idnits 2.17.1 draft-ietf-dkim-rfc4871bis-09.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 RFC5672, but the abstract doesn't seem to directly say this. It does mention RFC5672 though, so this could be OK. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 1312 has weird spacing: '... email elec...' -- The document date (May 1, 2011) is 4737 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) == Unused Reference: 'ITU-X660-1997' is defined on line 2841, but no explicit reference was found in the text -- Possible downref: Non-RFC (?) normative reference: ref. 'FIPS-180-2-2002' -- Possible downref: Non-RFC (?) normative reference: ref. 'ITU-X660-1997' ** Obsolete normative reference: RFC 3447 (Obsoleted by RFC 8017) ** Downref: Normative reference to an Informational RFC: RFC 5598 -- Obsolete informational reference (is this intentional?): RFC 4409 (Obsoleted by RFC 6409) -- Obsolete informational reference (is this intentional?): RFC 4870 (Obsoleted by RFC 4871) -- Obsolete informational reference (is this intentional?): RFC 4871 (Obsoleted by RFC 6376) -- Obsolete informational reference (is this intentional?): RFC 5226 (Obsoleted by RFC 8126) -- Obsolete informational reference (is this intentional?): RFC 5451 (Obsoleted by RFC 7001) -- Obsolete informational reference (is this intentional?): RFC 5751 (Obsoleted by RFC 8551) Summary: 2 errors (**), 0 flaws (~~), 3 warnings (==), 10 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Crocker, Ed. 3 Internet-Draft Brandenburg InternetWorking 4 Obsoletes: 4871, 5672 T. Hansen, Ed. 5 (if approved) AT&T Laboratories 6 Intended status: Standards Track M. Kucherawy, Ed. 7 Expires: November 2, 2011 Cloudmark 8 May 1, 2011 10 DomainKeys Identified Mail (DKIM) Signatures 11 draft-ietf-dkim-rfc4871bis-09 13 Abstract 15 DomainKeys Identified Mail (DKIM) permits a person, role, or 16 organization that owns the signing domain to claim some 17 responsibility for a message by associating the domain with the 18 message. This can be an author's organization, an operational relay 19 or one of their agents. DKIM separates the question of the identity 20 of the signer of the message from the purported author of the 21 message. Assertion of responsibility is validated through a 22 cryptographic signature and querying the signer's domain directly to 23 retrieve the appropriate public key. Message transit from author to 24 recipient is through relays that typically make no substantive change 25 to the message content and thus preserve the DKIM signature. 27 This memo obsoletes RFC4871 and RFC5672. 29 Status of this Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at http://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on November 2, 2011. 46 Copyright Notice 48 Copyright (c) 2011 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 1.1. DKIM Architecture Documents . . . . . . . . . . . . . . . 6 65 1.2. Signing Identity . . . . . . . . . . . . . . . . . . . . . 6 66 1.3. Scalability . . . . . . . . . . . . . . . . . . . . . . . 6 67 1.4. Simple Key Management . . . . . . . . . . . . . . . . . . 6 68 1.5. Data Integrity . . . . . . . . . . . . . . . . . . . . . . 7 69 2. Terminology and Definitions . . . . . . . . . . . . . . . . . 7 70 2.1. Signers . . . . . . . . . . . . . . . . . . . . . . . . . 7 71 2.2. Verifiers . . . . . . . . . . . . . . . . . . . . . . . . 7 72 2.3. Identity . . . . . . . . . . . . . . . . . . . . . . . . . 7 73 2.4. Identifier . . . . . . . . . . . . . . . . . . . . . . . . 8 74 2.5. Signing Domain Identifier (SDID) . . . . . . . . . . . . . 8 75 2.6. Agent or User Identifier (AUID) . . . . . . . . . . . . . 8 76 2.7. Identity Assessor . . . . . . . . . . . . . . . . . . . . 8 77 2.8. Whitespace . . . . . . . . . . . . . . . . . . . . . . . . 8 78 2.9. Imported ABNF Tokens . . . . . . . . . . . . . . . . . . . 9 79 2.10. Common ABNF Tokens . . . . . . . . . . . . . . . . . . . . 9 80 2.11. DKIM-Quoted-Printable . . . . . . . . . . . . . . . . . . 10 81 3. Protocol Elements . . . . . . . . . . . . . . . . . . . . . . 11 82 3.1. Selectors . . . . . . . . . . . . . . . . . . . . . . . . 11 83 3.2. Tag=Value Lists . . . . . . . . . . . . . . . . . . . . . 13 84 3.3. Signing and Verification Algorithms . . . . . . . . . . . 14 85 3.4. Canonicalization . . . . . . . . . . . . . . . . . . . . . 15 86 3.5. The DKIM-Signature Header Field . . . . . . . . . . . . . 19 87 3.6. Key Management and Representation . . . . . . . . . . . . 28 88 3.7. Computing the Message Hashes . . . . . . . . . . . . . . . 32 89 3.8. Input Requirements . . . . . . . . . . . . . . . . . . . . 35 90 3.9. Output Requirements . . . . . . . . . . . . . . . . . . . 35 91 3.10. Signing by Parent Domains . . . . . . . . . . . . . . . . 35 92 3.11. Relationship between SDID and AUID . . . . . . . . . . . . 36 93 4. Semantics of Multiple Signatures . . . . . . . . . . . . . . . 37 94 4.1. Example Scenarios . . . . . . . . . . . . . . . . . . . . 37 95 4.2. Interpretation . . . . . . . . . . . . . . . . . . . . . . 38 97 5. Signer Actions . . . . . . . . . . . . . . . . . . . . . . . . 39 98 5.1. Determine Whether the Email Should Be Signed and by 99 Whom . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 100 5.2. Select a Private Key and Corresponding Selector 101 Information . . . . . . . . . . . . . . . . . . . . . . . 40 102 5.3. Normalize the Message to Prevent Transport Conversions . . 40 103 5.4. Determine the Header Fields to Sign . . . . . . . . . . . 41 104 5.5. Recommended Signature Content . . . . . . . . . . . . . . 43 105 5.6. Compute the Message Hash and Signature . . . . . . . . . . 45 106 5.7. Insert the DKIM-Signature Header Field . . . . . . . . . . 45 107 6. Verifier Actions . . . . . . . . . . . . . . . . . . . . . . . 45 108 6.1. Extract Signatures from the Message . . . . . . . . . . . 46 109 6.2. Communicate Verification Results . . . . . . . . . . . . . 51 110 6.3. Interpret Results/Apply Local Policy . . . . . . . . . . . 52 111 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 53 112 7.1. DKIM-Signature Tag Specifications . . . . . . . . . . . . 53 113 7.2. DKIM-Signature Query Method Registry . . . . . . . . . . . 54 114 7.3. DKIM-Signature Canonicalization Registry . . . . . . . . . 54 115 7.4. _domainkey DNS TXT Record Tag Specifications . . . . . . . 55 116 7.5. DKIM Key Type Registry . . . . . . . . . . . . . . . . . . 55 117 7.6. DKIM Hash Algorithms Registry . . . . . . . . . . . . . . 56 118 7.7. DKIM Service Types Registry . . . . . . . . . . . . . . . 56 119 7.8. DKIM Selector Flags Registry . . . . . . . . . . . . . . . 56 120 7.9. DKIM-Signature Header Field . . . . . . . . . . . . . . . 57 121 8. Security Considerations . . . . . . . . . . . . . . . . . . . 57 122 8.1. Misuse of Body Length Limits ("l=" Tag) . . . . . . . . . 57 123 8.2. Misappropriated Private Key . . . . . . . . . . . . . . . 57 124 8.3. Key Server Denial-of-Service Attacks . . . . . . . . . . . 58 125 8.4. Attacks Against the DNS . . . . . . . . . . . . . . . . . 58 126 8.5. Replay Attacks . . . . . . . . . . . . . . . . . . . . . . 59 127 8.6. Limits on Revoking Keys . . . . . . . . . . . . . . . . . 59 128 8.7. Intentionally Malformed Key Records . . . . . . . . . . . 60 129 8.8. Intentionally Malformed DKIM-Signature Header Fields . . . 60 130 8.9. Information Leakage . . . . . . . . . . . . . . . . . . . 60 131 8.10. Remote Timing Attacks . . . . . . . . . . . . . . . . . . 60 132 8.11. Reordered Header Fields . . . . . . . . . . . . . . . . . 60 133 8.12. RSA Attacks . . . . . . . . . . . . . . . . . . . . . . . 61 134 8.13. Inappropriate Signing by Parent Domains . . . . . . . . . 61 135 8.14. Attacks Involving Addition of Header Fields . . . . . . . 61 136 8.15. Malformed Inputs . . . . . . . . . . . . . . . . . . . . . 62 137 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 63 138 9.1. Normative References . . . . . . . . . . . . . . . . . . . 63 139 9.2. Informative References . . . . . . . . . . . . . . . . . . 64 140 Appendix A. Example of Use (INFORMATIVE) . . . . . . . . . . . . 65 141 A.1. The User Composes an Email . . . . . . . . . . . . . . . . 65 142 A.2. The Email is Signed . . . . . . . . . . . . . . . . . . . 66 143 A.3. The Email Signature is Verified . . . . . . . . . . . . . 67 144 Appendix B. Usage Examples (INFORMATIVE) . . . . . . . . . . . . 68 145 B.1. Alternate Submission Scenarios . . . . . . . . . . . . . . 68 146 B.2. Alternate Delivery Scenarios . . . . . . . . . . . . . . . 70 147 Appendix C. Creating a Public Key (INFORMATIVE) . . . . . . . . . 72 148 C.1. Compatibility with DomainKeys Key Records . . . . . . . . 73 149 Appendix D. MUA Considerations . . . . . . . . . . . . . . . . . 73 150 Appendix E. Acknowledgements . . . . . . . . . . . . . . . . . . 74 151 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 75 153 1. Introduction 155 DomainKeys Identified Mail (DKIM) permits a person, role, or 156 organization to claim some responsibility for a message by 157 associating a domain name [RFC1034] with the message [RFC5322], which 158 they are authorized to use. This can be an author's organization, an 159 operational relay or one of their agents. Assertion of 160 responsibility is validated through a cryptographic signature and 161 querying the signer's domain directly to retrieve the appropriate 162 public key. Message transit from author to recipient is through 163 relays that typically make no substantive change to the message 164 content and thus preserve the DKIM signature. A message can contain 165 multiple signatures, from the same or different organizations 166 involved with the message. 168 The approach taken by DKIM differs from previous approaches to 169 message signing (e.g., Secure/Multipurpose Internet Mail Extensions 170 (S/MIME) [RFC1847], OpenPGP [RFC4880]) in that: 172 o the message signature is written as a message header field so that 173 neither human recipients nor existing MUA (Mail User Agent) 174 software is confused by signature-related content appearing in the 175 message body; 177 o there is no dependency on public and private key pairs being 178 issued by well-known, trusted certificate authorities; 180 o there is no dependency on the deployment of any new Internet 181 protocols or services for public key distribution or revocation; 183 o signature verification failure does not force rejection of the 184 message; 186 o no attempt is made to include encryption as part of the mechanism; 188 o message archiving is not a design goal. 190 DKIM: 192 o is compatible with the existing email infrastructure and 193 transparent to the fullest extent possible; 195 o requires minimal new infrastructure; 197 o can be implemented independently of clients in order to reduce 198 deployment time; 200 o can be deployed incrementally; 202 o allows delegation of signing to third parties. 204 1.1. DKIM Architecture Documents 206 Readers are advised to be familiar with the material in [RFC4686], 207 [RFC5585] and [RFC5863], which respectively provide the background 208 for the development of DKIM, an overview of the service, and 209 deployment and operations guidance and advice. 211 1.2. Signing Identity 213 DKIM separates the question of the identity of the signer of the 214 message from the purported author of the message. In particular, a 215 signature includes the identity of the signer. Verifiers can use the 216 signing information to decide how they want to process the message. 217 The signing identity is included as part of the signature header 218 field. 220 INFORMATIVE RATIONALE: The signing identity specified by a DKIM 221 signature is not required to match an address in any particular 222 header field because of the broad methods of interpretation by 223 recipient mail systems, including MUAs. 225 1.3. Scalability 227 DKIM is designed to support the extreme scalability requirements that 228 characterize the email identification problem. There are currently 229 over 70 million domains and a much larger number of individual 230 addresses. DKIM seeks to preserve the positive aspects of the 231 current email infrastructure, such as the ability for anyone to 232 communicate with anyone else without introduction. 234 1.4. Simple Key Management 236 DKIM differs from traditional hierarchical public-key systems in that 237 no Certificate Authority infrastructure is required; the verifier 238 requests the public key from a repository in the domain of the 239 claimed signer directly rather than from a third party. 241 The DNS is proposed as the initial mechanism for the public keys. 242 Thus, DKIM currently depends on DNS administration and the security 243 of the DNS system. DKIM is designed to be extensible to other key 244 fetching services as they become available. 246 1.5. Data Integrity 248 A DKIM signature associates the d= name with the computed hash of 249 some or all of the message (see Section 3.7) in order to prevent the 250 re-use of the signature with different messages. Verifying the 251 signature asserts that the hashed content has not changed since it 252 was signed, and asserts nothing else about "protecting" the end-to- 253 end integrity of the message. 255 2. Terminology and Definitions 257 This section defines terms used in the rest of the document. 259 DKIM is designed to operate within the Internet Mail service, as 260 defined in [RFC5598]. Basic email terminology is taken from that 261 specification. 263 Syntax descriptions use Augmented BNF (ABNF) [RFC5234]. 265 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 266 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 267 document are to be interpreted as described in [RFC2119]. 269 2.1. Signers 271 Elements in the mail system that sign messages on behalf of a domain 272 are referred to as signers. These may be MUAs (Mail User Agents), 273 MSAs (Mail Submission Agents), MTAs (Mail Transfer Agents), or other 274 agents such as mailing list exploders. In general, any signer will 275 be involved in the injection of a message into the message system in 276 some way. The key issue is that a message must be signed before it 277 leaves the administrative domain of the signer. 279 2.2. Verifiers 281 Elements in the mail system that verify signatures are referred to as 282 verifiers. These may be MTAs, Mail Delivery Agents (MDAs), or MUAs. 283 In most cases it is expected that verifiers will be close to an end 284 user (reader) of the message or some consuming agent such as a 285 mailing list exploder. 287 2.3. Identity 289 A person, role, or organization. In the context of DKIM, examples 290 include the author, the author's organization, an ISP along the 291 handling path, an independent trust assessment service, and a mailing 292 list operator. 294 2.4. Identifier 296 A label that refers to an identity. 298 2.5. Signing Domain Identifier (SDID) 300 A single domain name that is the mandatory payload output of DKIM and 301 that refers to the identity claiming some responsibility for the 302 message by signing it. It is specified in Section 3.5. 304 2.6. Agent or User Identifier (AUID) 306 A single identifier that refers to the agent or user on behalf of 307 whom the Signing Domain Identifier (SDID) has taken responsibility. 308 The AUID comprises a domain name and an optional . The 309 domain name is the same as that used for the SDID or is a sub-domain 310 of it. For DKIM processing, the domain name portion of the AUID has 311 only basic domain name semantics; any possible owner-specific 312 semantics are outside the scope of DKIM. It is specified in 313 Section 3.5 . 315 2.7. Identity Assessor 317 A module that consumes DKIM's mandatory payload, which is the 318 responsible Signing Domain Identifier (SDID). The module is 319 dedicated to the assessment of the delivered identifier. Other DKIM 320 (and non-DKIM) values can also be delivered to this module as well as 321 to a more general message evaluation filtering engine. However, this 322 additional activity is outside the scope of the DKIM signature 323 specification. 325 2.8. Whitespace 327 There are three forms of whitespace: 329 o WSP represents simple whitespace, i.e., a space or a tab character 330 (formal definition in [RFC5234]). 332 o LWSP is linear whitespace, defined as WSP plus CRLF (formal 333 definition in [RFC5234]). 335 o FWS is folding whitespace. It allows multiple lines separated by 336 CRLF followed by at least one whitespace, to be joined. 