFWS reasonspec ] 442 *( CFWS propspec ) 444 methodspec = [CFWS] method [CFWS] "=" [CFWS] result 445 ; indicates which authentication method was evaluated 446 ; and what its output was 448 reasonspec = "reason" [CFWS] "=" [CFWS] value 449 ; a free-form comment on the reason the given result 450 ; was returned 452 propspec = ptype [CFWS] "." [CFWS] property [CFWS] "=" pvalue 453 ; an indication of which properties of the message 454 ; were evaluated by the authentication scheme being 455 ; applied to yield the reported result 457 method = Keyword [ [CFWS] "/" [CFWS] method-version ] 458 ; a method indicates which method's result is 459 ; represented by "result", and is one of the methods 460 ; explicitly defined as valid in this document 461 ; or is an extension method as defined below 463 method-version = 1*DIGIT [CFWS] 464 ; indicates which version of the method specification is 465 ; in use, corresponding to the matching entry in the IANA 466 ; Email Authentication Methods registry; a value of "1" 467 ; is assumed if this version string is absent 469 result = Keyword 470 ; indicates the results of the attempt to authenticate 471 ; the message; see below for details 473 ptype = "smtp" / "header" / "body" / "policy" 474 ; indicates whether the property being evaluated was 475 ; a parameter to an [SMTP] command, or was a value taken 476 ; from a message header field, or was some property of 477 ; the message body, or some other property evaluated by 478 ; the receiving MTA 480 property = special-smtp-verb / Keyword 481 ; if "ptype" is "smtp", this indicates which [SMTP] 482 ; command provided the value that was evaluated by the 483 ; authentication scheme being applied; if "ptype" is 484 ; "header", this indicates from which header field the 485 ; value being evaluated was extracted; if "ptype" is 486 ; "body", this indicates where in the message body 487 ; a value being evaluated can be found (e.g., a specific 488 ; offset into the message or a reference to a MIME part); 489 ; if "ptype" is "policy" then this indicates the name 490 ; of the policy that caused this header field to be 491 ; added (see below) 493 special-smtp-verb = "mailfrom" / "rcptto" 494 ; special cases of [SMTP] commands that are made up 495 ; of multiple words 497 pvalue = [CFWS] ( value / [ [ local-part ] "@" ] domain-name ) 498 [CFWS] 499 ; the value extracted from the message property defined 500 ; by the "ptype.property" construction 502 "local-part" is defined in Section 3.4.1, and "CFWS" is defined in 503 Section 3.2.2, of [MAIL]. 505 "Keyword" is defined in Section 4.1.2 of [SMTP]. 507 The "value" is as defined in Section 5.1 of [MIME]. 509 The "domain-name" is as defined in Section 3.5 of [DKIM]. 511 The "Keyword" used in "result" above is further constrained by the 512 necessity of being enumerated in Section 2.5. 514 See Section 2.3 for a description of the "authserv-id" element. 516 If the value portion of a "pvalue" construction identifies something 517 intended to be an e-mail identity, then it MUST use the right hand 518 portion of that ABNF definition. 520 The list of commands eligible for use with the "smtp" ptype can be 521 found in Section 4.1 of [SMTP]. 523 The "propspec" may be omitted if, for example, the method was unable 524 to extract any properties to do its evaluation yet has a result to 525 report. 527 Where an SMTP command name is being reported as a "property", the 528 agent generating the header field represents that command by 529 converting it to lowercase and dropping any spaces (e.g., "MAIL FROM" 530 becomes "mailfrom", "RCPT TO" becomes "rcptto", etc.). 532 A "ptype" value of "policy" indicates a policy decision about the 533 message not specific to a property of the message that could be 534 extracted. For example, if a method would normally report a 535 "ptype.property" of "header.From" and no From: header field was 536 present, the method can use "policy" to indicate that no conclusion 537 about the authenticity of the message could be reached. 539 Examples of complete messages using this header field can be found in 540 Appendix C. 542 2.3. Authentication Identifier Field 544 Every Authentication-Results header field has an authentication 545 service identifier field ("authserv-id" above). Specifically, this 546 is any string intended to identify the authentication service within 547 the ADMD that conducted authentication checks on the message. This 548 identifier is intended to be machine-readable and not necessarily 549 meaningful to users. 551 Since agents consuming this field will use this identifier to 552 determine whether its contents are of interest (and are safe to use), 553 the uniqueness of the identifier MUST be guaranteed by the ADMD that 554 generates it and MUST pertain to that ADMD. MUAs or downstream 555 filters SHOULD use this identifier to determine whether or not the 556 data contained in an Authentication-Results header field ought to be 557 used or ignored. 559 For simplicity and scalability, the authentication service identifier 560 SHOULD be a common token used throughout the ADMD. Common practice 561 is to use the DNS domain name used by or within that ADMD, sometimes 562 called the "organizational domain", but this is not strictly 563 necessary. 565 For tracing and debugging purposes, the authentication identifier can 566 instead be the specific hostname of the MTA performing the 567 authentication check whose result is being reported. Moreover, some 568 implementations define a sub-structure to the identifier; these are 569 outside of the scope of this specification. 571 Note, however, that using a local, relative identifier like a flat 572 hostname, rather than a hierarchical and globally unique ADMD 573 identifier like a DNS domain name, makes configuration more difficult 574 for large sites. The hierarchical identifier permits aggregating 575 related, trusted systems together under a single, parent identifier, 576 which in turn permits assessing the trust relationship with a single 577 reference. The alternative is a flat namespace requiring 578 individually listing each trusted system. Since consumers will use 579 the identifier to determine whether to use the contents of the header 580 field: 582 o Changes to the identifier impose a large, centralized 583 administrative burden. 585 o Ongoing administrative changes require constantly updating this 586 centralized table, making it difficult to ensure that an MUA or 587 downstream filter will have access to accurate information for 588 assessing the usability of the header field's content. In 589 particular, consumers of the header field will need to know not 590 only the current identifier(s) in use, but previous ones as well 591 to account for delivery latency or later re-assessment of the 592 header field's contents. 594 Examples of valid authentication identifiers are "example.com", 595 "mail.example.org", "ms1.newyork.example.com", and "example-auth". 597 2.4. Version Tokens 599 The grammar above provides for the optional inclusion of versions on 600 both the header field itself (attached to the authserv-id token) and 601 on each of the methods being reported. The method version refers to 602 the method itself, which is specified in the documents describing 603 those methods, while the authserv-id version refers to this document 604 and thus the syntax of this header field. 606 The purpose of including these is to avoid misinterpretation of the 607 results. That is, if a parser finds a version after an authserv-id 608 that it does not explicitly know, it can immediately discontinue 609 trying to parse since what follows might not be in an expected 610 format. For a method version, the parser SHOULD ignore a method 611 result if the version is not supported in case the semantics of the 612 result have a different meaning than what is expected. For example, 613 if a hypothetical DKIM version 2 yielded a "pass" result for 614 different reasons than version 1 does, a consumer of this field might 615 not want to use the altered semantics. Allowing versions in the 616 syntax is a way to indicate this and let the consumer of the header 617 field decide. 619 2.5. Defined Methods and Result Values 621 Each individual authentication method returns one of a set of 622 specific result values. The subsections below provide references to 623 the documents defining the authentication methods specifically 624 supported by this document, and their corresponding result values. 625 Verifiers SHOULD use these values as described below. New methods 626 not specified in this document, but intended to be supported by the 627 header field defined here, MUST include a similar result table either 628 in its defining document or in a supplementary one. 630 2.5.1. DKIM and DomainKeys 632 DKIM is represented by the "dkim" method and is defined in [DKIM]. 633 DomainKeys is defined in [DOMAINKEYS] and is represented by the 634 "domainkeys" method. 636 A signature is "acceptable to the ADMD" if it passes local policy 637 checks (or there are no specific local policy checks). For example, 638 an ADMD policy might require that the signature(s) on the message be 639 added using the DNS domain present in the From: header field of the 640 message, thus making third-party signatures unacceptable even if they 641 verify. 643 Both DKIM and DomainKeys use the same result set, as follows: 645 none: The message was not signed. 