338 The formal ABNF for these are (WSP and LWSP are given for information 339 only): 340 WSP = SP / HTAB 341 LWSP = *(WSP / CRLF WSP) 342 FWS = [*WSP CRLF] 1*WSP 344 The definition of FWS is identical to that in [RFC5322] except for 345 the exclusion of obs-FWS. 347 2.9. Imported ABNF Tokens 349 The following tokens are imported from other RFCs as noted. Those 350 RFCs should be considered definitive. 352 The following tokens are imported from [RFC5321]: 354 o "Local-part" (implementation warning: this permits quoted strings) 356 o "sub-domain" 358 The following tokens are imported from [RFC5322]: 360 o "field-name" (name of a header field) 362 o "dot-atom-text" (in the Local-part of an email address) 364 The following tokens are imported from [RFC2045]: 366 o "qp-section" (a single line of quoted-printable-encoded text) 368 o "hex-octet" (a quoted-printable encoded octet) 370 INFORMATIVE NOTE: Be aware that the ABNF in [RFC2045] does not 371 obey the rules of [RFC5234] and must be interpreted accordingly, 372 particularly as regards case folding. 374 Other tokens not defined herein are imported from [RFC5234]. These 375 are intuitive primitives such as SP, HTAB, WSP, ALPHA, DIGIT, CRLF, 376 etc. 378 2.10. Common ABNF Tokens 379 The following ABNF tokens are used elsewhere in this document: 380 hyphenated-word = ALPHA [ *(ALPHA / DIGIT / "-") (ALPHA / DIGIT) ] 381 ALPHADIGITPS = (ALPHA / DIGIT / "+" / "/") 382 base64string = ALPHADIGITPS *([FWS] ALPHADIGITPS) 383 [ [FWS] "=" [ [FWS] "=" ] ] 384 hdr-name = field-name 385 qp-hdr-value = dkim-quoted-printable ; with "|" encoded 387 2.11. DKIM-Quoted-Printable 389 The DKIM-Quoted-Printable encoding syntax resembles that described in 390 Quoted-Printable [RFC2045], Section 6.7: any character MAY be encoded 391 as an "=" followed by two hexadecimal digits from the alphabet 392 "0123456789ABCDEF" (no lowercase characters permitted) representing 393 the hexadecimal-encoded integer value of that character. All control 394 characters (those with values < %x20), 8-bit characters (values > 395 %x7F), and the characters DEL (%x7F), SPACE (%x20), and semicolon 396 (";", %x3B) MUST be encoded. Note that all whitespace, including 397 SPACE, CR, and LF characters, MUST be encoded. After encoding, FWS 398 MAY be added at arbitrary locations in order to avoid excessively 399 long lines; such whitespace is NOT part of the value, and MUST be 400 removed before decoding. 402 ABNF: 404 dkim-quoted-printable = *(FWS / hex-octet / dkim-safe-char) 405 ; hex-octet is from RFC2045 406 dkim-safe-char = %x21-3A / %x3C / %x3E-7E 407 ; '!' - ':', '<', '>' - '~' 408 ; Characters not listed as "mail-safe" in 409 ; [RFC2049] are also not recommended. 411 INFORMATIVE NOTE: DKIM-Quoted-Printable differs from Quoted- 412 Printable as defined in [RFC2045] in several important ways: 414 1. Whitespace in the input text, including CR and LF, must be 415 encoded. [RFC2045] does not require such encoding, and does 416 not permit encoding of CR or LF characters that are part of a 417 CRLF line break. 419 2. Whitespace in the encoded text is ignored. This is to allow 420 tags encoded using DKIM-Quoted-Printable to be wrapped as 421 needed. In particular, [RFC2045] requires that line breaks in 422 the input be represented as physical line breaks; that is not 423 the case here. 425 3. The "soft line break" syntax ("=" as the last non-whitespace 426 character on the line) does not apply. 428 4. DKIM-Quoted-Printable does not require that encoded lines be 429 no more than 76 characters long (although there may be other 430 requirements depending on the context in which the encoded 431 text is being used). 433 3. Protocol Elements 435 Protocol Elements are conceptual parts of the protocol that are not 436 specific to either signers or verifiers. The protocol descriptions 437 for signers and verifiers are described in later sections (Signer 438 Actions (Section 5) and Verifier Actions (Section 6)). NOTE: This 439 section must be read in the context of those sections. 441 3.1. Selectors 443 To support multiple concurrent public keys per signing domain, the 444 key namespace is subdivided using "selectors". For example, 445 selectors might indicate the names of office locations (e.g., 446 "sanfrancisco", "coolumbeach", and "reykjavik"), the signing date 447 (e.g., "january2005", "february2005", etc.), or even an individual 448 user. 450 Selectors are needed to support some important use cases. For 451 example: 453 o Domains that want to delegate signing capability for a specific 454 address for a given duration to a partner, such as an advertising 455 provider or other outsourced function. 457 o Domains that want to allow frequent travelers to send messages 458 locally without the need to connect with a particular MSA. 460 o "Affinity" domains (e.g., college alumni associations) that 461 provide forwarding of incoming mail, but that do not operate a 462 mail submission agent for outgoing mail. 464 Periods are allowed in selectors and are component separators. When 465 keys are retrieved from the DNS, periods in selectors define DNS 466 label boundaries in a manner similar to the conventional use in 467 domain names. Selector components might be used to combine dates 468 with locations, for example, "march2005.reykjavik". In a DNS 469 implementation, this can be used to allow delegation of a portion of 470 the selector namespace. 472 ABNF: 473 selector = sub-domain *( "." sub-domain ) 475 The number of public keys and corresponding selectors for each domain 476 is determined by the domain owner. Many domain owners will be 477 satisfied with just one selector, whereas administratively 478 distributed organizations may choose to manage disparate selectors 479 and key pairs in different regions or on different email servers. 481 Beyond administrative convenience, selectors make it possible to 482 seamlessly replace public keys on a routine basis. If a domain 483 wishes to change from using a public key associated with selector 484 "january2005" to a public key associated with selector 485 "february2005", it merely makes sure that both public keys are 486 advertised in the public-key repository concurrently for the 487 transition period during which email may be in transit prior to 488 verification. At the start of the transition period, the outbound 489 email servers are configured to sign with the "february2005" private 490 key. At the end of the transition period, the "january2005" public 491 key is removed from the public-key repository. 493 INFORMATIVE NOTE: A key may also be revoked as described below. 494 The distinction between revoking and removing a key selector 495 record is subtle. When phasing out keys as described above, a 496 signing domain would probably simply remove the key record after 497 the transition period. However, a signing domain could elect to 498 revoke the key (but maintain the key record) for a further period. 499 There is no defined semantic difference between a revoked key and 500 a removed key. 502 While some domains may wish to make selector values well known, 503 others will want to take care not to allocate selector names in a way 504 that allows harvesting of data by outside parties. For example, if 505 per-user keys are issued, the domain owner will need to make the 506 decision as to whether to associate this selector directly with the 507 name of a registered end-user, or make it some unassociated random 508 value, such as a fingerprint of the public key. 510 INFORMATIVE OPERATIONS NOTE: Reusing a selector with a new key 511 (for example, changing the key associated with a user's name) 512 makes it impossible to tell the difference between a message that 513 didn't verify because the key is no longer valid versus a message 514 that is actually forged. For this reason, signers are ill-advised 515 to reuse selectors for new keys. A better strategy is to assign 516 new keys to new selectors. 518 3.2. Tag=Value Lists 520 DKIM uses a simple "tag=value" syntax in several contexts, including 521 in messages and domain signature records. 523 Values are a series of strings containing either plain text, "base64" 524 text (as defined in [RFC2045], Section 6.8), "qp-section" (ibid, 525 Section 6.7), or "dkim-quoted-printable" (as defined in 526 Section 2.11). The name of the tag will determine the encoding of 527 each value. Unencoded semicolon (";") characters MUST NOT occur in 528 the tag value, since that separates tag-specs. 530 INFORMATIVE IMPLEMENTATION NOTE: Although the "plain text" defined 531 below (as "tag-value") only includes 7-bit characters, an 532 implementation that wished to anticipate future standards would be 533 advised not to preclude the use of UTF8-encoded text in tag=value 534 lists. 536 Formally, the ABNF syntax rules are as follows: 537 tag-list = tag-spec 0*( ";" tag-spec ) [ ";" ] 538 tag-spec = [FWS] tag-name [FWS] "=" [FWS] tag-value [FWS] 539 tag-name = ALPHA 0*ALNUMPUNC 540 tag-value = [ tval 0*( 1*(WSP / FWS) tval ) ] 541 ; WSP and FWS prohibited at beginning and end 542 tval = 1*VALCHAR 543 VALCHAR = %x21-3A / %x3C-7E 544 ; EXCLAMATION to TILDE except SEMICOLON 545 ALNUMPUNC = ALPHA / DIGIT / "_" 547 Note that WSP is allowed anywhere around tags. In particular, any 548 WSP after the "=" and any WSP before the terminating ";" is not part 549 of the value; however, WSP inside the value is significant. 551 Tags MUST be interpreted in a case-sensitive manner. Values MUST be 552 processed as case sensitive unless the specific tag description of 553 semantics specifies case insensitivity. 555 Tags with duplicate names MUST NOT occur within a single tag-list; if 556 a tag name does occur more than once, the entire tag-list is invalid. 558 Whitespace within a value MUST be retained unless explicitly excluded 559 by the specific tag description. 561 Tag=value pairs that represent the default value MAY be included to 562 aid legibility. 564 Unrecognized tags MUST be ignored. 566 Tags that have an empty value are not the same as omitted tags. An 567 omitted tag is treated as having the default value; a tag with an 568 empty value explicitly designates the empty string as the value. 570 3.3. Signing and Verification Algorithms 572 DKIM supports multiple digital signature algorithms. Two algorithms 573 are defined by this specification at this time: rsa-sha1 and rsa- 574 sha256. Signers MUST implement and SHOULD sign using rsa-sha256. 575 Verifiers MUST implement rsa-sha256. 577 INFORMATIVE NOTE: Although rsa-sha256 is strongly encouraged, some 578 senders might prefer to use rsa-sha1 when balancing security 579 strength against performance, complexity, or other needs. 580 However, compliant verifiers might not implement rsa-sha1; they 581 will treat such messages as unsigned. 583 3.3.1. The rsa-sha1 Signing Algorithm 585 The rsa-sha1 Signing Algorithm computes a message hash as described 586 in Section 3.7 below using SHA-1 [FIPS-180-2-2002] as the hash-alg. 587 That hash is then signed by the signer using the RSA algorithm 588 (defined in PKCS#1 version 1.5 [RFC3447]) as the crypt-alg and the 589 signer's private key. The hash MUST NOT be truncated or converted 590 into any form other than the native binary form before being signed. 591 The signing algorithm SHOULD use a public exponent of 65537. 593 3.3.2. The rsa-sha256 Signing Algorithm 595 The rsa-sha256 Signing Algorithm computes a message hash as described 596 in Section 3.7 below using SHA-256 [FIPS-180-2-2002] as the hash-alg. 597 That hash is then signed by the signer using the RSA algorithm 598 (defined in PKCS#1 version 1.5 [RFC3447]) as the crypt-alg and the 599 signer's private key. The hash MUST NOT be truncated or converted 600 into any form other than the native binary form before being signed. 602 3.3.3. Key Sizes 604 Selecting appropriate key sizes is a trade-off between cost, 605 performance, and risk. Since short RSA keys more easily succumb to 606 off-line attacks, signers MUST use RSA keys of at least 1024 bits for 607 long-lived keys. Verifiers MUST be able to validate signatures with 608 keys ranging from 512 bits to 2048 bits, and they MAY be able to 609 validate signatures with larger keys. Verifier policies may use the 610 length of the signing key as one metric for determining whether a 611 signature is acceptable. 613 Factors that should influence the key size choice include the 614 following: 616 o The practical constraint that large (e.g., 4096 bit) keys may not 617 fit within a 512-byte DNS UDP response packet 619 o The security constraint that keys smaller than 1024 bits are 620 subject to off-line attacks 622 o Larger keys impose higher CPU costs to verify and sign email 624 o Keys can be replaced on a regular basis, thus their lifetime can 625 be relatively short 627 o The security goals of this specification are modest compared to 628 typical goals of other systems that employ digital signatures 630 See [RFC3766] for further discussion on selecting key sizes. 632 3.3.4. Other Algorithms 634 Other algorithms MAY be defined in the future. Verifiers MUST ignore 635 any signatures using algorithms that they do not implement. 637 3.4. Canonicalization 639 Some mail systems modify email in transit, potentially invalidating a 640 signature. For most signers, mild modification of email is 641 immaterial to validation of the DKIM domain name's use. For such 642 signers, a canonicalization algorithm that survives modest in-transit 643 modification is preferred. 645 Other signers demand that any modification of the email, however 646 minor, result in a signature verification failure. These signers 647 prefer a canonicalization algorithm that does not tolerate in-transit 648 modification of the signed email. 650 Some signers may be willing to accept modifications to header fields 651 that are within the bounds of email standards such as [RFC5322], but 652 are unwilling to accept any modification to the body of messages. 654 To satisfy all requirements, two canonicalization algorithms are 655 defined for each of the header and the body: a "simple" algorithm 656 that tolerates almost no modification and a "relaxed" algorithm that 657 tolerates common modifications such as whitespace replacement and 658 header field line rewrapping. A signer MAY specify either algorithm 659 for header or body when signing an email. If no canonicalization 660 algorithm is specified by the signer, the "simple" algorithm defaults 661 for both header and body. Verifiers MUST implement both 662 canonicalization algorithms. Note that the header and body may use 663 different canonicalization algorithms. Further canonicalization 664 algorithms MAY be defined in the future; verifiers MUST ignore any 665 signatures that use unrecognized canonicalization algorithms. 667 Canonicalization simply prepares the email for presentation to the 668 signing or verification algorithm. It MUST NOT change the 669 transmitted data in any way. Canonicalization of header fields and 670 body are described below. 672 NOTE: This section assumes that the message is already in "network 673 normal" format (text is ASCII encoded, lines are separated with CRLF 674 characters, etc.). See also Section 5.3 for information about 675 normalizing the message. 677 3.4.1. The "simple" Header Canonicalization Algorithm 679 The "simple" header canonicalization algorithm does not change header 680 fields in any way. Header fields MUST be presented to the signing or 681 verification algorithm exactly as they are in the message being 682 signed or verified. In particular, header field names MUST NOT be 683 case folded and whitespace MUST NOT be changed. 685 3.4.2. The "relaxed" Header Canonicalization Algorithm 687 The "relaxed" header canonicalization algorithm MUST apply the 688 following steps in order: 690 o Convert all header field names (not the header field values) to 691 lowercase. For example, convert "SUBJect: AbC" to "subject: AbC". 693 o Unfold all header field continuation lines as described in 694 [RFC5322]; in particular, lines with terminators embedded in 695 continued header field values (that is, CRLF sequences followed by 696 WSP) MUST be interpreted without the CRLF. Implementations MUST 697 NOT remove the CRLF at the end of the header field value. 699 o Convert all sequences of one or more WSP characters to a single SP 700 character. WSP characters here include those before and after a 701 line folding boundary. 