647 pass: The message was signed, the signature or signatures were 648 acceptable to the ADMD, and the signature(s) passed verification 649 tests. 651 fail: The message was signed and the signature or signatures were 652 acceptable to the ADMD, but they failed the verification test(s). 654 policy: The message was signed but some aspect of the signature or 655 signatures were not acceptable to the ADMD. 657 neutral: The message was signed but the signature or signatures 658 contained syntax errors or were not otherwise able to be 659 processed. This result is also used for other failures not 660 covered elsewhere in this list. 662 temperror: The message could not be verified due to some error that 663 is likely transient in nature, such as a temporary inability to 664 retrieve a public key. A later attempt may produce a final 665 result. 667 permerror: The message could not be verified due to some error that 668 is unrecoverable, such as a required header field being absent. A 669 later attempt is unlikely to produce a final result. 671 [DKIM] advises that if a message fails verification, it is to be 672 treated as an unsigned message. A report of "fail" here permits the 673 receiver of the report to decide how to handle the failure. A report 674 of "neutral" or "none" preempts that choice, ensuring the message 675 will be treated as if it had not been signed. 677 2.5.2. SPF and Sender-ID 679 SPF and Sender ID use the "spf" and "sender-id" method names, 680 respectively. The result values for SPF are defined in Section 2.5 681 of [SPF], and those definitions are included here by reference: 683 +-----------+-------------------------------+ 684 | Code | Meaning | 685 +-----------+-------------------------------+ 686 | none | [RFC4408] Section 2.5.1 | 687 +-----------+-------------------------------+ 688 | pass | [RFC4408] Section 2.5.3 | 689 +-----------+-------------------------------+ 690 | fail | [RFC4408] Section 2.5.4 | 691 +-----------+-------------------------------+ 692 | softfail | [RFC4408] Section 2.5.5 | 693 +-----------+-------------------------------+ 694 | policy | [this document] Section 2.5.2 | 695 +-----------+-------------------------------+ 696 | neutral | [RFC4408] Section 2.5.2 | 697 +-----------+-------------------------------+ 698 | temperror | [RFC4408] Section 2.5.6 | 699 +-----------+-------------------------------+ 700 | permerror | [RFC4408] Section 2.5.7 | 701 +-----------+-------------------------------+ 703 These result codes are used in the context of this specification to 704 reflect the result returned by the component conducting SPF 705 evaluation. 707 Similarly, the results for Sender-ID are listed and described in 708 Section 4.2 of [SENDERID], which in turn uses the SPF definitions. 710 Note that both of those documents specify result codes that use mixed 711 case, but they are typically used all-lowercase in this context. 713 In both cases, an additional result of "policy" is defined, which 714 means the client was authorized to inject or relay mail on behalf of 715 the sender's DNS domain according to the authentication method's 716 algorithm, but local policy dictates that the result is unacceptable. 717 For example, "policy" might be used if SPF returns a "pass" result, 718 but a local policy check matches the sending DNS domain to one found 719 in an explicit list of unacceptable DNS domains (e.g., spammers). 721 If the retrieved sender policies used to evaluate SPF and Sender ID 722 do not contain explicit provisions for authenticating the local-part 723 (see Section 3.4.1 of [MAIL]) of an address, the "pvalue" reported 724 along with results for these mechanisms SHOULD NOT include the local- 725 part. 727 2.5.3. "iprev" 729 The result values are used by the "iprev" method, defined in 730 Section 3, are as follows: 732 pass: The DNS evaluation succeeded, i.e., the "reverse" and 733 "forward" lookup results were returned and were in agreement. 735 fail: The DNS evaluation failed. In particular, the "reverse" and 736 "forward" lookups each produced results but they were not in 737 agreement, or the "forward" query completed but produced no 738 result, e.g., a DNS RCODE of 3, commonly known as NXDOMAIN, or an 739 RCODE of 0 (NOERROR) in a reply containing no answers, was 740 returned. 742 temperror: The DNS evaluation could not be completed due to some 743 error that is likely transient in nature, such as a temporary DNS 744 error, e.g., a DNS RCODE of 2, commonly known as SERVFAIL, or 745 other error condition resulted. A later attempt may produce a 746 final result. 748 permerror: The DNS evaluation could not be completed because no PTR 749 data are published for the connecting IP address, e.g., a DNS 750 RCODE of 3, commonly known as NXDOMAIN, or an RCODE of 0 (NOERROR) 751 in a reply containing no answers, was returned. This prevented 752 completion of the evaluation. A later attempt is unlikely to 753 produce a final result. 755 There is no "none" for this method since any TCP connection 756 delivering email has an IP address associated with it, so some kind 757 of evaluation will always be possible. 759 For discussion of the format of DNS replies, see Domain Names - 760 Implementation And Specification ([DNS]). 762 2.5.4. SMTP AUTH 764 SMTP AUTH (defined in [AUTH]) is represented by the "auth" method, 765 and its result values are as follows: 767 none: SMTP authentication was not attempted. 769 pass: The SMTP client authenticated to the server reporting the 770 result using the protocol described in [AUTH]. 772 fail: The SMTP client attempted to authenticate to the server using 773 the protocol described in [AUTH] but was not successful, yet 774 continued to send the message about which a result is being 775 reported. 777 temperror: The SMTP client attempted to authenticate using the 778 protocol described in [AUTH] but was not able to complete the 779 attempt due to some error which is likely transient in nature, 780 such as a temporary directory service lookup error. A later 781 attempt may produce a final result. 783 permerror: The SMTP client attempted to authenticate using the 784 protocol described in [AUTH] but was not able to complete the 785 attempt due to some error that is likely not transient in nature, 786 such as a permanent directory service lookup error. A later 787 attempt is not likely produce a final result. 789 An agent making use of the data provided by this header field SHOULD 790 consider "fail" and "temperror" to be synonymous in terms of message 791 authentication, i.e., the client did not authenticate in either case. 793 2.5.5. Other Registered Codes 795 Result codes were also registered in other RFCs for Vouch By 796 Reference (in [AR-VBR], represented by "vbr"), Authorized Third-Party 797 Signatures (in [ATPS], represented by "dkim-atps"), and the DKIM- 798 related Author Domain Signing Practices (in [ADSP], represented by 799 "dkim-adsp"). 801 2.5.6. Extension Methods 803 Additional authentication method identifiers (extension methods) may 804 be defined in the future by later revisions or extensions to this 805 specification. These method identifiers are registered with the 806 Internet Assigned Numbers Authority (IANA) and, preferably, published 807 in an RFC. See Section 6 for further details. 809 Extension methods can be defined for the following reasons: 811 1. To allow additional information from new authentication systems 812 to be communicated to MUAs or downstream filters. The names of 813 such identifiers ought to reflect the name of the method being 814 defined, but ought not be needlessly long. 816 2. To allow the creation of "sub-identifiers" that indicate 817 different levels of authentication and differentiate between 818 their relative strengths, e.g., "auth1-weak" and "auth1-strong". 820 Authentication method implementers are encouraged to provide adequate 821 information, via message header field comments if necessary, to allow 822 an MUA developer to understand or relay ancillary details of 823 authentication results. For example, if it might be of interest to 824 relay what data was used to perform an evaluation, such information 825 could be relayed as a comment in the header field, such as: 827 Authentication-Results: example.com; 828 foo=pass bar.baz=blob (2 of 3 tests OK) 830 Experimental method identifiers MUST only be used within ADMDs that 831 have explicitly consented to use them. These method identifiers and 832 the parameters associated with them are not documented in RFCs. 833 Therefore, they are subject to change at any time and not suitable 834 for production use. Any MTA, MUA, or downstream filter intended for 835 production use SHOULD ignore or delete any Authentication-Results 836 header field that includes an experimental (unknown) method 837 identifier. 839 2.5.7. Extension Result Codes 841 Additional result codes (extension results) might be defined in the 842 future by later revisions or extensions to this specification. 843 Result codes MUST be registered with the Internet Assigned Numbers 844 Authority (IANA) and preferably published in an RFC. See Section 6 845 for further details. 847 Extension results MUST only be used within ADMDs that have explicitly 848 consented to use them. These results and the parameters associated 849 with them are not formally documented. Therefore, they are subject 850 to change at any time and not suitable for production use. Any MTA, 851 MUA or downstream filter intended for production use SHOULD ignore or 852 delete any Authentication-Results header field that includes an 853 extension result. 