703 o Delete all WSP characters at the end of each unfolded header field 704 value. 706 o Delete any WSP characters remaining before and after the colon 707 separating the header field name from the header field value. The 708 colon separator MUST be retained. 710 3.4.3. The "simple" Body Canonicalization Algorithm 712 The "simple" body canonicalization algorithm ignores all empty lines 713 at the end of the message body. An empty line is a line of zero 714 length after removal of the line terminator. If there is no body or 715 no trailing CRLF on the message body, a CRLF is added. It makes no 716 other changes to the message body. In more formal terms, the 717 "simple" body canonicalization algorithm converts "0*CRLF" at the end 718 of the body to a single "CRLF". 720 Note that a completely empty or missing body is canonicalized as a 721 single "CRLF"; that is, the canonicalized length will be 2 octets. 723 The sha1 value (in base64) for an empty body (canonicalized to a 724 "CRLF") is: 725 uoq1oCgLlTqpdDX/iUbLy7J1Wic= 726 The sha256 value is: 727 frcCV1k9oG9oKj3dpUqdJg1PxRT2RSN/XKdLCPjaYaY= 729 3.4.4. The "relaxed" Body Canonicalization Algorithm 731 The "relaxed" body canonicalization algorithm MUST apply the 732 following steps (a) and (b) in order: 734 a. Reduce whitespace: 736 * Ignore all whitespace at the end of lines. Implementations 737 MUST NOT remove the CRLF at the end of the line. 739 * Reduce all sequences of WSP within a line to a single SP 740 character. 742 b. Ignore all empty lines at the end of the message body. "Empty 743 line" is defined in Section 3.4.3. If the body is non-empty, but 744 does not end with a CRLF, a CRLF is added. (For email, this is 745 only possible when using extensions to SMTP or non-SMTP transport 746 mechanisms.) 748 The sha1 value (in base64) for an empty body (canonicalized to a null 749 input) is: 750 2jmj7l5rSw0yVb/vlWAYkK/YBwk= 751 The sha256 value is: 752 47DEQpj8HBSa+/TImW+5JCeuQeRkm5NMpJWZG3hSuFU= 754 INFORMATIVE NOTE: It should be noted that the relaxed body 755 canonicalization algorithm may enable certain types of extremely 756 crude "ASCII Art" attacks where a message may be conveyed by 757 adjusting the spacing between words. If this is a concern, the 758 "simple" body canonicalization algorithm should be used instead. 760 3.4.5. Body Length Limits 762 A body length count MAY be specified to limit the signature 763 calculation to an initial prefix of the body text, measured in 764 octets. If the body length count is not specified, the entire 765 message body is signed. 767 INFORMATIVE RATIONALE: This capability is provided because it is 768 very common for mailing lists to add trailers to messages (e.g., 769 instructions how to get off the list). Until those messages are 770 also signed, the body length count is a useful tool for the 771 verifier since it may as a matter of policy accept messages having 772 valid signatures with extraneous data. 774 INFORMATIVE IMPLEMENTATION NOTE: Using body length limits enables 775 an attack in which an attacker modifies a message to include 776 content that solely benefits the attacker. It is possible for the 777 appended content to completely replace the original content in the 778 end recipient's eyes, such as via alterations to the MIME 779 structure or exploiting lax HTML parsing in the MUA, and to defeat 780 duplicate message detection algorithms. To avoid this attack, 781 signers should be wary of using this tag, and verifiers might wish 782 to ignore the tag, perhaps based on other criteria. 784 The body length count allows the signer of a message to permit data 785 to be appended to the end of the body of a signed message. The body 786 length count MUST be calculated following the canonicalization 787 algorithm; for example, any whitespace ignored by a canonicalization 788 algorithm is not included as part of the body length count. 790 A body length count of zero means that the body is completely 791 unsigned. 793 Signers wishing to ensure that no modification of any sort can occur 794 should specify the "simple" canonicalization algorithm for both 795 header and body and omit the body length count. 797 3.4.6. Canonicalization Examples (INFORMATIVE) 799 In the following examples, actual whitespace is used only for 800 clarity. The actual input and output text is designated using 801 bracketed descriptors: "" for a space character, "" for a 802 tab character, and "" for a carriage-return/line-feed sequence. 803 For example, "X Y" and "XY" represent the same three 804 characters. 806 Example 1: A message reading: 807 A: X 808 B : Y 809 Z 810 811 C 812 D E 813 814 816 when canonicalized using relaxed canonicalization for both header and 817 body results in a header reading: 818 a:X 819 b:Y Z 821 and a body reading: 822 C 823 D E 825 Example 2: The same message canonicalized using simple 826 canonicalization for both header and body results in a header 827 reading: 828 A: X 829 B : Y 830 Z 832 and a body reading: 833 C 834 D E 836 Example 3: When processed using relaxed header canonicalization and 837 simple body canonicalization, the canonicalized version has a header 838 of: 839 a:X 840 b:Y Z 842 and a body reading: 843 C 844 D E 846 3.5. The DKIM-Signature Header Field 848 The signature of the email is stored in the DKIM-Signature header 849 field. This header field contains all of the signature and key- 850 fetching data. The DKIM-Signature value is a tag-list as described 851 in Section 3.2. 853 The DKIM-Signature header field SHOULD be treated as though it were a 854 trace header field as defined in Section 3.6 of [RFC5322], and hence 855 SHOULD NOT be reordered and SHOULD be prepended to the message. 857 The DKIM-Signature header field being created or verified is always 858 included in the signature calculation, after the rest of the header 859 fields being signed; however, when calculating or verifying the 860 signature, the value of the "b=" tag (signature value) of that DKIM- 861 Signature header field MUST be treated as though it were an empty 862 string. Unknown tags in the DKIM-Signature header field MUST be 863 included in the signature calculation but MUST be otherwise ignored 864 by verifiers. Other DKIM-Signature header fields that are included 865 in the signature should be treated as normal header fields; in 866 particular, the "b=" tag is not treated specially. 868 The encodings for each field type are listed below. Tags described 869 as qp-section are encoded as described in Section 6.7 of MIME Part 870 One [RFC2045], with the additional conversion of semicolon characters 871 to "=3B"; intuitively, this is one line of quoted-printable encoded 872 text. The dkim-quoted-printable syntax is defined in Section 2.11. 874 Tags on the DKIM-Signature header field along with their type and 875 requirement status are shown below. Unrecognized tags MUST be 876 ignored. 878 v= Version (MUST be included). This tag defines the version of this 879 specification that applies to the signature record. It MUST have 880 the value "1". Note that verifiers must do a string comparison on 881 this value; for example, "1" is not the same as "1.0". 883 ABNF: 884 sig-v-tag = %x76 [FWS] "=" [FWS] "1" 886 INFORMATIVE NOTE: DKIM-Signature version numbers are expected 887 to increase arithmetically as new versions of this 888 specification are released. 890 a= The algorithm used to generate the signature (plain-text; 891 REQUIRED). Verifiers MUST support "rsa-sha1" and "rsa-sha256"; 892 signers SHOULD sign using "rsa-sha256". 894 ABNF: 896 sig-a-tag = %x61 [FWS] "=" [FWS] sig-a-tag-alg 897 sig-a-tag-alg = sig-a-tag-k "-" sig-a-tag-h 898 sig-a-tag-k = "rsa" / x-sig-a-tag-k 899 sig-a-tag-h = "sha1" / "sha256" / x-sig-a-tag-h 900 x-sig-a-tag-k = ALPHA *(ALPHA / DIGIT) 901 ; for later extension 902 x-sig-a-tag-h = ALPHA *(ALPHA / DIGIT) 903 ; for later extension 905 b= The signature data (base64; REQUIRED). Whitespace is ignored in 906 this value and MUST be ignored when reassembling the original 907 signature. In particular, the signing process can safely insert 908 FWS in this value in arbitrary places to conform to line-length 909 limits. See Signer Actions (Section 5) for how the signature is 910 computed. 912 ABNF: 913 sig-b-tag = %x62 [FWS] "=" [FWS] sig-b-tag-data 914 sig-b-tag-data = base64string 916 bh= The hash of the canonicalized body part of the message as 917 limited by the "l=" tag (base64; REQUIRED). Whitespace is ignored 918 in this value and MUST be ignored when reassembling the original 919 signature. In particular, the signing process can safely insert 920 FWS in this value in arbitrary places to conform to line-length 921 limits. See Section 3.7 for how the body hash is computed. 923 ABNF: 924 sig-bh-tag = %x62 %x68 [FWS] "=" [FWS] sig-bh-tag-data 925 sig-bh-tag-data = base64string 927 c= Message canonicalization (plain-text; OPTIONAL, default is 928 "simple/simple"). This tag informs the verifier of the type of 929 canonicalization used to prepare the message for signing. It 930 consists of two names separated by a "slash" (%d47) character, 931 corresponding to the header and body canonicalization algorithms 932 respectively. These algorithms are described in Section 3.4. If 933 only one algorithm is named, that algorithm is used for the header 934 and "simple" is used for the body. For example, "c=relaxed" is 935 treated the same as "c=relaxed/simple". 937 ABNF: 938 sig-c-tag = %x63 [FWS] "=" [FWS] sig-c-tag-alg 939 ["/" sig-c-tag-alg] 940 sig-c-tag-alg = "simple" / "relaxed" / x-sig-c-tag-alg 941 x-sig-c-tag-alg = hyphenated-word ; for later extension 943 d= The SDID claiming responsibility for an introduction of a message 944 into the mail stream (plain-text; REQUIRED). Hence, the SDID 945 value is used to form the query for the public key. The SDID MUST 946 correspond to a valid DNS name under which the DKIM key record is 947 published. The conventions and semantics used by a signer to 948 create and use a specific SDID are outside the scope of the DKIM 949 Signing specification, as is any use of those conventions and 950 semantics. When presented with a signature that does not meet 951 these requirements, verifiers MUST consider the signature invalid. 953 Internationalized domain names MUST be encoded as A-Labels, as 954 described in Section 2.3 of [RFC5890]. 956 ABNF: 958 sig-d-tag = %x64 [FWS] "=" [FWS] domain-name 959 domain-name = sub-domain 1*("." sub-domain) 960 ; from RFC5321 Domain, excluding address-literal 962 h= Signed header fields (plain-text, but see description; REQUIRED). 963 A colon-separated list of header field names that identify the 964 header fields presented to the signing algorithm. The field MUST 965 contain the complete list of header fields in the order presented 966 to the signing algorithm. The field MAY contain names of header 967 fields that do not exist when signed; nonexistent header fields do 968 not contribute to the signature computation (that is, they are 969 treated as the null input, including the header field name, the 970 separating colon, the header field value, and any CRLF 971 terminator). The field MUST NOT include the DKIM-Signature header 972 field that is being created or verified, but may include others. 973 Folding whitespace (FWS) MAY be included on either side of the 974 colon separator. Header field names MUST be compared against 975 actual header field names in a case-insensitive manner. This list 976 MUST NOT be empty. See Section 5.4 for a discussion of choosing 977 header fields to sign. 979 ABNF: 980 sig-h-tag = %x68 [FWS] "=" [FWS] hdr-name 981 0*( [FWS] ":" [FWS] hdr-name ) 983 INFORMATIVE EXPLANATION: By "signing" header fields that do not 984 actually exist, a signer can prevent insertion of those header 985 fields before verification. However, since a signer cannot 986 possibly know what header fields might be created in the 987 future, and that some MUAs might present header fields that are 988 embedded inside a message (e.g., as a message/rfc822 content 989 type), the security of this solution is not total. 991 INFORMATIVE EXPLANATION: The exclusion of the header field name 992 and colon as well as the header field value for non-existent 993 header fields prevents an attacker from inserting an actual 994 header field with a null value. 996 i= The Agent or User Identifier (AUID) on behalf of which the SDID is 997 taking responsibility (dkim-quoted-printable; OPTIONAL, default is 998 an empty Local-part followed by an "@" followed by the domain from 999 the "d=" tag). 1001 The syntax is a standard email address where the Local-part MAY be 1002 omitted. The domain part of the address MUST be the same as, or a 1003 subdomain of, the value of the "d=" tag. 1005 Internationalized domain names MUST be encoded as A-Labels, as 1006 described in Section 2.3 of [RFC5890]. 1008 ABNF: 1010 sig-i-tag = %x69 [FWS] "=" [FWS] [ Local-part ] 1011 "@" domain-name 1013 The AUID is specified as having the same syntax as an email 1014 address, but is not required to have the same semantics. Notably, 1015 the domain name is not required to be registered in the DNS -- so 1016 it might not resolve in a query -- and the Local-part MAY be drawn 1017 from a namespace unrelated to any mailbox. The details of the 1018 structure and semantics for the namespace are determined by the 1019 Signer. Any knowledge or use of those details by verifiers or 1020 assessors is outside the scope of the DKIM Signing specification. 1021 The Signer MAY choose to use the same namespace for its AUIDs as 1022 its users' email addresses or MAY choose other means of 1023 representing its users. However, the signer SHOULD use the same 1024 AUID for each message intended to be evaluated as being within the 1025 same sphere of responsibility, if it wishes to offer receivers the 1026 option of using the AUID as a stable identifier that is finer 1027 grained than the SDID. 1029 INFORMATIVE NOTE: The Local-part of the "i=" tag is optional 1030 because in some cases a signer may not be able to establish a 1031 verified individual identity. In such cases, the signer might 1032 wish to assert that although it is willing to go as far as 1033 signing for the domain, it is unable or unwilling to commit to 1034 an individual user name within their domain. It can do so by 1035 including the domain part but not the Local-part of the 1036 identity. 1038 INFORMATIVE DISCUSSION: This specification does not require the 1039 value of the "i=" tag to match the identity in any message 1040 header fields. This is considered to be a verifier policy 1041 issue. Constraints between the value of the "i=" tag and other 1042 identities in other header fields seek to apply basic 1043 authentication into the semantics of trust associated with a 1044 role such as content author. Trust is a broad and complex 1045 topic and trust mechanisms are subject to highly creative 1046 attacks. The real-world efficacy of any but the most basic 1047 bindings between the "i=" value and other identities is not 1048 well established, nor is its vulnerability to subversion by an 1049 attacker. Hence reliance on the use of these options should be 1050 strictly limited. In particular, it is not at all clear to 1051 what extent a typical end-user recipient can rely on any 1052 assurances that might be made by successful use of the "i=" 1053 options. 1055 l= Body length count (plain-text unsigned decimal integer; OPTIONAL, 1056 default is entire body). This tag informs the verifier of the 1057 number of octets in the body of the email after canonicalization 1058 included in the cryptographic hash, starting from 0 immediately 1059 following the CRLF preceding the body. This value MUST NOT be 1060 larger than the actual number of octets in the canonicalized 1061 message body. 1063 INFORMATIVE IMPLEMENTATION WARNING: Use of the "l=" tag might 1064 allow display of fraudulent content without appropriate warning 1065 to end users. The "l=" tag is intended for increasing 1066 signature robustness when sending to mailing lists that both 1067 modify their content and do not sign their messages. However, 1068 using the "l=" tag enables attacks in which an intermediary 1069 with malicious intent modifies a message to include content 1070 that solely benefits the attacker. It is possible for the 1071 appended content to completely replace the original content in 1072 the end recipient's eyes and to defeat duplicate message 1073 detection algorithms. Examples are described in Security 1074 Considerations Section 8. To avoid this attack, signers should 1075 be extremely wary of using this tag, and verifiers might wish 1076 to ignore the tag. 1078 INFORMATIVE NOTE: The value of the "l=" tag is constrained to 1079 76 decimal digits. This constraint is not intended to predict 1080 the size of future messages or to require implementations to 1081 use an integer representation large enough to represent the 1082 maximum possible value, but is intended to remind the 1083 implementer to check the length of this and all other tags 1084 during verification and to test for integer overflow when 1085 decoding the value. Implementers may need to limit the actual 1086 value expressed to a value smaller than 10^76, e.g., to allow a 1087 message to fit within the available storage space. 