855 3. The "iprev" Authentication Method 857 This section defines an additional authentication method called 858 "iprev". 860 "iprev" is an attempt to verify that a client appears to be valid 861 based on some DNS queries, which is to say that the IP address is 862 explicitly associated with a domain name. Upon receiving a session 863 initiation of some kind from a client, the IP address of the client 864 peer is queried for matching names (i.e., a number-to-name 865 translation, also known as a "reverse lookup" or a "PTR" record 866 query). Once that result is acquired, a lookup of each of the names 867 (i.e., a name-to-number translation, or an "A" or "AAAA" record 868 query) thus retrieved is done. The response to this second check 869 will typically result in at least one mapping back to the client's IP 870 address. 872 Expressed as an algorithm: If the client peer's IP address is I, the 873 list of names to which I maps (after a "PTR" query) is the set N, and 874 the union of IP addresses to which each member of N maps (after 875 corresponding "A" and "AAAA" queries) is L, then this test is 876 successful if I is an element of L. 878 The response to a PTR query could contain multiple names. To prevent 879 heavy DNS loads, agents performing these queries MUST be implemented 880 such that the number of names evaluated by generation of 881 corresponding A or AAAA queries is limited so as not unduly taxing to 882 the DNS infrastructure, though it MAY be configurable by an 883 administrator. As an example, Section 5.5 of [SPF] chose a limit of 884 10 for its implementation of this algorithm. 886 DNS Extensions to Support IP Version 6 ([DNS-IP6]) discusses the 887 query formats for the IPv6 case. 889 There is some contention regarding the wisdom and reliability of this 890 test. For example, in some regions it can be difficult for this test 891 ever to pass because the practice of arranging to match the forward 892 and reverse DNS is infrequently observed. Therefore, the precise 893 implementation details of how a verifier performs an "iprev" test are 894 not specified here. The verifier MAY report a successful or failed 895 "iprev" test at its discretion having done some kind of check of the 896 validity of the connection's identity using DNS. It is incumbent 897 upon an agent making use of the reported "iprev" result to understand 898 what exactly that particular verifier is attempting to report. 900 Extensive discussion of reverse DNS mapping and its implications can 901 be found in Considerations for the use of DNS Reverse Mapping 902 ([DNSOP-REVERSE]). In particular, it recommends that applications 903 avoid using this test as a means of authentication or security. Its 904 presence in this document is not an endorsement, but is merely 905 acknowledgement that the method remains common and provides the means 906 to relay the results of that test. 908 4. Adding the Header Field to a Message 910 This specification makes no attempt to evaluate the relative 911 strengths of various message authentication methods that may become 912 available. The methods listed are an order-independent set; their 913 sequence does not indicate relative strength or importance of one 914 method over another. Instead, the MUA or downstream filter consuming 915 this header field is to interpret the result of each method based on 916 its own knowledge of what that method evaluates. 918 Each "method" MUST refer to an authentication method declared in the 919 IANA registry, or an extension method as described in Section 2.5.6, 920 and each "result" MUST refer to a result code declared in the IANA 921 registry, or an extension result code as defined in Section 2.5.7. 922 See Section 6 for further information about the registered methods 923 and result codes. 925 An MTA compliant with this specification adds this header field 926 (after performing one or more message authentication tests) to 927 indicate which MTA or ADMD performed the test, which test got 928 applied, and what the result was. If an MTA applies more than one 929 such test, it adds this header field either once per test, or once 930 indicating all of the results. An MTA MUST NOT add a result to an 931 existing header field. 933 An MTA MAY add this header field containing only the authentication 934 identifier portion and the "none" token (see Section 2.2) to indicate 935 explicitly that no message authentication schemes were applied prior 936 to delivery of this message. 938 An MTA adding this header field has to take steps to identify it as 939 legitimate to the MUAs or downstream filters that will ultimately 940 consume its content. One process to do so is described in Section 5. 941 Further measures may be necessary in some environments. Some 942 possible solutions are enumerated in Section 8.1. This document does 943 not mandate any specific solution to this issue as each environment 944 has its own facilities and limitations. 946 Most known message authentication methods focus on a particular 947 identifier to evaluate. SPF and Sender ID differ in that they can 948 yield a result based on more than one identifier; specifically, SPF 949 can evaluate the RFC5321.Helo parameter or the RFC5321.MailFrom 950 parameter, and Sender ID can evaluate the RFC5321.MailFrom parameter 951 or the PRA identity. When generating this field to report those 952 results, only the parameter that yielded the result is included. 954 For MTAs that add this header field, adding header fields in order 955 (at the top), per Section 3.6 of [MAIL], is particularly important. 956 Moreover, this header field SHOULD be inserted above any other trace 957 header fields such MTAs might prepend. This placement allows easy 958 detection of header fields that can be trusted. 960 End users making direct use of this header field might inadvertently 961 trust information that has not been properly vetted. If, for 962 example, a basic SPF result were to be relayed that claims an 963 authenticated addr-spec, the local-part of that addr-spec has 964 actually not been authenticated. Thus, an MTA adding this header 965 field SHOULD NOT include any data that has not been authenticated by 966 the method(s) being applied. Moreover, MUAs SHOULD NOT render to 967 users such information if it is presented by a method known not to 968 authenticate it. 970 4.1. Header Field Position and Interpretation 972 In order to ensure non-ambiguous results and avoid the impact of 973 false header fields, MUAs and downstream filters SHOULD NOT interpret 974 this header field unless specifically configured to do so by the user 975 or administrator. That is, this interpretation should not be "on by 976 default". Naturally then, users or administrators ought not activate 977 such a feature unless they are certain the header field will be 978 validly added by an agent within the ADMD that accepts the mail that 979 is ultimately read by the MUA, and instances of the header field 980 appearing to originate within the ADMD but are actually added by 981 foreign MTAs will be removed before delivery. 983 Furthermore, MUAs and downstream filters SHOULD NOT interpret this 984 header field unless the authentication service identifier it bears 985 appears to be one used within its own ADMD as configured by the user 986 or administrator. 988 MUAs and downstream filters MUST ignore any result reported using a 989 "result" not specified in the result code registry, or a "ptype" not 990 listed in the corresponding registry for such values as defined in 991 Section 6. Moreover, such agents MUST ignore a result indicated for 992 any "method" they do not specifically support. 994 An MUA SHOULD NOT reveal these results to end users, absent careful 995 human factors design considerations and testing, for the presentation 996 of trust related materials. For example, an attacker could register 997 examp1e.com (note the digit "one") and send signed mail to intended 998 victims; a verifier would detect that the signature was valid and 999 report a "pass" even though it's clear the DNS domain name was 1000 intended to mislead. See Section 8.2 for further discussion. 1002 As stated in Section 2.1, this header field MUST be treated as though 1003 it were a trace header field as defined in Section 3.6.7 of [MAIL], 1004 and hence MUST NOT be reordered and MUST be prepended to the message, 1005 so that there is generally some indication upon delivery of where in 1006 the chain of handling MTAs the message authentication was done. 1008 Note that there are a few message handlers that are only capable of 1009 appending new header fields to a message. Strictly speaking, these 1010 handlers are not compliant with this specification. They can still 1011 add the header field to carry authentication details, but any signal 1012 about where in the handling chain the work was done may be lost. 1013 Consumers SHOULD be designed such that this can be tolerated, 1014 especially from a producer known to have this limitation. 1016 MUAs SHOULD ignore instances of this header field discovered within 1017 message/rfc822 MIME attachments. 1019 Further discussion of these topics can be found in Section 8 below. 