1089 ABNF: 1090 sig-l-tag = %x6c [FWS] "=" [FWS] 1091 1*76DIGIT 1093 q= A colon-separated list of query methods used to retrieve the 1094 public key (plain-text; OPTIONAL, default is "dns/txt"). Each 1095 query method is of the form "type[/options]", where the syntax and 1096 semantics of the options depend on the type and specified options. 1097 If there are multiple query mechanisms listed, the choice of query 1098 mechanism MUST NOT change the interpretation of the signature. 1099 Implementations MUST use the recognized query mechanisms in the 1100 order presented. Unrecognized query mechanisms MUST be ignored. 1102 Currently, the only valid value is "dns/txt", which defines the 1103 DNS TXT record lookup algorithm described elsewhere in this 1104 document. The only option defined for the "dns" query type is 1105 "txt", which MUST be included. Verifiers and signers MUST support 1106 "dns/txt". 1108 ABNF: 1109 sig-q-tag = %x71 [FWS] "=" [FWS] sig-q-tag-method 1110 *([FWS] ":" [FWS] sig-q-tag-method) 1111 sig-q-tag-method = "dns/txt" / x-sig-q-tag-type 1112 ["/" x-sig-q-tag-args] 1113 x-sig-q-tag-type = hyphenated-word ; for future extension 1114 x-sig-q-tag-args = qp-hdr-value 1116 s= The selector subdividing the namespace for the "d=" (domain) tag 1117 (plain-text; REQUIRED). 1119 Internationalized selector names MUST be encoded as A-Labels, as 1120 described in Section 2.3 of [RFC5890]. 1122 ABNF: 1123 sig-s-tag = %x73 [FWS] "=" [FWS] selector 1125 t= Signature Timestamp (plain-text unsigned decimal integer; 1126 RECOMMENDED, default is an unknown creation time). The time that 1127 this signature was created. The format is the number of seconds 1128 since 00:00:00 on January 1, 1970 in the UTC time zone. The value 1129 is expressed as an unsigned integer in decimal ASCII. This value 1130 is not constrained to fit into a 31- or 32-bit integer. 1131 Implementations SHOULD be prepared to handle values up to at least 1132 10^12 (until approximately AD 200,000; this fits into 40 bits). 1133 To avoid denial-of-service attacks, implementations MAY consider 1134 any value longer than 12 digits to be infinite. Leap seconds are 1135 not counted. Implementations MAY ignore signatures that have a 1136 timestamp in the future. 1138 ABNF: 1139 sig-t-tag = %x74 [FWS] "=" [FWS] 1*12DIGIT 1141 x= Signature Expiration (plain-text unsigned decimal integer; 1142 RECOMMENDED, default is no expiration). The format is the same as 1143 in the "t=" tag, represented as an absolute date, not as a time 1144 delta from the signing timestamp. The value is expressed as an 1145 unsigned integer in decimal ASCII, with the same constraints on 1146 the value in the "t=" tag. Signatures MAY be considered invalid 1147 if the verification time at the verifier is past the expiration 1148 date. The verification time should be the time that the message 1149 was first received at the administrative domain of the verifier if 1150 that time is reliably available; otherwise the current time should 1151 be used. The value of the "x=" tag MUST be greater than the value 1152 of the "t=" tag if both are present. 1154 INFORMATIVE NOTE: The "x=" tag is not intended as an anti- 1155 replay defense. 1157 INFORMATIVE NOTE: Due to clock drift, the receiver's notion of 1158 when to consider the signature expired may not match exactly 1159 when the sender is expecting. Receivers MAY add a 'fudge 1160 factor' to allow for such possible drift. 1162 ABNF: 1163 sig-x-tag = %x78 [FWS] "=" [FWS] 1164 1*12DIGIT 1166 z= Copied header fields (dkim-quoted-printable, but see description; 1167 OPTIONAL, default is null). A vertical-bar-separated list of 1168 selected header fields present when the message was signed, 1169 including both the field name and value. It is not required to 1170 include all header fields present at the time of signing. This 1171 field need not contain the same header fields listed in the "h=" 1172 tag. The header field text itself must encode the vertical bar 1173 ("|", %x7C) character (i.e., vertical bars in the "z=" text are 1174 meta-characters, and any actual vertical bar characters in a 1175 copied header field must be encoded). Note that all whitespace 1176 must be encoded, including whitespace between the colon and the 1177 header field value. After encoding, FWS MAY be added at arbitrary 1178 locations in order to avoid excessively long lines; such 1179 whitespace is NOT part of the value of the header field, and MUST 1180 be removed before decoding. 1182 The header fields referenced by the "h=" tag refer to the fields 1183 in the [RFC5322] header of the message, not to any copied fields 1184 in the "z=" tag. Copied header field values are for diagnostic 1185 use. 1187 ABNF: 1188 sig-z-tag = %x7A [FWS] "=" [FWS] sig-z-tag-copy 1189 *( "|" [FWS] sig-z-tag-copy ) 1190 sig-z-tag-copy = hdr-name [FWS] ":" qp-hdr-value 1192 INFORMATIVE EXAMPLE of a signature header field spread across 1193 multiple continuation lines: 1194 DKIM-Signature: v=1; a=rsa-sha256; d=example.net; s=brisbane; 1195 c=simple; q=dns/txt; i=@eng.example.net; 1196 t=1117574938; x=1118006938; 1197 h=from:to:subject:date; 1198 z=From:foo@eng.example.net|To:joe@example.com| 1199 Subject:demo=20run|Date:July=205,=202005=203:44:08=20PM=20-0700; 1200 bh=MTIzNDU2Nzg5MDEyMzQ1Njc4OTAxMjM0NTY3ODkwMTI=; 1201 b=dzdVyOfAKCdLXdJOc9G2q8LoXSlEniSbav+yuU4zGeeruD00lszZVoG4ZHRNiYzR 1203 3.6. Key Management and Representation 1205 Signature applications require some level of assurance that the 1206 verification public key is associated with the claimed signer. Many 1207 applications achieve this by using public key certificates issued by 1208 a trusted third party. However, DKIM can achieve a sufficient level 1209 of security, with significantly enhanced scalability, by simply 1210 having the verifier query the purported signer's DNS entry (or some 1211 security-equivalent) in order to retrieve the public key. 1213 DKIM keys can potentially be stored in multiple types of key servers 1214 and in multiple formats. The storage and format of keys are 1215 irrelevant to the remainder of the DKIM algorithm. 1217 Parameters to the key lookup algorithm are the type of the lookup 1218 (the "q=" tag), the domain of the signer (the "d=" tag of the DKIM- 1219 Signature header field), and the selector (the "s=" tag). 1220 public_key = dkim_find_key(q_val, d_val, s_val) 1221 This document defines a single binding, using DNS TXT records to 1222 distribute the keys. Other bindings may be defined in the future. 1224 3.6.1. Textual Representation 1226 It is expected that many key servers will choose to present the keys 1227 in an otherwise unstructured text format (for example, an XML form 1228 would not be considered to be unstructured text for this purpose). 1229 The following definition MUST be used for any DKIM key represented in 1230 an otherwise unstructured textual form. 1232 The overall syntax is a tag-list as described in Section 3.2. The 1233 current valid tags are described below. Other tags MAY be present 1234 and MUST be ignored by any implementation that does not understand 1235 them. 1237 v= Version of the DKIM key record (plain-text; RECOMMENDED, default 1238 is "DKIM1"). If specified, this tag MUST be set to "DKIM1" 1239 (without the quotes). This tag MUST be the first tag in the 1240 record. Records beginning with a "v=" tag with any other value 1241 MUST be discarded. Note that verifiers must do a string 1242 comparison on this value; for example, "DKIM1" is not the same as 1243 "DKIM1.0". 1245 ABNF: 1246 key-v-tag = %x76 [FWS] "=" [FWS] %x44 %x4B %x49 %x4D %x31 1248 h= Acceptable hash algorithms (plain-text; OPTIONAL, defaults to 1249 allowing all algorithms). A colon-separated list of hash 1250 algorithms that might be used. Unrecognized algorithms MUST be 1251 ignored. Refer to Section 3.3 for a discussion of the hash 1252 algorithms implemented by Signers and Verifiers. The set of 1253 algorithms listed in this tag in each record is an operational 1254 choice made by the Signer. 1256 ABNF: 1257 key-h-tag = %x68 [FWS] "=" [FWS] key-h-tag-alg 1258 0*( [FWS] ":" [FWS] key-h-tag-alg ) 1259 key-h-tag-alg = "sha1" / "sha256" / x-key-h-tag-alg 1260 x-key-h-tag-alg = hyphenated-word ; for future extension 1261 k= Key type (plain-text; OPTIONAL, default is "rsa"). Signers and 1262 verifiers MUST support the "rsa" key type. The "rsa" key type 1263 indicates that an ASN.1 DER-encoded [ITU-X660-1997] RSAPublicKey 1264 [RFC3447] (see Sections Section 3.1 and A.1.1) is being used in 1265 the "p=" tag. (Note: the "p=" tag further encodes the value using 1266 the base64 algorithm.) Unrecognized key types MUST be ignored. 1268 ABNF: 1269 key-k-tag = %x76 [FWS] "=" [FWS] key-k-tag-type 1270 key-k-tag-type = "rsa" / x-key-k-tag-type 1271 x-key-k-tag-type = hyphenated-word ; for future extension 1273 n= Notes that might be of interest to a human (qp-section; OPTIONAL, 1274 default is empty). No interpretation is made by any program. 1275 This tag should be used sparingly in any key server mechanism that 1276 has space limitations (notably DNS). This is intended for use by 1277 administrators, not end users. 1279 ABNF: 1280 key-n-tag = %x6e [FWS] "=" [FWS] qp-section 1282 p= Public-key data (base64; REQUIRED). An empty value means that 1283 this public key has been revoked. The syntax and semantics of 1284 this tag value before being encoded in base64 are defined by the 1285 "k=" tag. 1287 INFORMATIVE RATIONALE: If a private key has been compromised or 1288 otherwise disabled (e.g., an outsourcing contract has been 1289 terminated), a signer might want to explicitly state that it 1290 knows about the selector, but all messages using that selector 1291 should fail verification. Verifiers should ignore any DKIM- 1292 Signature header fields with a selector referencing a revoked 1293 key. 1295 ABNF: 1296 key-p-tag = %x70 [FWS] "=" [ [FWS] base64string] 1298 INFORMATIVE NOTE: A base64string is permitted to include white 1299 space (FWS) at arbitrary places; however, any CRLFs must be 1300 followed by at least one WSP character. Implementors and 1301 administrators are cautioned to ensure that selector TXT 1302 records conform to this specification. 1304 s= Service Type (plain-text; OPTIONAL; default is "*"). A colon- 1305 separated list of service types to which this record applies. 1306 Verifiers for a given service type MUST ignore this record if the 1307 appropriate type is not listed. Unrecognized service types MUST 1308 be ignored. Currently defined service types are as follows: 1310 * matches all service types 1312 email electronic mail (not necessarily limited to SMTP) 1314 This tag is intended to constrain the use of keys for other 1315 purposes, should use of DKIM be defined by other services in the 1316 future. 1318 ABNF: 1319 key-s-tag = %x73 [FWS] "=" [FWS] key-s-tag-type 1320 0*( [FWS] ":" [FWS] key-s-tag-type ) 1321 key-s-tag-type = "email" / "*" / x-key-s-tag-type 1322 x-key-s-tag-type = hyphenated-word ; for future extension 1324 t= Flags, represented as a colon-separated list of names (plain- 1325 text; OPTIONAL, default is no flags set). Unrecognized flags MUST 1326 be ignored. The defined flags are as follows: 1328 y This domain is testing DKIM. Verifiers MUST NOT treat messages 1329 from signers in testing mode differently from unsigned email, even 1330 should the signature fail to verify. Verifiers MAY wish to track 1331 testing mode results to assist the signer. 1333 s Any DKIM-Signature header fields using the "i=" tag MUST have the 1334 same domain value on the right-hand side of the "@" in the "i=" 1335 tag and the value of the "d=" tag. That is, the "i=" domain MUST 1336 NOT be a subdomain of "d=". Use of this flag is RECOMMENDED 1337 unless subdomaining is required. 1339 ABNF: 1340 key-t-tag = %x74 [FWS] "=" [FWS] key-t-tag-flag 1341 0*( [FWS] ":" [FWS] key-t-tag-flag ) 1342 key-t-tag-flag = "y" / "s" / x-key-t-tag-flag 1343 x-key-t-tag-flag = hyphenated-word ; for future extension 1345 Unrecognized flags MUST be ignored. 1347 3.6.2. DNS Binding 1349 A binding using DNS TXT records as a key service is hereby defined. 1350 All implementations MUST support this binding. 1352 3.6.2.1. Namespace 1354 All DKIM keys are stored in a subdomain named "_domainkey". Given a 1355 DKIM-Signature field with a "d=" tag of "example.com" and an "s=" tag 1356 of "foo.bar", the DNS query will be for 1357 "foo.bar._domainkey.example.com". 1359 3.6.2.2. Resource Record Types for Key Storage 1361 The DNS Resource Record type used is specified by an option to the 1362 query-type ("q=") tag. The only option defined in this base 1363 specification is "txt", indicating the use of a TXT Resource Record 1364 (RR). A later extension of this standard may define another RR type. 1366 Strings in a TXT RR MUST be concatenated together before use with no 1367 intervening whitespace. TXT RRs MUST be unique for a particular 1368 selector name; that is, if there are multiple records in an RRset, 1369 the results are undefined. 1371 TXT RRs are encoded as described in Section 3.6.1 1373 3.7. Computing the Message Hashes 1375 Both signing and verifying message signatures start with a step of 1376 computing two cryptographic hashes over the message. Signers will 1377 choose the parameters of the signature as described in Signer Actions 1378 Section 5; verifiers will use the parameters specified in the DKIM- 1379 Signature header field being verified. In the following discussion, 1380 the names of the tags in the DKIM-Signature header field that either 1381 exists (when verifying) or will be created (when signing) are used. 1382 Note that canonicalization (Section 3.4) is only used to prepare the 1383 email for signing or verifying; it does not affect the transmitted 1384 email in any way. 1386 The signer/verifier MUST compute two hashes, one over the body of the 1387 message and one over the selected header fields of the message. 1389 Signers MUST compute them in the order shown. Verifiers MAY compute 1390 them in any order convenient to the verifier, provided that the 1391 result is semantically identical to the semantics that would be the 1392 case had they been computed in this order. 1394 In hash step 1, the signer/verifier MUST hash the message body, 1395 canonicalized using the body canonicalization algorithm specified in 1396 the "c=" tag and then truncated to the length specified in the "l=" 1397 tag. That hash value is then converted to base64 form and inserted 1398 into (signers) or compared to (verifiers) the "bh=" tag of the DKIM- 1399 Signature header field. 1401 In hash step 2, the signer/verifier MUST pass the following to the 1402 hash algorithm in the indicated order. 1404 1. The header fields specified by the "h=" tag, in the order 1405 specified in that tag, and canonicalized using the header 1406 canonicalization algorithm specified in the "c=" tag. Each 1407 header field MUST be terminated with a single CRLF. 1409 2. The DKIM-Signature header field that exists (verifying) or will 1410 be inserted (signing) in the message, with the value of the "b=" 1411 tag (including all surrounding whitespace) deleted (i.e., treated 1412 as the empty string), canonicalized using the header 1413 canonicalization algorithm specified in the "c=" tag, and without 1414 a trailing CRLF. 1416 All tags and their values in the DKIM-Signature header field are 1417 included in the cryptographic hash with the sole exception of the 1418 value portion of the "b=" (signature) tag, which MUST be treated as 1419 the null string. All tags MUST be included even if they might not be 1420 understood by the verifier. The header field MUST be presented to 1421 the hash algorithm after the body of the message rather than with the 1422 rest of the header fields and MUST be canonicalized as specified in 1423 the "c=" (canonicalization) tag. The DKIM-Signature header field 1424 MUST NOT be included in its own h= tag, although other DKIM-Signature 1425 header fields MAY be signed (see Section 4). 