1021 4.2. Local Policy Enforcement 1023 Some sites have a local policy that considers any particular 1024 authentication policy's non-recoverable failure results (typically 1025 "fail" or similar) as justification for rejecting the message. In 1026 such cases, the border MTA SHOULD issue an SMTP rejection response to 1027 the message, rather than adding this header field and allowing the 1028 message to proceed toward delivery. This is more desirable than 1029 allowing the message to reach an internal host's MTA or spam filter, 1030 thus possibly generating a local rejection such as a [DSN] to a 1031 forged originator. Such generated rejections are colloquially known 1032 as "backscatter". 1034 The same MAY also be done for local policy decisions overriding the 1035 results of the authentication methods (e.g., the "policy" result 1036 codes described in Section 2.5). 1038 Such rejections at the SMTP protocol level are not possible if local 1039 policy is enforced at the MUA and not the MTA. 1041 5. Removing Existing Header Fields 1043 For security reasons, any MTA conforming to this specification MUST 1044 delete any discovered instance of this header field that claims, by 1045 virtue of its authentication service identifier, to have been added 1046 within its trust boundary but that did not come directly from another 1047 trusted MTA. For example, an MTA for example.com receiving a message 1048 MUST delete or otherwise obscure any instance of this header field 1049 bearing an authentication service identifier indicating the header 1050 field was added within example.com prior to adding its own header 1051 fields. This could mean each MTA will have to be equipped with a 1052 list of internal MTAs known to be compliant (and hence trustworthy). 1054 For simplicity and maximum security, a border MTA could remove all 1055 instances of this header field on mail crossing into its trust 1056 boundary. However, this may conflict with the desire to access 1057 authentication results performed by trusted external service 1058 providers. It may also invalidate signed messages whose signatures 1059 cover external instances of this header field. A more robust border 1060 MTA could allow a specific list of authenticating MTAs whose 1061 information is to be admitted, removing the header field originating 1062 from all others. 1064 As stated in Section 1.2, a formal definition of "trust boundary" is 1065 deliberately not made here. It is entirely possible that a border 1066 MTA for example.com will explicitly trust authentication results 1067 asserted by upstream host example.net even though they exist in 1068 completely disjoint administrative boundaries. In that case, the 1069 border MTA MAY elect not to delete those results; moreover, the 1070 upstream host doing some authentication work could apply a signing 1071 technology such as [DKIM] on its own results to assure downstream 1072 hosts of their authenticity. An example of this is provided in 1073 Appendix C. 1075 Similarly, in the case of messages signed using [DKIM] or other 1076 message signing methods that sign header fields, this removal action 1077 could invalidate one or more signatures on the message if they 1078 covered the header field to be removed. This behavior can be 1079 desirable since there's little value in validating the signature on a 1080 message with forged header fields. However, signing agents MAY 1081 therefore elect to omit these header fields from signing to avoid 1082 this situation. 1084 An MTA SHOULD remove any instance of this header field bearing a 1085 version (express or implied) that it does not support. However, an 1086 MTA MUST remove such a header field if the [SMTP] connection relaying 1087 the message is not from a trusted internal MTA. This means the MTA 1088 needs to be able to understand versions of this header field at last 1089 as late as the ones understood by the MUAs or other consumers within 1090 its ADMD. 1092 6. IANA Considerations 1094 IANA has registered the defined header field and created two tables 1095 as described below. These registry actions were originally defined 1096 by [RFC5451] and are repeated here to provide a single, current 1097 reference. 1099 6.1. The Authentication-Results Header Field 1101 [RFC5451] added the "Authentication-Results" header field to the IANA 1102 Permanent Message Header Field Registry, per the procedure found in 1103 [IANA-HEADERS]. That entry is to be updated to reference this 1104 document. The following is the registration template: 1106 Header field name: Authentication-Results 1107 Applicable protocol: mail ([MAIL]) 1108 Status: Standard 1109 Author/Change controller: IETF 1110 Specification document(s): [this memo] 1111 Related information: 1112 Requesting review of any proposed changes and additions to 1113 this field is recommended. 1115 6.2. Email Authentication Method Name Registry 1117 Names of message authentication methods supported by this 1118 specification are to be registered with IANA, with the exception of 1119 experimental names as described in Section 2.5.6. A registry was 1120 created by [RFC5451] for this purpose. This document changes the 1121 rules governing that registry. 1123 New entries are assigned only for values that have received Expert 1124 Review, per [IANA-CONSIDERATIONS]. The Designated Expert shall be 1125 appointed by the IESG. The Designated Expert has discretion to 1126 request that a publication be referenced if a clear, concise 1127 definition of the authentication method cannot be provided such that 1128 interoperability is assured. Registrations should otherwise be 1129 permitted. The Designated Expert can also handle requests to mark 1130 any current registration as "deprecated". 1132 Each method must register a name, the specification that defines it, 1133 a version number associated with the method being registered 1134 (preferably starting at "1"), and zero or more "ptype" values 1135 appropriate for use with that method, which "property" value(s) 1136 should be reported by that method, and a description of the "value" 1137 to be used with each. 1139 All existing registry entries that reference [RFC5451] are to be 1140 updated to reference this document. [RFC Editor note: Section 1141 numbers may have to change as well since they appear in the registry, 1142 but numbering may change between now and publication. We can deal 1143 with this during the IANA phase and/or AUTH48.]. 1145 IANA is also requested to add a "version" field to all existing 1146 registry entries. All current methods are to be recorded as version 1147 "1". 1149 6.3. Email Authentication Result Names Registry 1151 Names of message authentication result codes supported by this 1152 specification must be registered with IANA, with the exception of 1153 experimental codes as described in Section 2.5.7. A registry was 1154 created by [RFC5451] for this purpose. This document changes the 1155 rules governing that registry. 1157 New entries are assigned only for values that have received Expert 1158 Review, per [IANA-CONSIDERATIONS]. The Designated Expert shall be 1159 appointed by the IESG. The Designated Expert has discretion to 1160 request that a publication be referenced if a clear, concise 1161 definition of the authentication result cannot be provided such that 1162 interoperability is assured. Registrations should otherwise be 1163 permitted. The Designated Expert can also handle requests to mark 1164 any current registration as "deprecated". 1166 All existing registry entries that reference [RFC5451] are to be 1167 updated to reference this document. 1169 The definitions for the SPF and Sender ID authentication methods are 1170 updated using the references found in Section 2.5.2. 1172 7. Implementation Status 1174 [RFC Editor: Please delete this section prior to publication.] 1176 This section records the status of known implementations of the 1177 protocol defined by this specification at the time of posting of this 1178 Internet-Draft, and is based on a proposal described in 1179 draft-sheffer-running-code. The description of implementations in 1180 this section is intended to assist the IETF in its decision processes 1181 in progressing drafts to RFCs. Please note that the listing of any 1182 individual implementation here does not imply endorsement by the 1183 IETF. Furthermore, no effort has been spent to verify the 1184 information presented here that was supplied by IETF contributors. 1186 This is not intended as, and must not be construed to be, a catalog 1187 of available implementations or their features. Readers are advised 1188 to note that other implementations may exist. 1190 According to draft-sheffer-running-code, "this will allow reviewers 1191 and working groups to assign due consideration to documents that have 1192 the benefit of running code, by considering the running code as 1193 evidence of valuable experimentation and feedback that has made the 1194 implemented protocols more mature. It is up to the individual 1195 working groups to use this information as they see fit". 