1427 When calculating the hash on messages that will be transmitted using 1428 base64 or quoted-printable encoding, signers MUST compute the hash 1429 after the encoding. Likewise, the verifier MUST incorporate the 1430 values into the hash before decoding the base64 or quoted-printable 1431 text. However, the hash MUST be computed before transport level 1432 encodings such as SMTP "dot-stuffing" (the modification of lines 1433 beginning with a "." to avoid confusion with the SMTP end-of-message 1434 marker, as specified in [RFC5321]). 1436 With the exception of the canonicalization procedure described in 1437 Section 3.4, the DKIM signing process treats the body of messages as 1438 simply a string of octets. DKIM messages MAY be either in plain-text 1439 or in MIME format; no special treatment is afforded to MIME content. 1440 Message attachments in MIME format MUST be included in the content 1441 that is signed. 1443 More formally, pseudo-code for the signature algorithm is: 1444 body-hash = hash-alg (canon-body, l-param) 1445 data-hash = hash-alg (h-headers, D-SIG, content-hash) 1446 signature = sig-alg (d-domain, selector, data-hash) 1448 where: 1450 body-hash: is the output from hashing the body, using hash-alg. 1452 hash-alg: is the hashing algorithm specified in the "a" 1453 parameter. 1455 canon-body: is a canonicalized representation of the body, 1456 produced by using the body algorithm specified in the "c" 1457 parameter, as defined in Section 3.4 and excluding the 1458 DKIM-Signature field. 1460 l-param: is the length-of-body value of the "l" parameter. 1462 data-hash: is the output from using the hash-alg algorithm, to 1463 hash the header including the DKIM-Signature header, and the 1464 body hash. 1466 h-headers: is the list of headers to be signed, as specified in 1467 the "h" parameter. 1469 D-SIG: is the canonicalized DKIM-Signature field without the 1470 signature value portion of the parameter, itself; that is, an 1471 empty parameter value. 1473 signature: is the signature value produced by the signing 1474 algorithm. 1476 sig-alg: is the signature algorithm specified by the "a" 1477 parameter. 1479 d-domain: is the domain name specified in the "d" parameter. 1481 selector: is the selector value specified in the "s" parameter. 1483 NOTE: Many digital signature APIs provide both hashing and 1484 application of the RSA private key using a single "sign()" 1485 primitive. When using such an API, the last two steps in the 1486 algorithm would probably be combined into a single call that would 1487 perform both the "a-hash-alg" and the "sig-alg". 1489 3.8. Input Requirements 1491 DKIM's design is predicated on valid input. Therefore, signers and 1492 verifiers SHOULD take reasonable steps to ensure that the messages 1493 they are processing are valid according to [RFC5322], [RFC2045], and 1494 any other relevant message format standards. See Section 8.15 for 1495 additional discussion and references. 1497 3.9. Output Requirements 1499 For each signature that verifies successfully or produces a TEMPFAIL 1500 result, the output of a DKIM verifier module MUST include the set of: 1502 o The domain name, taken from the "d=" signature tag; and 1504 o The result of the verification attempt for that signature. 1506 The output MAY include other signature properties or result meta- 1507 data, including PERMFAILed or otherwise ignored signatures, for use 1508 by modules that consume those results. 1510 See Section 6.1 for discussion of signature validation result codes. 1512 3.10. Signing by Parent Domains 1514 In some circumstances, it is desirable for a domain to apply a 1515 signature on behalf of any of its subdomains without the need to 1516 maintain separate selectors (key records) in each subdomain. By 1517 default, private keys corresponding to key records can be used to 1518 sign messages for any subdomain of the domain in which they reside; 1519 for example, a key record for the domain example.com can be used to 1520 verify messages where the AUID ("i=" tag of the signature) is 1521 sub.example.com, or even sub1.sub2.example.com. In order to limit 1522 the capability of such keys when this is not intended, the "s" flag 1523 MAY be set in the "t=" tag of the key record, to constrain the 1524 validity of the domain of the AUID. If the referenced key record 1525 contains the "s" flag as part of the "t=" tag, the domain of the AUID 1526 ("i=" flag) MUST be the same as that of the SDID (d=) domain. If 1527 this flag is absent, the domain of the AUID MUST be the same as, or a 1528 subdomain of, the SDID. 1530 3.11. Relationship between SDID and AUID 1532 DKIM's primary task is to communicate from the Signer to a recipient- 1533 side Identity Assessor a single Signing Domain Identifier (SDID) that 1534 refers to a responsible identity. DKIM MAY optionally provide a 1535 single responsible Agent or User Identifier (AUID). 1537 Hence, DKIM's mandatory output to a receive-side Identity Assessor is 1538 a single domain name. Within the scope of its use as DKIM output, 1539 the name has only basic domain name semantics; any possible owner- 1540 specific semantics are outside the scope of DKIM. That is, within 1541 its role as a DKIM identifier, additional semantics cannot be assumed 1542 by an Identity Assessor. 1544 Upon successfully verifying the signature, a receive-side DKIM 1545 verifier MUST communicate the Signing Domain Identifier (d=) to a 1546 consuming Identity Assessor module and MAY communicate the Agent or 1547 User Identifier (i=) if present. 1549 To the extent that a receiver attempts to intuit any structured 1550 semantics for either of the identifiers, this is a heuristic function 1551 that is outside the scope of DKIM's specification and semantics. 1552 Hence, it is relegated to a higher-level service, such as a delivery 1553 handling filter that integrates a variety of inputs and performs 1554 heuristic analysis of them. 1556 INFORMATIVE DISCUSSION: This document does not require the value 1557 of the SDID or AUID to match an identifier in any other message 1558 header field. This requirement is, instead, an assessor policy 1559 issue. The purpose of such a linkage would be to authenticate the 1560 value in that other header field. This, in turn, is the basis for 1561 applying a trust assessment based on the identifier value. Trust 1562 is a broad and complex topic and trust mechanisms are subject to 1563 highly creative attacks. The real-world efficacy of any but the 1564 most basic bindings between the SDID or AUID and other identities 1565 is not well established, nor is its vulnerability to subversion by 1566 an attacker. Hence, reliance on the use of such bindings should 1567 be strictly limited. In particular, it is not at all clear to 1568 what extent a typical end-user recipient can rely on any 1569 assurances that might be made by successful use of the SDID or 1570 AUID. 1572 4. Semantics of Multiple Signatures 1574 4.1. Example Scenarios 1576 There are many reasons why a message might have multiple signatures. 1577 For example, a given signer might sign multiple times, perhaps with 1578 different hashing or signing algorithms during a transition phase. 1580 INFORMATIVE EXAMPLE: Suppose SHA-256 is in the future found to be 1581 insufficiently strong, and DKIM usage transitions to SHA-1024. A 1582 signer might immediately sign using the newer algorithm, but 1583 continue to sign using the older algorithm for interoperability 1584 with verifiers that had not yet upgraded. The signer would do 1585 this by adding two DKIM-Signature header fields, one using each 1586 algorithm. Older verifiers that did not recognize SHA-1024 as an 1587 acceptable algorithm would skip that signature and use the older 1588 algorithm; newer verifiers could use either signature at their 1589 option, and all other things being equal might not even attempt to 1590 verify the other signature. 1592 Similarly, a signer might sign a message including all headers and no 1593 "l=" tag (to satisfy strict verifiers) and a second time with a 1594 limited set of headers and an "l=" tag (in anticipation of possible 1595 message modifications in route to other verifiers). Verifiers could 1596 then choose which signature they preferred. 1598 INFORMATIVE EXAMPLE: A verifier might receive a message with two 1599 signatures, one covering more of the message than the other. If 1600 the signature covering more of the message verified, then the 1601 verifier could make one set of policy decisions; if that signature 1602 failed but the signature covering less of the message verified, 1603 the verifier might make a different set of policy decisions. 1605 Of course, a message might also have multiple signatures because it 1606 passed through multiple signers. A common case is expected to be 1607 that of a signed message that passes through a mailing list that also 1608 signs all messages. Assuming both of those signatures verify, a 1609 recipient might choose to accept the message if either of those 1610 signatures were known to come from trusted sources. 1612 INFORMATIVE EXAMPLE: Recipients might choose to whitelist mailing 1613 lists to which they have subscribed and that have acceptable anti- 1614 abuse policies so as to accept messages sent to that list even 1615 from unknown authors. They might also subscribe to less trusted 1616 mailing lists (e.g., those without anti-abuse protection) and be 1617 willing to accept all messages from specific authors, but insist 1618 on doing additional abuse scanning for other messages. 1620 Another related example of multiple signers might be forwarding 1621 services, such as those commonly associated with academic alumni 1622 sites. 1624 INFORMATIVE EXAMPLE: A recipient might have an address at 1625 members.example.org, a site that has anti-abuse protection that is 1626 somewhat less effective than the recipient would prefer. Such a 1627 recipient might have specific authors whose messages would be 1628 trusted absolutely, but messages from unknown authors that had 1629 passed the forwarder's scrutiny would have only medium trust. 1631 4.2. Interpretation 1633 A signer that is adding a signature to a message merely creates a new 1634 DKIM-Signature header, using the usual semantics of the h= option. A 1635 signer MAY sign previously existing DKIM-Signature header fields 1636 using the method described in Section 5.4 to sign trace header 1637 fields. 1639 INFORMATIVE NOTE: Signers should be cognizant that signing DKIM- 1640 Signature header fields may result in signature failures with 1641 intermediaries that do not recognize that DKIM-Signature header 1642 fields are trace header fields and unwittingly reorder them, thus 1643 breaking such signatures. For this reason, signing existing DKIM- 1644 Signature header fields is unadvised, albeit legal. 1646 INFORMATIVE NOTE: If a header field with multiple instances is 1647 signed, those header fields are always signed from the bottom up. 1648 Thus, it is not possible to sign only specific DKIM-Signature 1649 header fields. For example, if the message being signed already 1650 contains three DKIM-Signature header fields A, B, and C, it is 1651 possible to sign all of them, B and C only, or C only, but not A 1652 only, B only, A and B only, or A and C only. 1654 A signer MAY add more than one DKIM-Signature header field using 1655 different parameters. For example, during a transition period a 1656 signer might want to produce signatures using two different hash 1657 algorithms. 1659 Signers SHOULD NOT remove any DKIM-Signature header fields from 1660 messages they are signing, even if they know that the signatures 1661 cannot be verified. 1663 When evaluating a message with multiple signatures, a verifier SHOULD 1664 evaluate signatures independently and on their own merits. For 1665 example, a verifier that by policy chooses not to accept signatures 1666 with deprecated cryptographic algorithms would consider such 1667 signatures invalid. Verifiers MAY process signatures in any order of 1668 their choice; for example, some verifiers might choose to process 1669 signatures corresponding to the From field in the message header 1670 before other signatures. See Section 6.1 for more information about 1671 signature choices. 1673 INFORMATIVE IMPLEMENTATION NOTE: Verifier attempts to correlate 1674 valid signatures with invalid signatures in an attempt to guess 1675 why a signature failed are ill-advised. In particular, there is 1676 no general way that a verifier can determine that an invalid 1677 signature was ever valid. 1679 Verifiers SHOULD ignore those signatures that produce a PERMFAIL 1680 result (see Section 6.1), acting as though they were not present in 1681 the message. Verifiers SHOULD continue to check signatures until a 1682 signature successfully verifies to the satisfaction of the verifier. 1683 To limit potential denial-of-service attacks, verifiers MAY limit the 1684 total number of signatures they will attempt to verify. 1686 5. Signer Actions 1688 The following steps are performed in order by signers. 1690 5.1. Determine Whether the Email Should Be Signed and by Whom 1692 A signer can obviously only sign email for domains for which it has a 1693 private key and the necessary knowledge of the corresponding public 1694 key and selector information. However, there are a number of other 1695 reasons beyond the lack of a private key why a signer could choose 1696 not to sign an email. 1698 INFORMATIVE NOTE: Signing modules may be incorporated into any 1699 portion of the mail system as deemed appropriate, including an 1700 MUA, a SUBMISSION server, or an MTA. Wherever implemented, 1701 signers should beware of signing (and thereby asserting 1702 responsibility for) messages that may be problematic. In 1703 particular, within a trusted enclave the signing address might be 1704 derived from the header according to local policy; SUBMISSION 1705 servers might only sign messages from users that are properly 1706 authenticated and authorized. 1708 INFORMATIVE IMPLEMENTER ADVICE: SUBMISSION servers should not sign 1709 Received header fields if the outgoing gateway MTA obfuscates 1710 Received header fields, for example, to hide the details of 1711 internal topology. 1713 If an email cannot be signed for some reason, it is a local policy 1714 decision as to what to do with that email. 1716 5.2. Select a Private Key and Corresponding Selector Information 1718 This specification does not define the basis by which a signer should 1719 choose which private key and selector information to use. Currently, 1720 all selectors are equal as far as this specification is concerned, so 1721 the decision should largely be a matter of administrative 1722 convenience. Distribution and management of private keys is also 1723 outside the scope of this document. 1725 INFORMATIVE OPERATIONS ADVICE: A signer should not sign with a 1726 private key when the selector containing the corresponding public 1727 key is expected to be revoked or removed before the verifier has 1728 an opportunity to validate the signature. The signer should 1729 anticipate that verifiers may choose to defer validation, perhaps 1730 until the message is actually read by the final recipient. In 1731 particular, when rotating to a new key pair, signing should 1732 immediately commence with the new private key and the old public 1733 key should be retained for a reasonable validation interval before 1734 being removed from the key server. 1736 5.3. Normalize the Message to Prevent Transport Conversions 1738 Some messages, particularly those using 8-bit characters, are subject 1739 to modification during transit, notably conversion to 7-bit form. 1740 Such conversions will break DKIM signatures. In order to minimize 1741 the chances of such breakage, signers SHOULD convert the message to a 1742 suitable MIME content transfer encoding such as quoted-printable or 1743 base64 as described in [RFC2045] before signing. Such conversion is 1744 outside the scope of DKIM; the actual message SHOULD be converted to 1745 7-bit MIME by an MUA or MSA prior to presentation to the DKIM 1746 algorithm. 