1197 7.1. Google Mail 1199 Responsible Organization: Google 1201 Implementation: Gmail; http://mail.google.com 1203 Brief Description: Gmail is a popular, free, web-based mailbox 1204 service provider 1206 Maturity level: widely used 1208 Coverage: all syntax required to report SPF and DKIM results 1210 Licensing: proprietary 1212 Implementation Experience: n/a 1214 Contact Information: http://mail.google.com 1216 7.2. Yahoo! Mail 1218 Responsible Organization: Yahoo!, Inc. 1220 Implementation: Yahoo! Mail; http://mail.yahoo.com 1222 Brief Description: Yahoo! Mail is a popular, free, web-based 1223 mailbox service provider 1225 Maturity level: widely used 1227 Coverage: all syntax required to report SPF and DKIM results 1229 Licensing: proprietary 1230 Implementation Experience: n/a 1232 Contact Information: http://mail.yahoo.com 1234 7.3. Hotmail 1236 Responsible Organization: Microsoft Corp. 1238 Implementation: Hotmail; http://www.hotmail.com 1240 Brief Description: Hotmail is a popular, free, web-based mailbox 1241 service provider 1243 Maturity level: widely used 1245 Coverage: all syntax required to report SPF and DKIM results 1247 Licensing: proprietary 1249 Implementation Experience: n/a 1251 Contact Information: http://www.hotmail.com 1253 7.4. Courier MTA 1255 Responsible Organization: Double Precision, Inc. 1257 Implementation: Courier MTA; http://www.courier-mta.org 1259 Brief Description: Courier MTA is an open source mail server 1261 Maturity level: production 1263 Coverage: all syntax required to report SPF and DKIM results 1265 Licensing: GPL 1267 Implementation Experience: n/a 1269 Contact Information: http://www.courier-mta.org 1271 7.5. sid-milter 1273 Responsible Organization: Sendmail, Inc. 1275 Implementation: sid-milter; 1276 http://sourceforge.net/projects/sid-milter 1278 Brief Description: sid-milter is an open source MTA plugin that 1279 implements both Sender ID and SPF 1281 Maturity level: production (no longer maintained) 1283 Coverage: all syntax required to report SPF and Sender ID results 1285 Licensing: Sendmail Open Source License 1287 Implementation Experience: n/a 1289 Contact Information: http://www.sendmail.com 1291 7.6. opendkim 1293 Responsible Organization: The Trusted Domain Project 1295 Implementation: opendkim; http://www.opendkim.org 1297 Brief Description: opendkim includes a library that implements DKIM 1298 and an MTA plugin that uses this library to provide DKIM and 1299 related services 1301 Maturity level: widely used (Facebook, AOL, etc.) 1303 Coverage: all syntax required to report DKIM, VBR, and some 1304 extension results 1306 Licensing: BSD two-clause license 1308 Implementation Experience: n/a 1310 Contact Information: http://www.trusteddomain.org 1312 7.7. opendmarc 1314 Responsible Organization: The Trusted Domain Project 1316 Implementation: opendmarc; http://www.trusteddomain.org/opendmarc 1318 Brief Description: opendmarc includes a library that implements 1319 DMARC and an MTA plugin that uses this library to provide DMARC 1320 and related services 1322 Maturity level: production 1324 Coverage: all syntax required to report DMARC results 1326 Licensing: BSD two-clause license 1328 Implementation Experience: n/a 1330 Contact Information: http://www.trusteddomain.org 1332 7.8. authres 1334 Responsible Organization: Julian Mehnle, Scott Kitterman 1336 Implementation: authres; https://pypi.python.org/pypi/authres/ 1338 Brief Description: authres is a python module for parsing and 1339 generating the Authentication-Results header field. It is used by 1340 "pyspf", a python module that implements the SPF validation 1341 service. 1343 Maturity level: production 1345 Coverage: supports all aspects of the protocol, including one new 1346 method that is pending publication 1348 Licensing: Apache 2.0 License 1350 Implementation Experience: n/a 1352 Contact Information: 1353 https://launchpad.net/authentication-results-python 1355 8. Security Considerations 1357 The following security considerations apply when adding or processing 1358 the "Authentication-Results" header field: 1360 8.1. Forged Header Fields 1362 An MUA or filter that accesses a mailbox whose messages are handled 1363 by a non-conformant MTA, and understands Authentication-Results 1364 header fields, could potentially make false conclusions based on 1365 forged header fields. A malicious user or agent could forge a header 1366 field using the DNS domain of a receiving ADMD as the authserv-id 1367 token in the value of the header field, and with the rest of the 1368 value claim that the message was properly authenticated. The non- 1369 conformant MTA would fail to strip the forged header field, and the 1370 MUA could inappropriately trust it. 1372 For this reason, it is best not to have processing of the 1373 "Authentication-Results" header field enabled by default; instead it 1374 should be ignored, at least for the purposes of enacting filtering 1375 decisions, unless specifically enabled by the user or administrator 1376 after verifying that the border MTA is compliant. It is acceptable 1377 to have an MUA aware of this specification, but have an explicit list 1378 of hostnames whose "Authentication-Results" header fields are 1379 trustworthy; however, this list should initially be empty. 1381 Proposed alternative solutions to this problem were made some time 1382 ago, and are listed below. To date, they have not been developed due 1383 to lack of demand, but are documented here should the information be 1384 useful at some point in the future: 1386 1. Possibly the simplest is a digital signature protecting the 1387 header field, such as using [DKIM], that can be verified by an 1388 MUA by using a posted public key. Although one of the main 1389 purposes of this document is to relieve the burden of doing 1390 message authentication work at the MUA, this only requires that 1391 the MUA learn a single authentication scheme even if a number of 1392 them are in use at the border MTA. Note that [DKIM] requires 1393 that the From header field be signed, although in this 1394 application, the signing agent (a trusted MTA) likely cannot 1395 authenticate that value, so the fact that it is signed should be 1396 ignored. Where the authserv-id is the ADMD's domain name, the 1397 authserv-id matching this valid internal signature's "d=" DKIM 1398 value is sufficient. 1400 2. Another would be a means to interrogate the MTA that added the 1401 header field to see if it is actually providing any message 1402 authentication services and saw the message in question, but this 1403 isn't especially palatable given the work required to craft and 1404 implement such a scheme. 1406 3. Yet another might be a method to interrogate the internal MTAs 1407 that apparently handled the message (based on Received: header 1408 fields) to determine whether any of them conform to Section 5 of 1409 this memo. This, too, has potentially high barriers to entry. 1411 4. Extensions to [IMAP], [SMTP], and [POP3] could be defined to 1412 allow an MUA or filtering agent to acquire the "authserv-id" in 1413 use within an ADMD, thus allowing it to identify which 1414 Authentication-Results header fields it can trust. 1416 5. On the presumption that internal MTAs are fully compliant with 1417 Section 3.6 of [MAIL], and the compliant internal MTAs are using 1418 their own host names or the ADMD's DNS domain name as the 1419 "authserv-id" token, the header field proposed here should always 1420 appear above a Received: header added by a trusted MTA. This can 1421 be used as a test for header field validity. 1423 Support for some of these is being considered for future work. 1425 In any case, a mechanism needs to exist for an MUA or filter to 1426 verify that the host that appears to have added the header field (a) 1427 actually did so, and (b) is legitimately adding that header field for 1428 this delivery. Given the variety of messaging environments deployed 1429 today, consensus appears to be that specifying a particular mechanism 1430 for doing so is not appropriate for this document. 1432 Mitigation of the forged header field attack can also be accomplished 1433 by moving the authentication results data into meta-data associated 1434 with the message. In particular, an [SMTP] extension could be 1435 established that is used to communicate authentication results from 1436 the border MTA to intermediate and delivery MTAs; the latter of these 1437 could arrange to store the authentication results as meta-data 1438 retrieved and rendered along with the message by an [IMAP] client 1439 aware of a similar extension in that protocol. The delivery MTA 1440 would be told to trust data via this extension only from MTAs it 1441 trusts, and border MTAs would not accept data via this extension from 1442 any source. There is no vector in such an arrangement for forgery of 1443 authentication data by an outside agent. 1445 8.2. Misleading Results 1447 Until some form of service for querying the reputation of a sending 1448 agent is widely deployed, the existence of this header field 1449 indicating a "pass" does not render the message trustworthy. It is 1450 possible for an arriving piece of spam or other undesirable mail to 1451 pass checks by several of the methods enumerated above (e.g., a piece 1452 of spam signed using [DKIM] by the originator of the spam, which 1453 might be a spammer or a compromised system). In particular, this 1454 issue is not resolved by forged header field removal discussed above. 1456 Hence, MUAs and downstream filters must take some care with use of 1457 this header even after possibly malicious headers are scrubbed. 