1748 Similarly, a message that is not compliant with RFC5322, RFC2045 and 1749 RFC2047 can be subject to attempts by intermediaries to correct or 1750 interpret such content. See Section 8 of [RFC4409] for examples of 1751 changes that are commonly made. Such "corrections" may break DKIM 1752 signatures or have other undesirable effects. Therefore, a verifier 1753 SHOULD NOT validate a message that is not compliant with those 1754 specifications. 1756 If the message is submitted to the signer with any local encoding 1757 that will be modified before transmission, that modification to 1758 canonical [RFC5322] form MUST be done before signing. In particular, 1759 bare CR or LF characters (used by some systems as a local line 1760 separator convention) MUST be converted to the SMTP-standard CRLF 1761 sequence before the message is signed. Any conversion of this sort 1762 SHOULD be applied to the message actually sent to the recipient(s), 1763 not just to the version presented to the signing algorithm. 1765 More generally, the signer MUST sign the message as it is expected to 1766 be received by the verifier rather than in some local or internal 1767 form. 1769 5.4. Determine the Header Fields to Sign 1771 The From header field MUST be signed (that is, included in the "h=" 1772 tag of the resulting DKIM-Signature header field). Signers SHOULD 1773 NOT sign an existing header field likely to be legitimately modified 1774 or removed in transit. In particular, [RFC5321] explicitly permits 1775 modification or removal of the Return-Path header field in transit. 1776 Signers MAY include any other header fields present at the time of 1777 signing at the discretion of the signer. 1779 INFORMATIVE OPERATIONS NOTE: The choice of which header fields to 1780 sign is non-obvious. One strategy is to sign all existing, non- 1781 repeatable header fields. An alternative strategy is to sign only 1782 header fields that are likely to be displayed to or otherwise be 1783 likely to affect the processing of the message at the receiver. A 1784 third strategy is to sign only "well known" headers. Note that 1785 verifiers may treat unsigned header fields with extreme 1786 skepticism, including refusing to display them to the end user or 1787 even ignoring the signature if it does not cover certain header 1788 fields. For this reason, signing fields present in the message 1789 such as Date, Subject, Reply-To, Sender, and all MIME header 1790 fields are highly advised. 1792 The DKIM-Signature header field is always implicitly signed and MUST 1793 NOT be included in the "h=" tag except to indicate that other 1794 preexisting signatures are also signed. 1796 Signers MAY claim to have signed header fields that do not exist 1797 (that is, signers MAY include the header field name in the "h=" tag 1798 even if that header field does not exist in the message). When 1799 computing the signature, the non-existing header field MUST be 1800 treated as the null string (including the header field name, header 1801 field value, all punctuation, and the trailing CRLF). 1803 INFORMATIVE RATIONALE: This allows signers to explicitly assert 1804 the absence of a header field; if that header field is added later 1805 the signature will fail. 1807 INFORMATIVE NOTE: A header field name need only be listed once 1808 more than the actual number of that header field in a message at 1809 the time of signing in order to prevent any further additions. 1810 For example, if there is a single Comments header field at the 1811 time of signing, listing Comments twice in the "h=" tag is 1812 sufficient to prevent any number of Comments header fields from 1813 being appended; it is not necessary (but is legal) to list 1814 Comments three or more times in the "h=" tag. 1816 Signers choosing to sign an existing header field that occurs more 1817 than once in the message (such as Received) MUST sign the physically 1818 last instance of that header field in the header block. Signers 1819 wishing to sign multiple instances of such a header field MUST 1820 include the header field name multiple times in the h= tag of the 1821 DKIM-Signature header field, and MUST sign such header fields in 1822 order from the bottom of the header field block to the top. The 1823 signer MAY include more instances of a header field name in h= than 1824 there are actual corresponding header fields to indicate that 1825 additional header fields of that name SHOULD NOT be added. 1827 INFORMATIVE EXAMPLE: 1829 If the signer wishes to sign two existing Received header fields, 1830 and the existing header contains: 1831 Received: 1832 Received: 1833 Received: 1835 then the resulting DKIM-Signature header field should read: 1837 DKIM-Signature: ... h=Received : Received :... 1838 and Received header fields and will be signed in that 1839 order. 1841 Signers should be careful of signing header fields that might have 1842 additional instances added later in the delivery process, since such 1843 header fields might be inserted after the signed instance or 1844 otherwise reordered. Trace header fields (such as Received) and 1845 Resent-* blocks are the only fields prohibited by [RFC5322] from 1846 being reordered. In particular, since DKIM-Signature header fields 1847 may be reordered by some intermediate MTAs, signing existing DKIM- 1848 Signature header fields is error-prone. 1850 INFORMATIVE ADMONITION: Despite the fact that [RFC5322] permits 1851 header fields to be reordered (with the exception of Received 1852 header fields), reordering of signed header fields with multiple 1853 instances by intermediate MTAs will cause DKIM signatures to be 1854 broken; such anti-social behavior should be avoided. 1856 INFORMATIVE IMPLEMENTER'S NOTE: Although not required by this 1857 specification, all end-user visible header fields should be signed 1858 to avoid possible "indirect spamming". For example, if the 1859 Subject header field is not signed, a spammer can resend a 1860 previously signed mail, replacing the legitimate subject with a 1861 one-line spam. 1863 5.5. Recommended Signature Content 1865 The purpose of the DKIM cryptographic algorithm is to affix an 1866 identifier to the message in a way that is both robust against normal 1867 transit-related changes and resistant to kinds of replay attacks. An 1868 essential aspect of satisfying these requirements is choosing what 1869 header fields to include in the hash and what fields to exclude. 1871 The basic rule for choosing fields to include is to select those 1872 fields that constitute the "core" of the message content. Hence, any 1873 replay attack will have to include these in order to have the 1874 signature succeed; but with these included, the core of the message 1875 is valid, even if sent on to new recipients. 1877 Common examples of fields with addresses and fields with textual 1878 content related to the body are: 1880 o From (REQUIRED; see Section 5.4) 1882 o Reply-To 1884 o Subject 1886 o Date 1888 o To, Cc 1890 o Resent-Date, Resent-From, Resent-To, Resent-Cc 1892 o In-Reply-To, References 1894 o List-Id, List-Help, List-Unsubscribe, List-Subscribe, List-Post, 1895 List-Owner, List-Archive 1897 If the "l=" signature tag is in use (see Section 3.5), the Content- 1898 Type field is also a candidate for being included as it could be 1899 replaced in a way that causes completely different content to be 1900 rendered to the receiving user. 1902 There are tradeoffs in the decision of what constitutes the "core" of 1903 the message, which for some fields is a subjective concept. 1905 Including fields such as "Message-ID" for example is useful if one 1906 considers a mechanism for being able to distinguish separate 1907 instances of the same message to be core content. Similarly, "In- 1908 Reply-To" and "References" might be desirable to include if one 1909 considers message threading to be a core part of the message. 1911 Another class of fields that may be of interest are those that convey 1912 security-related information about the message, such as 1913 Authentication-Results [RFC5451]. 1915 The basic rule for choosing field to exclude is to select those 1916 fields for which there are multiple fields with the same name, and 1917 fields that are modified in transit. Examples of these are: 1919 o Return-Path 1921 o Received 1923 o Comments, Keywords 1925 Note that the DKIM-Signature field is also excluded from the header 1926 hash, because its handling is specified separately. 1928 Typically, it is better to exclude other, optional fields because of 1929 the potential that additional fields of the same name will be 1930 legitimately added or re-ordered prior to verification. There are 1931 likely to be legitimate exceptions to this rule, because of the wide 1932 variety of application-specific header fields that might be applied 1933 to a message, some of which are unlikely to be duplicated, modified, 1934 or reordered. 1936 Signers SHOULD choose canonicalization algorithms based on the types 1937 of messages they process and their aversion to risk. For example, 1938 e-commerce sites sending primarily purchase receipts, which are not 1939 expected to be processed by mailing lists or other software likely to 1940 modify messages, will generally prefer "simple" canonicalization. 1941 Sites sending primarily person-to-person email will likely prefer to 1942 be more resilient to modification during transport by using "relaxed" 1943 canonicalization. 1945 Signers SHOULD NOT use "l=" unless they intend to accommodate 1946 intermediate mail processors that append text to a message. For 1947 example, many mailing list processors append "unsubscribe" 1948 information to message bodies. If signers use "l=", they SHOULD 1949 include the entire message body existing at the time of signing in 1950 computing the count. In particular, signers SHOULD NOT specify a 1951 body length of 0 since this may be interpreted as a meaningless 1952 signature by some verifiers. 1954 5.6. Compute the Message Hash and Signature 1956 The signer MUST compute the message hash as described in Section 3.7 1957 and then sign it using the selected public-key algorithm. This will 1958 result in a DKIM-Signature header field that will include the body 1959 hash and a signature of the header hash, where that header includes 1960 the DKIM-Signature header field itself. 1962 Entities such as mailing list managers that implement DKIM and that 1963 modify the message or a header field (for example, inserting 1964 unsubscribe information) before retransmitting the message SHOULD 1965 check any existing signature on input and MUST make such 1966 modifications before re-signing the message. 1968 The signer MAY elect to limit the number of bytes of the body that 1969 will be included in the hash and hence signed. The length actually 1970 hashed should be inserted in the "l=" tag of the DKIM-Signature 1971 header field. 1973 5.7. Insert the DKIM-Signature Header Field 1975 Finally, the signer MUST insert the DKIM-Signature header field 1976 created in the previous step prior to transmitting the email. The 1977 DKIM-Signature header field MUST be the same as used to compute the 1978 hash as described above, except that the value of the "b=" tag MUST 1979 be the appropriately signed hash computed in the previous step, 1980 signed using the algorithm specified in the "a=" tag of the DKIM- 1981 Signature header field and using the private key corresponding to the 1982 selector given in the "s=" tag of the DKIM-Signature header field, as 1983 chosen above in Section 5.2 1985 The DKIM-Signature header field MUST be inserted before any other 1986 DKIM-Signature fields in the header block. 1988 INFORMATIVE IMPLEMENTATION NOTE: The easiest way to achieve this 1989 is to insert the DKIM-Signature header field at the beginning of 1990 the header block. In particular, it may be placed before any 1991 existing Received header fields. This is consistent with treating 1992 DKIM-Signature as a trace header field. 1994 6. Verifier Actions 1996 Since a signer MAY remove or revoke a public key at any time, it is 1997 recommended that verification occur in a timely manner. In many 1998 configurations, the most timely place is during acceptance by the 1999 border MTA or shortly thereafter. In particular, deferring 2000 verification until the message is accessed by the end user is 2001 discouraged. 2003 A border or intermediate MTA MAY verify the message signature(s). An 2004 MTA who has performed verification MAY communicate the result of that 2005 verification by adding a verification header field to incoming 2006 messages. This considerably simplifies things for the user, who can 2007 now use an existing mail user agent. Most MUAs have the ability to 2008 filter messages based on message header fields or content; these 2009 filters would be used to implement whatever policy the user wishes 2010 with respect to unsigned mail. 2012 A verifying MTA MAY implement a policy with respect to unverifiable 2013 mail, regardless of whether or not it applies the verification header 2014 field to signed messages. 2016 Verifiers MUST produce a result that is semantically equivalent to 2017 applying the following steps in the order listed. In practice, 2018 several of these steps can be performed in parallel in order to 2019 improve performance. 2021 6.1. Extract Signatures from the Message 2023 The order in which verifiers try DKIM-Signature header fields is not 2024 defined; verifiers MAY try signatures in any order they like. For 2025 example, one implementation might try the signatures in textual 2026 order, whereas another might try signatures by identities that match 2027 the contents of the From header field before trying other signatures. 2028 Verifiers MUST NOT attribute ultimate meaning to the order of 2029 multiple DKIM-Signature header fields. In particular, there is 2030 reason to believe that some relays will reorder the header fields in 2031 potentially arbitrary ways. 2033 INFORMATIVE IMPLEMENTATION NOTE: Verifiers might use the order as 2034 a clue to signing order in the absence of any other information. 2035 However, other clues as to the semantics of multiple signatures 2036 (such as correlating the signing host with Received header fields) 2037 might also be considered. 2039 A verifier SHOULD NOT treat a message that has one or more bad 2040 signatures and no good signatures differently from a message with no 2041 signature at all; such treatment is a matter of local policy and is 2042 beyond the scope of this document. 2044 When a signature successfully verifies, a verifier will either stop 2045 processing or attempt to verify any other signatures, at the 2046 discretion of the implementation. A verifier MAY limit the number of 2047 signatures it tries to avoid denial-of-service attacks. 2049 INFORMATIVE NOTE: An attacker could send messages with large 2050 numbers of faulty signatures, each of which would require a DNS 2051 lookup and corresponding CPU time to verify the message. This 2052 could be an attack on the domain that receives the message, by 2053 slowing down the verifier by requiring it to do a large number of 2054 DNS lookups and/or signature verifications. It could also be an 2055 attack against the domains listed in the signatures, essentially 2056 by enlisting innocent verifiers in launching an attack against the 2057 DNS servers of the actual victim. 2059 In the following description, text reading "return status 2060 (explanation)" (where "status" is one of "PERMFAIL" or "TEMPFAIL") 2061 means that the verifier MUST immediately cease processing that 2062 signature. The verifier SHOULD proceed to the next signature, if any 2063 is present, and completely ignore the bad signature. If the status 2064 is "PERMFAIL", the signature failed and should not be reconsidered. 2065 If the status is "TEMPFAIL", the signature could not be verified at 2066 this time but may be tried again later. A verifier MAY either 2067 arrange to defer the message for later processing, or try another 2068 signature; if no good signature is found and any of the signatures 2069 resulted in a TEMPFAIL status, the verifier MAY arrange to defer the 2070 message for later processing. The "(explanation)" is not normative 2071 text; it is provided solely for clarification. 2073 Verifiers SHOULD ignore any DKIM-Signature header fields where the 2074 signature does not validate. Verifiers that are prepared to validate 2075 multiple signature header fields SHOULD proceed to the next signature 2076 header field, if one exists. However, verifiers MAY make note of the 2077 fact that an invalid signature was present for consideration at a 2078 later step. 2080 INFORMATIVE NOTE: The rationale of this requirement is to permit 2081 messages that have invalid signatures but also a valid signature 2082 to work. For example, a mailing list exploder might opt to leave 2083 the original submitter signature in place even though the exploder 2084 knows that it is modifying the message in some way that will break 2085 that signature, and the exploder inserts its own signature. In 2086 this case, the message should succeed even in the presence of the 2087 known-broken signature. 2089 For each signature to be validated, the following steps should be 2090 performed in such a manner as to produce a result that is 2091 semantically equivalent to performing them in the indicated order. 2093 6.1.1. Validate the Signature Header Field 2095 Implementers MUST meticulously validate the format and values in the 2096 DKIM-Signature header field; any inconsistency or unexpected values 2097 MUST cause the header field to be completely ignored and the verifier 2098 to return PERMFAIL (signature syntax error). Being "liberal in what 2099 you accept" is definitely a bad strategy in this security context. 2100 Note however that this does not include the existence of unknown tags 2101 in a DKIM-Signature header field, which are explicitly permitted. 2102 Verifiers MUST ignore DKIM-Signature header fields with a "v=" tag 2103 that is inconsistent with this specification and return PERMFAIL 2104 (incompatible version). 2106 INFORMATIVE IMPLEMENTATION NOTE: An implementation may, of course, 2107 choose to also verify signatures generated by older versions of 2108 this specification. 2110 If any tag listed as "required" in Section 3.5 is omitted from the 2111 DKIM-Signature header field, the verifier MUST ignore the DKIM- 2112 Signature header field and return PERMFAIL (signature missing 2113 required tag). 2115 INFORMATIONAL NOTE: The tags listed as required in Section 3.5 are 2116 "v=", "a=", "b=", "bh=", "d=", "h=", and "s=". Should there be a 2117 conflict between this note and Section 3.5, Section 3.5 is 2118 normative. 2120 If the DKIM-Signature header field does not contain the "i=" tag, the 2121 verifier MUST behave as though the value of that tag were "@d", where 2122 "d" is the value from the "d=" tag. 2124 Verifiers MUST confirm that the domain specified in the "d=" tag is 2125 the same as or a parent domain of the domain part of the "i=" tag. 2126 If not, the DKIM-Signature header field MUST be ignored and the 2127 verifier should return PERMFAIL (domain mismatch). 2129 If the "h=" tag does not include the From header field, the verifier 2130 MUST ignore the DKIM-Signature header field and return PERMFAIL (From 2131 field not signed). 2133 Verifiers MAY ignore the DKIM-Signature header field and return 2134 PERMFAIL (signature expired) if it contains an "x=" tag and the 2135 signature has expired. 2137 Verifiers MAY ignore the DKIM-Signature header field if the domain 2138 used by the signer in the "d=" tag is not associated with a valid 2139 signing entity. For example, signatures with "d=" values such as 2140 "com" and "co.uk" may be ignored. The list of unacceptable domains 2141 SHOULD be configurable. 2143 Verifiers MAY ignore the DKIM-Signature header field and return 2144 PERMFAIL (unacceptable signature header) for any other reason, for 2145 example, if the signature does not sign header fields that the 2146 verifier views to be essential. As a case in point, if MIME header 2147 fields are not signed, certain attacks may be possible that the 2148 verifier would prefer to avoid. 2150 6.1.2. Get the Public Key 2152 The public key for a signature is needed to complete the verification 2153 process. The process of retrieving the public key depends on the 2154 query type as defined by the "q=" tag in the DKIM-Signature header 2155 field. Obviously, a public key need only be retrieved if the process 2156 of extracting the signature information is completely successful. 2157 Details of key management and representation are described in 2158 Section 3.6. The verifier MUST validate the key record and MUST 2159 ignore any public key records that are malformed. 2161 NOTE: The use of a wildcard TXT record that covers a queried DKIM 2162 domain name will produce a response to a DKIM query that is 2163 unlikely to be valid DKIM key record. This problem is not 2164 specific to DKIM and applies to many other types of queries. 2165 Client software that processes DNS responses needs to take this 2166 problem into account. 2168 When validating a message, a verifier MUST perform the following 2169 steps in a manner that is semantically the same as performing them in 2170 the order indicated -- in some cases the implementation may 2171 parallelize or reorder these steps, as long as the semantics remain 2172 unchanged: 2174 1. Retrieve the public key as described in Section 3.6 using the 2175 algorithm in the "q=" tag, the domain from the "d=" tag, and the 2176 selector from the "s=" tag. 2178 2. If the query for the public key fails to respond, the verifier 2179 MAY seek a later verification attempt by returning TEMPFAIL (key 2180 unavailable). 2182 3. If the query for the public key fails because the corresponding 2183 key record does not exist, the verifier MUST immediately return 2184 PERMFAIL (no key for signature). 2186 4. If the query for the public key returns multiple key records, the 2187 verifier may choose one of the key records or may cycle through 2188 the key records performing the remainder of these steps on each 2189 record at the discretion of the implementer. The order of the 2190 key records is unspecified. If the verifier chooses to cycle 2191 through the key records, then the "return ..." wording in the 2192 remainder of this section means "try the next key record, if any; 2193 if none, return to try another signature in the usual way". 2195 5. If the result returned from the query does not adhere to the 2196 format defined in this specification, the verifier MUST ignore 2197 the key record and return PERMFAIL (key syntax error). Verifiers 2198 are urged to validate the syntax of key records carefully to 2199 avoid attempted attacks. In particular, the verifier MUST ignore 2200 keys with a version code ("v=" tag) that they do not implement. 2202 6. If the "h=" tag exists in the public key record and the hash 2203 algorithm implied by the "a=" tag in the DKIM-Signature header 2204 field is not included in the contents of the "h=" tag, the 2205 verifier MUST ignore the key record and return PERMFAIL 2206 (inappropriate hash algorithm). 2208 7. If the public key data (the "p=" tag) is empty, then this key has 2209 been revoked and the verifier MUST treat this as a failed 2210 signature check and return PERMFAIL (key revoked). There is no 2211 defined semantic difference between a key that has been revoked 2212 and a key record that has been removed. 2214 8. If the public key data is not suitable for use with the algorithm 2215 and key types defined by the "a=" and "k=" tags in the DKIM- 2216 Signature header field, the verifier MUST immediately return 2217 PERMFAIL (inappropriate key algorithm). 2219 6.1.3. Compute the Verification 2221 Given a signer and a public key, verifying a signature consists of 2222 actions semantically equivalent to the following steps. 2224 1. Based on the algorithm defined in the "c=" tag, the body length 2225 specified in the "l=" tag, and the header field names in the "h=" 2226 tag, prepare a canonicalized version of the message as is 2227 described in Section 3.7 (note that this version does not 2228 actually need to be instantiated). When matching header field 2229 names in the "h=" tag against the actual message header field, 2230 comparisons MUST be case-insensitive. 2232 2. Based on the algorithm indicated in the "a=" tag, compute the 2233 message hashes from the canonical copy as described in 2234 Section 3.7. 2236 3. Verify that the hash of the canonicalized message body computed 2237 in the previous step matches the hash value conveyed in the "bh=" 2238 tag. If the hash does not match, the verifier SHOULD ignore the 2239 signature and return PERMFAIL (body hash did not verify). 2241 4. Using the signature conveyed in the "b=" tag, verify the 2242 signature against the header hash using the mechanism appropriate 2243 for the public key algorithm described in the "a=" tag. If the 2244 signature does not validate, the verifier SHOULD ignore the 2245 signature and return PERMFAIL (signature did not verify). 2247 5. Otherwise, the signature has correctly verified. 2249 INFORMATIVE IMPLEMENTER'S NOTE: Implementations might wish to 2250 initiate the public-key query in parallel with calculating the 2251 hash as the public key is not needed until the final decryption is 2252 calculated. Implementations may also verify the signature on the 2253 message header before validating that the message hash listed in 2254 the "bh=" tag in the DKIM-Signature header field matches that of 2255 the actual message body; however, if the body hash does not match, 2256 the entire signature must be considered to have failed. 2258 A body length specified in the "l=" tag of the signature limits the 2259 number of bytes of the body passed to the verification algorithm. 2260 All data beyond that limit is not validated by DKIM. Hence, 2261 verifiers might treat a message that contains bytes beyond the 2262 indicated body length with suspicion, such as by declaring the 2263 signature invalid (e.g., by returning PERMFAIL (unsigned content)), 2264 or conveying the partial verification to the policy module. 2266 6.2. Communicate Verification Results 2268 Verifiers wishing to communicate the results of verification to other 2269 parts of the mail system may do so in whatever manner they see fit. 2270 For example, implementations might choose to add an email header 2271 field to the message before passing it on. Any such header field 2272 SHOULD be inserted before any existing DKIM-Signature or preexisting 2273 authentication status header fields in the header field block. The 2274 Authentication-Results: header field ([RFC5451]) MAY be used for this 2275 purpose. 2277 INFORMATIVE ADVICE to MUA filter writers: Patterns intended to 2278 search for results header fields to visibly mark authenticated 2279 mail for end users should verify that such header field was added 2280 by the appropriate verifying domain and that the verified identity 2281 matches the author identity that will be displayed by the MUA. In 2282 particular, MUA filters should not be influenced by bogus results 2283 header fields added by attackers. To circumvent this attack, 2284 verifiers MAY wish to request deletion of existing results header 2285 fields after verification and before arranging to add a new header 2286 field. 2288 6.3. Interpret Results/Apply Local Policy 2290 It is beyond the scope of this specification to describe what actions 2291 an Identity Assessor can make, but mail carrying a validated SDID 2292 presents an opportunity to an Identity Assessor that unauthenticated 2293 email does not. Specifically, an authenticated email creates a 2294 predictable identifier by which other decisions can reliably be 2295 managed, such as trust and reputation. Conversely, unauthenticated 2296 email lacks a reliable identifier that can be used to assign trust 2297 and reputation. It is reasonable to treat unauthenticated email as 2298 lacking any trust and having no positive reputation. 2300 In general, modules that consume DKIM verification output SHOULD NOT 2301 determine message acceptability based solely on a lack of any 2302 signature or on an unverifiable signature; such rejection would cause 2303 severe interoperability problems. If an MTA does wish to reject such 2304 messages during an SMTP session (for example, when communicating with 2305 a peer who, by prior agreement, agrees to only send signed messages), 2306 and a signature is missing or does not verify, the handling MTA 2307 SHOULD use a 550/5.7.x reply code. 2309 Where the verifier is integrated within the MTA and it is not 2310 possible to fetch the public key, perhaps because the key server is 2311 not available, a temporary failure message MAY be generated using a 2312 451/4.7.5 reply code, such as: 2313 451 4.7.5 Unable to verify signature - key server unavailable 2315 Temporary failures such as inability to access the key server or 2316 other external service are the only conditions that SHOULD use a 4xx 2317 SMTP reply code. In particular, cryptographic signature verification 2318 failures MUST NOT provoke 4xx SMTP replies. 2320 Once the signature has been verified, that information MUST be 2321 conveyed to the Identity Assessor (such as an explicit allow/ 2322 whitelist and reputation system) and/or to the end user. If the SDID 2323 is not the same as the address in the From: header field, the mail 2324 system SHOULD take pains to ensure that the actual SDID is clear to 2325 the reader. 2327 While the symptoms of a failed verification are obvious -- the 2328 signature doesn't verify -- establishing the exact cause can be more 2329 difficult. If a selector cannot be found, is that because the 2330 selector has been removed, or was the value changed somehow in 2331 transit? If the signature line is missing, is that because it was 2332 never there, or was it removed by an overzealous filter? For 2333 diagnostic purposes, the exact reason why the verification fails 2334 SHOULD be made available and possibly recorded in the system logs. 2335 If the email cannot be verified, then it SHOULD be treated the same 2336 as all unverified email regardless of whether or not it looks like it 2337 was signed. 2339 7. IANA Considerations 2341 DKIM has registered namespaces with IANA. In all cases, new values 2342 are assigned only for values that have been documented in a published 2343 RFC that has IETF Consensus [RFC5226]. 2345 This memo updates these registries as described below. Of note is 2346 the addition of a new "status" column. All registrations into these 2347 namespaces MUST include the name being registered, the document in 2348 which it was registered or updated, and an indication of its current 2349 status which MUST be one of "active" (in current use) or "historic" 2350 (no longer in current use). 2352 7.1. DKIM-Signature Tag Specifications 2354 A DKIM-Signature provides for a list of tag specifications. IANA has 2355 established the DKIM-Signature Tag Specification Registry for tag 2356 specifications that can be used in DKIM-Signature fields. 2358 The updated entries in the registry comprise: 2360 +------+-----------------+--------+ 2361 | TYPE | REFERENCE | STATUS | 2362 +------+-----------------+--------+ 2363 | v | (this document) | active | 2364 | a | (this document) | active | 2365 | b | (this document) | active | 2366 | bh | (this document) | active | 2367 | c | (this document) | active | 2368 | d | (this document) | active | 2369 | h | (this document) | active | 2370 | i | (this document) | active | 2371 | l | (this document) | active | 2372 | q | (this document) | active | 2373 | s | (this document) | active | 2374 | t | (this document) | active | 2375 | x | (this document) | active | 2376 | z | (this document) | active | 2377 +------+-----------------+--------+ 2379 Table 1: DKIM-Signature Tag Specification Registry Updated Values 2381 7.2. DKIM-Signature Query Method Registry 2383 The "q=" tag-spec (specified in Section 3.5) provides for a list of 2384 query methods. 2386 IANA has established the DKIM-Signature Query Method Registry for 2387 mechanisms that can be used to retrieve the key that will permit 2388 validation processing of a message signed using DKIM. 2390 The updated entry in the registry comprises: 2392 +------+--------+-----------------+--------+ 2393 | TYPE | OPTION | REFERENCE | STATUS | 2394 +------+--------+-----------------+--------+ 2395 | dns | txt | (this document) | active | 2396 +------+--------+-----------------+--------+ 2398 DKIM-Signature Query Method Registry Updated Values 2400 7.3. DKIM-Signature Canonicalization Registry 2402 The "c=" tag-spec (specified in Section 3.5) provides for a specifier 2403 for canonicalization algorithms for the header and body of the 2404 message. 