1459 8.3. Header Field Position 1461 Despite the requirements of [MAIL], header fields can sometimes be 1462 reordered enroute by intermediate MTAs. The goal of requiring header 1463 field addition only at the top of a message is an acknowledgement 1464 that some MTAs do reorder header fields, but most do not. Thus, in 1465 the general case, there will be some indication of which MTAs (if 1466 any) handled the message after the addition of the header field 1467 defined here. 1469 8.4. Reverse IP Query Denial-of-Service Attacks 1471 Section 5.5 of [SPF] describes a DNS-based denial-of-service attack 1472 for verifiers that attempt DNS-based identity verification of 1473 arriving client connections. A verifier wishing to do this check and 1474 report this information needs to take care not to go to unbounded 1475 lengths to resolve "A" and "PTR" queries. MUAs or other filters 1476 making use of an "iprev" result specified by this document need to be 1477 aware of the algorithm used by the verifier reporting the result and, 1478 especially, its limitations. 1480 8.5. Mitigation of Backscatter 1482 Failing to follow the instructions of Section 4.2 can result in a 1483 denial-of-service attack caused by the generation of [DSN] messages 1484 (or equivalent) to addresses that did not send the messages being 1485 rejected. 1487 8.6. Internal MTA Lists 1489 Section 5 describes a procedure for scrubbing header fields that may 1490 contain forged authentication results about a message. A compliant 1491 installation will have to include, at each MTA, a list of other MTAs 1492 known to be compliant and trustworthy. Failing to keep this list 1493 current as internal infrastructure changes may expose an ADMD to 1494 attack. 1496 8.7. Attacks against Authentication Methods 1498 If an attack becomes known against an authentication method, clearly 1499 then the agent verifying that method can be fooled into thinking an 1500 inauthentic message is authentic, and thus the value of this header 1501 field can be misleading. It follows that any attack against the 1502 authentication methods supported by this document is also a security 1503 consideration here. 1505 8.8. Intentionally Malformed Header Fields 1507 It is possible for an attacker to add an Authentication-Results 1508 header field that is extraordinarily large or otherwise malformed in 1509 an attempt to discover or exploit weaknesses in header field parsing 1510 code. Implementers must thoroughly verify all such header fields 1511 received from MTAs and be robust against intentionally as well as 1512 unintentionally malformed header fields. 1514 8.9. Compromised Internal Hosts 1516 An internal MUA or MTA that has been compromised could generate mail 1517 with a forged From header field and a forged Authentication-Results 1518 header field that endorses it. Although it is clearly a larger 1519 concern to have compromised internal machines than it is to prove the 1520 value of this header field, this risk can be mitigated by arranging 1521 that internal MTAs will remove this header field if it claims to have 1522 been added by a trusted border MTA (as described above), yet the 1523 [SMTP] connection is not coming from an internal machine known to be 1524 running an authorized MTA. However, in such a configuration, 1525 legitimate MTAs will have to add this header field when legitimate 1526 internal-only messages are generated. This is also covered in 1527 Section 5. 1529 8.10. Encapsulated Instances 1531 MIME messages can contain attachments of type "message/rfc822", which 1532 contain other messages. Such an encapsulated message can also 1533 contain an Authentication-Results header field. Although the 1534 processing of these is outside of the intended scope of this document 1535 (see Section 1.3), some early guidance to MUA developers is 1536 appropriate here. 1538 Since MTAs are unlikely to strip Authentication-Results header fields 1539 after mailbox delivery, MUAs are advised in Section 4.1 to ignore 1540 such instances within MIME attachments. Moreover, when extracting a 1541 message digest to separate mail store messages or other media, such 1542 header fields should be removed so that they will never be 1543 interpreted improperly by MUAs that might later consume them. 1545 8.11. Reverse Mapping 1547 Although Section 3 of this memo includes explicit support for the 1548 "iprev" method, its value as an authentication mechanism is limited. 1549 Implementers of both this proposal and agents that use the data it 1550 relays are encouraged to become familiar with the issues raised by 1551 [DNSOP-REVERSE] when deciding whether or not to include support for 1552 "iprev". 1554 9. References 1556 9.1. Normative References 1558 [ABNF] Crocker, D. and P. Overell, "Augmented BNF for 1559 Syntax Specifications: ABNF", STD 68, 1560 RFC 5234, January 2008. 1562 [IANA-HEADERS] Klyne, G., Nottingham, M., and J. Mogul, 1563 "Registration Procedures for Message Header 1564 Fields", BCP 90, RFC 3864, September 2004. 1566 [KEYWORDS] Bradner, S., "Key words for use in RFCs to 1567 Indicate Requirement Levels", BCP 14, 1568 RFC 2119, March 1997. 1570 [MAIL] Resnick, P., Ed., "Internet Message Format", 1571 RFC 5322, October 2008. 1573 [MIME] Freed, N. and N. Borenstein, "Multipurpose 1574 Internet Mail Extensions (MIME) Part One: 1575 Format of Internet Message Bodies", RFC 2045, 1576 November 1996. 1578 [SMTP] Klensin, J., "Simple Mail Transfer Protocol", 1579 RFC 5321, October 2008. 1581 9.2. Informative References 1583 [ADSP] Allman, E., Fenton, J., Delany, M., and J. 1584 Levine, "DomainKeys Identified Mail (DKIM) 1585 Author Domain Signing Practices (ADSP)", 1586 RFC 5617, August 2009. 1588 [AR-VBR] Kucherawy, M., "Authentication-Results 1589 Registration for Vouch by Reference Results", 1590 RFC 6212, April 2011. 1592 [ATPS] Kucherawy, M., "DomainKeys Identified Mail 1593 (DKIM) Authorized Third-Party Signatures", 1594 RFC 6541, February 2012. 1596 [AUTH] Siemborski, R. and A. Melnikov, "SMTP Service 1597 Extension for Authentication", RFC 4954, 1598 July 2007. 1600 [DKIM] Crocker, D., Hansen, T., and M. Kucherawy, 1601 "DomainKeys Identified Mail (DKIM) 1602 Signatures", RFC 6376, September 2011. 1604 [DNS] Mockapetris, P., "Domain names - 1605 implementation and specification", STD 13, 1606 RFC 1035, November 1987. 1608 [DNS-IP6] Thomson, S., Huitema, C., Ksinant, V., and M. 1609 Souissi, "DNS Extensions to Support IP Version 1610 6", RFC 3596, October 2003. 1612 [DNSOP-REVERSE] Senie, D. and A. Sullivan, "Considerations for 1613 the use of DNS Reverse Mapping", Work 1614 in Progress, March 2008. 1616 [DOMAINKEYS] Delany, M., "Domain-Based Email Authentication 1617 Using Public Keys Advertised in the DNS 1618 (DomainKeys)", RFC 4870, May 2007. 1620 [DSN] Moore, K. and G. Vaudreuil, "An Extensible 1621 Message Format for Delivery Status 1622 Notifications", RFC 3464, January 2003. 1624 [EMAIL-ARCH] Crocker, D., "Internet Mail Architecture", 1625 RFC 5598, October 2008. 1627 [IANA-CONSIDERATIONS] Narten, T. and H. Alvestrand, "Guidelines for 1628 Writing an IANA Considerations Section in 1629 RFCs", BCP 26, RFC 5226, May 2008. 1631 [IMAP] Crispin, M., "INTERNET MESSAGE ACCESS PROTOCOL 1632 - VERSION 4rev1", RFC 3501, March 2003. 1634 [POP3] Myers, J. and M. Rose, "Post Office Protocol - 1635 Version 3", STD 53, RFC 1939, May 1996. 1637 [RFC5451] Kucherawy, M., "Message Header Field for 1638 Indicating Message Authentication Status", 1639 RFC 5451, April 2009. 1641 [SECURITY] Rescorla, E. and B. Korver, "Guidelines for 1642 Writing RFC Text on Security Considerations", 1643 BCP 72, RFC 3552, July 2003. 1645 [SENDERID] Lyon, J. and M. Wong, "Sender ID: 1646 Authenticating E-Mail", RFC 4406, April 2006. 1648 [SPF] Wong, M. and W. Schlitt, "Sender Policy 1649 Framework (SPF) for Authorizing Use of Domains 1650 in E-Mail, Version 1", RFC 4408, April 2006. 1652 [VBR] Hoffman, P., Levine, J., and A. Hathcock, 1653 "Vouch By Reference", RFC 5518, April 2009. 1655 Appendix A. Acknowledgements 1657 The author wishes to acknowledge the following for their review and 1658 constructive criticism of this update: Dave Cridland, Dave Crocker, 1659 Bjoern Hoehrmann, Scott Kitterman, John Levine, Alexey Melnikov, S. 1660 Moonesamy, and Alessandro Vesely. 1662 Appendix B. Legacy MUAs 1664 Implementers of this protocol should be aware that many MUAs are 1665 unlikely to be retrofitted to support the new header field and its 1666 semantics. In the interests of convenience and quicker adoption, a 1667 delivery MTA might want to consider adding things that are processed 1668 by existing MUAs in addition to the Authentication-Results header 1669 field. One suggestion is to include a Priority header field, on 1670 messages that don't already have such a header field, containing a 1671 value that reflects the strength of the authentication that was 1672 accomplished, e.g., "low" for weak or no authentication, "normal" or 1673 "high" for good or strong authentication. 1675 Some modern MUAs can already filter based on the content of this 1676 header field. However, there is keen interest in having MUAs make 1677 some kind of graphical representation of this header field's meaning 1678 to end users. Until this capability is added, other interim means of 1679 conveying authentication results may be necessary while this proposal 1680 and its successors are adopted. 1682 Appendix C. Authentication-Results Examples 1684 This section presents some examples of the use of this header field 1685 to indicate authentication results. 1687 C.1. Trivial Case; Header Field Not Present 1689 The trivial case: 1691 Received: from mail-router.example.com 1692 (mail-router.example.com [192.0.2.1]) 1693 by server.example.org (8.11.6/8.11.6) 1694 with ESMTP id g1G0r1kA003489; 1695 Fri, Feb 15 2002 17:19:07 -0800 1696 From: sender@example.com 1697 Date: Fri, Feb 15 2002 16:54:30 -0800 1698 To: receiver@example.org 1699 Message-Id: <12345.abc@example.com> 1700 Subject: here's a sample 1702 Hello! Goodbye! 1704 Example 1: Trivial case 1706 The "Authentication-Results" header field is completely absent. The 1707 MUA may make no conclusion about the validity of the message. This 1708 could be the case because the message authentication services were 1709 not available at the time of delivery, or no service is provided, or 1710 the MTA is not in compliance with this specification. 1712 C.2. Nearly Trivial Case; Service Provided, But No Authentication Done 1714 A message that was delivered by an MTA that conforms to this 1715 specification but provides no actual message authentication service: 1717 Authentication-Results: example.org 1; none 1718 Received: from mail-router.example.com 1719 (mail-router.example.com [192.0.2.1]) 1720 by server.example.org (8.11.6/8.11.6) 1721 with ESMTP id g1G0r1kA003489; 1722 Fri, Feb 15 2002 17:19:07 -0800 1723 From: sender@example.com 1724 Date: Fri, Feb 15 2002 16:54:30 -0800 1725 To: receiver@example.org 1726 Message-Id: <12345.abc@example.com> 1727 Subject: here's a sample 1729 Hello! Goodbye! 1731 Example 2: Header present but no authentication done 1733 The "Authentication-Results" header field is present, showing that 1734 the delivering MTA conforms to this specification. It used its DNS 1735 domain name as the authserv-id. The presence of "none" (and the 1736 absence of any method and result tokens) indicates that no message 1737 authentication was done. The version number of the specification to 1738 which the field's content conforms is explicitly provided. 1740 C.3. Service Provided, Authentication Done 1742 A message that was delivered by an MTA that conforms to this 1743 specification and applied some message authentication: 1745 Authentication-Results: example.com; 1746 spf=pass smtp.mailfrom=example.net 1747 Received: from dialup-1-2-3-4.example.net 1748 (dialup-1-2-3-4.example.net [192.0.2.200]) 1749 by mail-router.example.com (8.11.6/8.11.6) 1750 with ESMTP id g1G0r1kA003489; 1751 Fri, Feb 15 2002 17:19:07 -0800 1752 From: sender@example.net 1753 Date: Fri, Feb 15 2002 16:54:30 -0800 1754 To: receiver@example.com 1755 Message-Id: <12345.abc@example.net> 1756 Subject: here's a sample 1758 Hello! Goodbye! 1760 Example 3: Header reporting results 1762 The "Authentication-Results" header field is present, indicating that 1763 the border MTA conforms to this specification. The authserv-id is 1764 once again the DNS domain name. Furthermore, the message was 1765 authenticated by that MTA via the method specified in [SPF]. Note 1766 that since that method cannot authenticate the local-part, it has 1767 been omitted from the result's value. The MUA could extract and 1768 relay this extra information if desired. 1770 C.4. Service Provided, Several Authentications Done, Single MTA 1772 A message that was relayed inbound via a single MTA that conforms to 1773 this specification and applied three different message authentication 1774 checks: 1776 Authentication-Results: example.com; 1777 auth=pass (cram-md5) smtp.auth=sender@example.net; 1778 spf=pass smtp.mailfrom=example.net 1779 Authentication-Results: example.com; 1780 sender-id=pass header.from=example.net 1781 Received: from dialup-1-2-3-4.example.net (8.11.6/8.11.6) 1782 (dialup-1-2-3-4.example.net [192.0.2.200]) 1783 by mail-router.example.com (8.11.6/8.11.6) 1784 with ESMTP id g1G0r1kA003489; 1785 Fri, Feb 15 2002 17:19:07 -0800 1786 Date: Fri, Feb 15 2002 16:54:30 -0800 1787 To: receiver@example.com 1788 From: sender@example.net 1789 Message-Id: <12345.abc@example.net> 1790 Subject: here's a sample 1792 Hello! Goodbye! 1794 Example 4: Headers reporting results from one MTA 1796 The "Authentication-Results" header field is present, indicating the 1797 delivering MTA conforms to this specification. Once again, the 1798 receiving DNS domain name is used as the authserv-id. Furthermore, 1799 the sender authenticated herself/himself to the MTA via a method 1800 specified in [AUTH], and both SPF and Sender ID checks were done and 1801 passed. The MUA could extract and relay this extra information if 1802 desired. 1804 Two "Authentication-Results" header fields are not required since the 1805 same host did all of the checking. The authenticating agent could 1806 have consolidated all the results into one header field. 1808 This example illustrates a scenario in which a remote user on a 1809 dialup connection (example.net) sends mail to a border MTA 1810 (example.com) using SMTP authentication to prove identity. The 1811 dialup provider has been explicitly authorized to relay mail as 1812 "example.com" resulting in passes by the SPF and SenderID checks. 1814 C.5. Service Provided, Several Authentications Done, Different MTAs 1816 A message that was relayed inbound by two different MTAs that conform 1817 to this specification and applied multiple message authentication 1818 checks: 1820 Authentication-Results: example.com; 1821 sender-id=fail header.from=example.com; 1822 dkim=pass (good signature) header.d=example.com 1823 Received: from mail-router.example.com 1824 (mail-router.example.com [192.0.2.1]) 1825 by auth-checker.example.com (8.11.6/8.11.6) 1826 with ESMTP id i7PK0sH7021929; 1827 Fri, Feb 15 2002 17:19:22 -0800 1828 DKIM-Signature: v=1; a=rsa-sha256; s=gatsby; d=example.com; 1829 t=1188964191; c=simple/simple; h=From:Date:To:Subject: 1830 Message-Id:Authentication-Results; 1831 bh=sEuZGD/pSr7ANysbY3jtdaQ3Xv9xPQtS0m70; 1832 b=EToRSuvUfQVP3Bkz ... rTB0t0gYnBVCM= 1833 Authentication-Results: example.com; 1834 auth=pass (cram-md5) smtp.auth=sender@example.com; 1835 spf=fail smtp.mailfrom=example.com 1836 Received: from dialup-1-2-3-4.example.net 1837 (dialup-1-2-3-4.example.net [192.0.2.200]) 1838 by mail-router.example.com (8.11.6/8.11.6) 1839 with ESMTP id g1G0r1kA003489; 1840 Fri, Feb 15 2002 17:19:07 -0800 1841 From: sender@example.com 1842 Date: Fri, Feb 15 2002 16:54:30 -0800 1843 To: receiver@example.com 1844 Message-Id: <12345.abc@example.com> 1845 Subject: here's a sample 1847 Hello! Goodbye! 1849 Example 5: Headers reporting results from multiple MTAs 1851 The "Authentication-Results" header field is present, indicating 1852 conformance to this specification. Once again, the authserv-id used 1853 is the recipient's DNS domain name. The header field is present 1854 twice because two different MTAs in the chain of delivery did 1855 authentication tests. The first, "mail-router.example.com" reports 1856 that SMTP AUTH and SPF were both used, and the former passed while 1857 the latter failed. In the SMTP AUTH case, additional information is 1858 provided in the comment field, which the MUA can choose to render if 1859 desired. 1861 The second MTA, "auth-checker.example.com", reports that it did a 1862 Sender ID test (which failed) and a DKIM test (which passed). Again, 1863 additional data about one of the tests is provided as a comment, 1864 which the MUA may choose to render. Also noteworthy here is the fact 1865 that there is a DKIM signature added by example.com that assured the 1866 integrity of the lower Authentication-Results field. 1868 Since different hosts did the two sets of authentication checks, the 1869 header fields cannot be consolidated in this example. 1871 This example illustrates more typical transmission of mail into 1872 "example.com" from a user on a dialup connection "example.net". The 1873 user appears to be legitimate as he/she had a valid password allowing 1874 authentication at the border MTA using SMTP AUTH. The SPF and Sender 1875 ID tests failed since "example.com" has not granted "example.net" 1876 authority to relay mail on its behalf. However, the DKIM test passed 1877 because the sending user had a private key matching one of 1878 "example.com"'s published public keys and used it to sign the 1879 message. 1881 C.6. Service Provided, Multi-Tiered Authentication Done 1883 A message that had authentication done at various stages, one of 1884 which was outside the receiving ADMD: 1886 Authentication-Results: example.com; 1887 dkim=pass reason="good signature" 1888 header.i=@mail-router.example.net; 1889 dkim=fail reason="bad signature" 1890 header.i=@newyork.example.com 1891 Received: from mail-router.example.net 1892 (mail-router.example.net [192.0.2.250]) 1893 by chicago.example.com (8.11.6/8.11.6) 1894 for 1895 with ESMTP id i7PK0sH7021929; 1896 Fri, Feb 15 2002 17:19:22 -0800 1897 DKIM-Signature: v=1; a=rsa-sha256; s=furble; 1898 d=mail-router.example.net; t=1188964198; c=relaxed/simple; 1899 h=From:Date:To:Message-Id:Subject:Authentication-Results; 1900 bh=ftA9J6GtX8OpwUECzHnCkRzKw1uk6FNiLfJl5Nmv49E=; 1901 b=oINEO8hgn/gnunsg ... 9n9ODSNFSDij3= 1902 Authentication-Results: example.net; 1903 dkim=pass (good signature) header.i=@newyork.example.com 1904 Received: from smtp.newyork.example.com 1905 (smtp.newyork.example.com [192.0.2.220]) 1906 by mail-router.example.net (8.11.6/8.11.6) 1907 with ESMTP id g1G0r1kA003489; 1908 Fri, Feb 15 2002 17:19:07 -0800 1909 DKIM-Signature: v=1; a=rsa-sha256; s=gatsby; 1910 d=newyork.example.com; 1911 t=1188964191; c=simple/simple; 1912 h=From:Date:To:Message-Id:Subject; 1913 bh=sEu28nfs9fuZGD/pSr7ANysbY3jtdaQ3Xv9xPQtS0m7=; 1914 b=EToRSuvUfQVP3Bkz ... rTB0t0gYnBVCM= 1915 From: sender@newyork.example.com 1916 Date: Fri, Feb 15 2002 16:54:30 -0800 1917 To: meetings@example.net 1918 Message-Id: <12345.abc@newyork.example.com> 1919 Subject: here's a sample 1921 Example 6: Headers reporting results from multiple MTAs in different 1922 ADMDs 1924 In this example we see multi-tiered authentication with an extended 1925 trust boundary. 1927 The message was sent from someone at example.com's New York office 1928 (newyork.example.com) to a mailing list managed at an intermediary. 1930 The message was signed at the origin using DKIM. 1932 The message was sent to a mailing list service provider called 1933 example.net, which is used by example.com. There, 1934 meetings@example.net is expanded to a long list of recipients, one of 1935 whom is at the Chicago office. In this example, we will assume that 1936 the trust boundary for chicago.example.com includes the mailing list 1937 server at example.net. 1939 The mailing list server there first authenticated the message and 1940 affixed an Authentication-Results header field indicating such using 1941 its DNS domain name for the authserv-id. It then altered the message 1942 by affixing some footer text to the body, including some 1943 administrivia such as unsubscription instructions. Finally, the 1944 mailing list server affixes a second DKIM signature and begins 1945 distribution of the message. 1947 The border MTA for chicago.example.com explicitly trusts results from 1948 mail-router.example.net so that header field is not removed. It 1949 performs evaluation of both signatures and determines that the first 1950 (most recent) is a "pass" but, because of the aforementioned 1951 modifications, the second is a "fail". However, the first signature 1952 included the Authentication-Results header added at mail- 1953 router.example.net that validated the second signature. Thus, 1954 indirectly, it can be determined that the authentications claimed by 1955 both signatures are indeed valid. 1957 Note that two styles of presenting meta-data about the result are in 1958 use here. In one case, the "reason=" clause is present which is 1959 intended for easy extraction by parsers; in the other case, the CFWS 1960 production of the ABNF is used to include such data as a header field 1961 comment. The latter can be harder for parsers to extract given the 1962 varied supported syntaxes of mail header fields. 1964 C.7. Comment-Heavy Example 1966 The formal syntax permits comments within the content in a number of 1967 places. For the sake of illustration, this example is also legal: 1969 Authentication-Results: foo.example.net (foobar) 1 (baz); 1970 dkim (Because I like it) / 1 (One yay) = (wait for it) fail 1971 policy (A dot can go here) . (like that) expired 1972 (this surprised me) = (as I wasn't expecting it) 1362471462 1974 Example 7: A very comment-heavy but perfectly legal example 1976 Appendix D. Operational Considerations about Message Authentication 1978 This protocol is predicated on the idea that authentication (and 1979 presumably in the future, reputation) work is typically done by 1980 border MTAs rather than MUAs or intermediate MTAs; the latter merely 1981 make use of the results determined by the former. Certainly this is 1982 not mandatory for participation in electronic mail or message 1983 authentication, but this protocol and its deployment to date are 1984 based on that model. The assumption satisfies several common ADMD 1985 requirements: 1987 1. Service operators prefer to resolve the handling of problem 1988 messages as close to the border of the ADMD as possible. This 1989 enables, for example, rejection of messages at the SMTP level 1990 rather than generating a DSN internally. Thus, doing any of the 1991 authentication or reputation work exclusively at the MUA or 1992 intermediate MTA renders this desire unattainable. 1994 2. Border MTAs are more likely to have direct access to external 1995 sources of authentication or reputation information since modern 1996 MUAs are more likely to be heavily firewalled. Thus, some MUAs 1997 might not even be able to complete the task of performing 1998 authentication or reputation evaluations without complex proxy 1999 configurations or similar burdens. 2001 3. MUAs rely upon the upstream MTAs within their trust boundaries to 2002 make correct (as much as that is possible) evaluations about the 2003 message's envelope, header, and content. Thus, MUAs don't need 2004 to know how to do the work that upstream MTAs do; they only need 2005 the results of that work. 2007 4. Evaluations about the quality of a message, from simple token 2008 matching (e.g., a list of preferred DNS domains) to cryptanalysis 2009 (e.g., public/private key work), are at least a little bit 2010 expensive and thus need to be minimized. To that end, performing 2011 those tests at the border MTA is far preferred to doing that work 2012 at each MUA that handles a message. If an ADMD's environment 2013 adheres to common messaging protocols, a reputation query or an 2014 authentication check performed by a border MTA would return the 2015 same result as the same query performed by an MUA. By contrast, 2016 in an environment where the MUA does the work, a message arriving 2017 for multiple recipients would thus cause authentication or 2018 reputation evaluation to be done more than once for the same 2019 message (i.e., at each MUA) causing needless amplification of 2020 resource use and creating a possible denial-of-service attack 2021 vector. 2023 5. Minimizing change is good. As new authentication and reputation 2024 methods emerge, the list of methods supported by this header 2025 field would presumably be extended. If MUAs simply consume the 2026 contents of this header field rather than actually attempting to 2027 do authentication and/or reputation work, then MUAs only need to 2028 learn to parse this header field once; emergence of new methods 2029 requires only a configuration change at the MUAs and software 2030 changes at the MTAs (which are presumably fewer in number). When 2031 choosing to implement these functions in MTAs vs. MUAs, the 2032 issues of individual flexibility, infrastructure inertia, and 2033 scale of effort must be considered. It is typically easier to 2034 change a single MUA than an MTA because the modification affects 2035 fewer users and can be pursued with less care. However, changing 2036 many MUAs is more effort than changing a smaller number of MTAs. 2038 6. For decisions affecting message delivery and display, assessment 2039 based on authentication and reputation is best performed close to 2040 the time of message transit, as a message makes its journey 2041 toward a user's inbox, not afterwards. DKIM keys and IP address 2042 reputations, etc., can change over time or even become invalid, 2043 and users can take a long time to read a message once delivered. 2044 The value of this work thus degrades, perhaps quickly, once the 2045 delivery process has completed. This seriously diminishes the 2046 value of this work when done other than at MTAs. 2048 Many operational choices are possible within an ADMD, including the 2049 venue for performing authentication and/or reputation assessment. 2050 The current specification does not dictate any of those choices. 2051 Rather, it facilitates those cases in which information produced by 2052 one stage of analysis needs to be transported with the message to the 2053 next stage. 2055 Appendix E. Changes since RFC5451 2057 [Note to IESG: This can be dropped prior to publication unless it's 2058 desirable to carry the changes visibly in this way.] 2060 o Errata #2617 was addressed in RFC6577 and was incorporated here 2062 o Request Internet Standard status 2064 o Change IANA rules to Designated Expert from IETF Review 2066 o Update existing IANA registries from the old RFC to this one 2068 o Add references to ADSP, ATPS, VBR 2069 o Remove all the "X-" stuff, per BCP178 2071 o Adjust language to indicate that this header field was already 2072 defined, and we're just refreshing and revising 2074 o In a few places, RFC2119 language had been used in lowercase 2075 terms; fixed here 2077 o Errata #2818 addressed 2079 o Errata #3195 addressed 2081 o Some minor wordsmithing and removal of odd prose 2083 o ABNF: change "dot-atom" to "Keyword" since "dot-atom" allows "=", 2084 which leads to ambiguous productions 2086 o ABNF: the authserv-id can be a "value", not a "dot-atom" 2088 o ABNF: separate the spec version from the method version; they're 2089 syntactically the same but semantically different; add a section 2090 discussing them 2092 o Call out the SMTP verb exceptions ("mailfrom" and "rcptto"); the 2093 previous RFC didn't do this, leading to interoperability problems 2095 o Rather than deleting suspect header fields, they could also be 2096 renamed to something harmless; there is at least one 2097 implementation of this 2099 o Update IANA method registry to include version numbers 2101 o Rather than repeating what RFC4408[bis] says the SPF results are, 2102 just refer to those documents 2104 o Constrain inclusion of unnecessary properties to avoid confusing 2105 consumers 2107 o Review "should" vs. SHOULD 2109 o Update prose around authserv-id (Section 2.3) 2111 o Merge Sections 2.5 and 2.6 (defined methods and result codes) 2113 Author's Address 2115 Murray S. Kucherawy 2116 270 Upland Drive 2117 San Francisco, CA 94127 2118 US 2120 EMail: superuser@gmail.com