2406 IANA has established the DKIM-Signature Canonicalization Algorithm 2407 Registry for algorithms for converting a message into a canonical 2408 form before signing or verifying using DKIM. 2410 The updated entries in the header registry comprise: 2412 +---------+-----------------+--------+ 2413 | TYPE | REFERENCE | STATUS | 2414 +---------+-----------------+--------+ 2415 | simple | (this document) | active | 2416 | relaxed | (this document) | active | 2417 +---------+-----------------+--------+ 2419 DKIM-Signature Header Canonicalization Algorithm Registry 2420 Updated Values 2422 The updated entries in the body registry comprise: 2424 +---------+-----------------+--------+ 2425 | TYPE | REFERENCE | STATUS | 2426 +---------+-----------------+--------+ 2427 | simple | (this document) | active | 2428 | relaxed | (this document) | active | 2429 +---------+-----------------+--------+ 2431 DKIM-Signature Body Canonicalization Algorithm Registry 2432 Updated Values 2434 7.4. _domainkey DNS TXT Record Tag Specifications 2436 A _domainkey DNS TXT record provides for a list of tag 2437 specifications. IANA has established the DKIM _domainkey DNS TXT Tag 2438 Specification Registry for tag specifications that can be used in DNS 2439 TXT Records. 2441 The updated entries in the registry comprise: 2443 +------+-----------------+----------+ 2444 | TYPE | REFERENCE | STATUS | 2445 +------+-----------------+----------+ 2446 | v | (this document) | active | 2447 | g | [RFC4871] | historic | 2448 | h | (this document) | active | 2449 | k | (this document) | active | 2450 | n | (this document) | active | 2451 | p | (this document) | active | 2452 | s | (this document) | active | 2453 | t | (this document) | active | 2454 +------+-----------------+----------+ 2456 DKIM _domainkey DNS TXT Record Tag Specification Registry 2457 Updated Values 2459 7.5. DKIM Key Type Registry 2461 The "k=" (specified in Section 3.6.1) and the "a=" (specified in Section 3.5) tags provide for a list of 2463 mechanisms that can be used to decode a DKIM signature. 2465 IANA has established the DKIM Key Type Registry for such mechanisms. 2467 The updated entry in the registry comprises: 2469 +------+-----------+--------+ 2470 | TYPE | REFERENCE | STATUS | 2471 +------+-----------+--------+ 2472 | rsa | [RFC3447] | active | 2473 +------+-----------+--------+ 2475 DKIM Key Type Updated Values 2477 7.6. DKIM Hash Algorithms Registry 2479 The "h=" (specified in Section 3.6.1) and the "a=" (specified in Section 3.5) tags provide for a list of 2481 mechanisms that can be used to produce a digest of message data. 2483 IANA has established the DKIM Hash Algorithms Registry for such 2484 mechanisms. 2486 The updated entries in the registry comprise: 2488 +--------+-------------------+--------+ 2489 | TYPE | REFERENCE | STATUS | 2490 +--------+-------------------+--------+ 2491 | sha1 | [FIPS-180-2-2002] | active | 2492 | sha256 | [FIPS-180-2-2002] | active | 2493 +--------+-------------------+--------+ 2495 DKIM Hash Algorithms Updated Values 2497 7.7. DKIM Service Types Registry 2499 The "s=" tag (specified in Section 3.6.1) provides for a 2500 list of service types to which this selector may apply. 2502 IANA has established the DKIM Service Types Registry for service 2503 types. 2505 The updated entries in the registry comprise: 2507 +-------+-----------------+--------+ 2508 | TYPE | REFERENCE | STATUS | 2509 +-------+-----------------+--------+ 2510 | email | (this document) | active | 2511 | * | (this document) | active | 2512 +-------+-----------------+--------+ 2514 DKIM Service Types Registry Updated Values 2516 7.8. DKIM Selector Flags Registry 2518 The "t=" tag (specified in Section 3.6.1) provides for a 2519 list of flags to modify interpretation of the selector. 2521 IANA has established the DKIM Selector Flags Registry for additional 2522 flags. 2524 The updated entries in the registry comprise: 2526 +------+-----------------+--------+ 2527 | TYPE | REFERENCE | STATUS | 2528 +------+-----------------+--------+ 2529 | y | (this document) | active | 2530 | s | (this document) | active | 2531 +------+-----------------+--------+ 2533 DKIM Selector Flags Registry Updated Values 2535 7.9. DKIM-Signature Header Field 2537 IANA has added DKIM-Signature to the "Permanent Message Header 2538 Fields" registry (see [RFC3864]) for the "mail" protocol, using this 2539 document as the reference. 2541 8. Security Considerations 2543 It has been observed that any mechanism that is introduced that 2544 attempts to stem the flow of spam is subject to intensive attack. 2545 DKIM needs to be carefully scrutinized to identify potential attack 2546 vectors and the vulnerability to each. See also [RFC4686]. 2548 8.1. Misuse of Body Length Limits ("l=" Tag) 2550 As noted in Section 3.4.5, use of the "l=" signature tag enables a 2551 variety of attacks in which added content can partially or completely 2552 change the recipient's view of the message. 2554 8.2. Misappropriated Private Key 2556 If the private key for a user is resident on their computer and is 2557 not protected by an appropriately secure mechanism, it is possible 2558 for malware to send mail as that user and any other user sharing the 2559 same private key. The malware would not, however, be able to 2560 generate signed spoofs of other signers' addresses, which would aid 2561 in identification of the infected user and would limit the 2562 possibilities for certain types of attacks involving socially 2563 engineered messages. This threat applies mainly to MUA-based 2564 implementations; protection of private keys on servers can be easily 2565 achieved through the use of specialized cryptographic hardware. 2567 A larger problem occurs if malware on many users' computers obtains 2568 the private keys for those users and transmits them via a covert 2569 channel to a site where they can be shared. The compromised users 2570 would likely not know of the misappropriation until they receive 2571 "bounce" messages from messages they are purported to have sent. 2572 Many users might not understand the significance of these bounce 2573 messages and would not take action. 2575 One countermeasure is to use a user-entered passphrase to encrypt the 2576 private key, although users tend to choose weak passphrases and often 2577 reuse them for different purposes, possibly allowing an attack 2578 against DKIM to be extended into other domains. Nevertheless, the 2579 decoded private key might be briefly available to compromise by 2580 malware when it is entered, or might be discovered via keystroke 2581 logging. The added complexity of entering a passphrase each time one 2582 sends a message would also tend to discourage the use of a secure 2583 passphrase. 2585 A somewhat more effective countermeasure is to send messages through 2586 an outgoing MTA that can authenticate the submitter using existing 2587 techniques (e.g., SMTP Authentication), possibly validate the message 2588 itself (e.g., verify that the header is legitimate and that the 2589 content passes a spam content check), and sign the message using a 2590 key appropriate for the submitter address. Such an MTA can also 2591 apply controls on the volume of outgoing mail each user is permitted 2592 to originate in order to further limit the ability of malware to 2593 generate bulk email. 2595 8.3. Key Server Denial-of-Service Attacks 2597 Since the key servers are distributed (potentially separate for each 2598 domain), the number of servers that would need to be attacked to 2599 defeat this mechanism on an Internet-wide basis is very large. 2600 Nevertheless, key servers for individual domains could be attacked, 2601 impeding the verification of messages from that domain. This is not 2602 significantly different from the ability of an attacker to deny 2603 service to the mail exchangers for a given domain, although it 2604 affects outgoing, not incoming, mail. 2606 A variation on this attack is that if a very large amount of mail 2607 were to be sent using spoofed addresses from a given domain, the key 2608 servers for that domain could be overwhelmed with requests. However, 2609 given the low overhead of verification compared with handling of the 2610 email message itself, such an attack would be difficult to mount. 2612 8.4. Attacks Against the DNS 2614 Since the DNS is a required binding for key services, specific 2615 attacks against the DNS must be considered. 2617 While the DNS is currently insecure [RFC3833], these security 2618 problems are the motivation behind DNS Security (DNSSEC) [RFC4033], 2619 and all users of the DNS will reap the benefit of that work. 2621 DKIM is only intended as a "sufficient" method of proving 2622 authenticity. It is not intended to provide strong cryptographic 2623 proof about authorship or contents. Other technologies such as 2624 OpenPGP [RFC4880] and S/MIME [RFC5751] address those requirements. 2626 A second security issue related to the DNS revolves around the 2627 increased DNS traffic as a consequence of fetching selector-based 2628 data as well as fetching signing domain policy. Widespread 2629 deployment of DKIM will result in a significant increase in DNS 2630 queries to the claimed signing domain. In the case of forgeries on a 2631 large scale, DNS servers could see a substantial increase in queries. 2633 A specific DNS security issue that should be considered by DKIM 2634 verifiers is the name chaining attack described in Section 2.3 of 2635 [RFC3833]. A DKIM verifier, while verifying a DKIM-Signature header 2636 field, could be prompted to retrieve a key record of an attacker's 2637 choosing. This threat can be minimized by ensuring that name 2638 servers, including recursive name servers, used by the verifier 2639 enforce strict checking of "glue" and other additional information in 2640 DNS responses and are therefore not vulnerable to this attack. 2642 8.5. Replay Attacks 2644 In this attack, a spammer sends a message to be spammed to an 2645 accomplice, which results in the message being signed by the 2646 originating MTA. The accomplice resends the message, including the 2647 original signature, to a large number of recipients, possibly by 2648 sending the message to many compromised machines that act as MTAs. 2649 The messages, not having been modified by the accomplice, have valid 2650 signatures. 2652 Partial solutions to this problem involve the use of reputation 2653 services to convey the fact that the specific email address is being 2654 used for spam and that messages from that signer are likely to be 2655 spam. This requires a real-time detection mechanism in order to 2656 react quickly enough. However, such measures might be prone to 2657 abuse, if for example an attacker resent a large number of messages 2658 received from a victim in order to make them appear to be a spammer. 2660 Large verifiers might be able to detect unusually large volumes of 2661 mails with the same signature in a short time period. Smaller 2662 verifiers can get substantially the same volume of information via 2663 existing collaborative systems. 2665 8.6. Limits on Revoking Keys 2667 When a large domain detects undesirable behavior on the part of one 2668 of its users, it might wish to revoke the key used to sign that 2669 user's messages in order to disavow responsibility for messages that 2670 have not yet been verified or that are the subject of a replay 2671 attack. However, the ability of the domain to do so can be limited 2672 if the same key, for scalability reasons, is used to sign messages 2673 for many other users. Mechanisms for explicitly revoking keys on a 2674 per-address basis have been proposed but require further study as to 2675 their utility and the DNS load they represent. 2677 8.7. Intentionally Malformed Key Records 2679 It is possible for an attacker to publish key records in DNS that are 2680 intentionally malformed, with the intent of causing a denial-of- 2681 service attack on a non-robust verifier implementation. The attacker 2682 could then cause a verifier to read the malformed key record by 2683 sending a message to one of its users referencing the malformed 2684 record in a (not necessarily valid) signature. Verifiers MUST 2685 thoroughly verify all key records retrieved from the DNS and be 2686 robust against intentionally as well as unintentionally malformed key 2687 records. 2689 8.8. Intentionally Malformed DKIM-Signature Header Fields 2691 Verifiers MUST be prepared to receive messages with malformed DKIM- 2692 Signature header fields, and thoroughly verify the header field 2693 before depending on any of its contents. 2695 8.9. Information Leakage 2697 An attacker could determine when a particular signature was verified 2698 by using a per-message selector and then monitoring their DNS traffic 2699 for the key lookup. This would act as the equivalent of a "web bug" 2700 for verification time rather than when the message was read. 2702 8.10. Remote Timing Attacks 2704 In some cases it may be possible to extract private keys using a 2705 remote timing attack [BONEH03]. Implementations should consider 2706 obfuscating the timing to prevent such attacks. 2708 8.11. Reordered Header Fields 2710 Existing standards allow intermediate MTAs to reorder header fields. 2711 If a signer signs two or more header fields of the same name, this 2712 can cause spurious verification errors on otherwise legitimate 2713 messages. In particular, signers that sign any existing DKIM- 2714 Signature fields run the risk of having messages incorrectly fail to 2715 verify. 2717 8.12. RSA Attacks 2719 An attacker could create a large RSA signing key with a small 2720 exponent, thus requiring that the verification key have a large 2721 exponent. This will force verifiers to use considerable computing 2722 resources to verify the signature. Verifiers might avoid this attack 2723 by refusing to verify signatures that reference selectors with public 2724 keys having unreasonable exponents. 2726 In general, an attacker might try to overwhelm a verifier by flooding 2727 it with messages requiring verification. This is similar to other 2728 MTA denial-of-service attacks and should be dealt with in a similar 2729 fashion. 2731 8.13. Inappropriate Signing by Parent Domains 2733 The trust relationship described in Section 3.10 could conceivably be 2734 used by a parent domain to sign messages with identities in a 2735 subdomain not administratively related to the parent. For example, 2736 the ".com" registry could create messages with signatures using an 2737 "i=" value in the example.com domain. There is no general solution 2738 to this problem, since the administrative cut could occur anywhere in 2739 the domain name. For example, in the domain "example.podunk.ca.us" 2740 there are three administrative cuts (podunk.ca.us, ca.us, and us), 2741 any of which could create messages with an identity in the full 2742 domain. 2744 INFORMATIVE NOTE: This is considered an acceptable risk for the 2745 same reason that it is acceptable for domain delegation. For 2746 example, in the example above any of the domains could potentially 2747 simply delegate "example.podunk.ca.us" to a server of their choice 2748 and completely replace all DNS-served information. Note that a 2749 verifier MAY ignore signatures that come from an unlikely domain 2750 such as ".com", as discussed in Section 6.1.1. 2752 8.14. Attacks Involving Addition of Header Fields 2754 Many email implementations do not enforce [RFC5322] with strictness. 2755 As discussed in Section 5.3 DKIM processing is predicated on a valid 2756 mail message as its input. However, DKIM implementers should be 2757 aware of the potential effect of having loose enforcement by email 2758 components interacting with DKIM modules. 2760 For example, a message with multiple From: header fields violates 2761 Section 3.6 of [RFC5322]. With the intent of providing a better user 2762 experience, many agents tolerate these violations and deliver the 2763 message anyway. An MUA then might elect to render to the user the 2764 value of the first, or "top", From: field. This may also be done 2765 simply out of the expectation that there is only one, where a "find 2766 first" algorithm would have the same result. Such code in an MUA can 2767 be exploited to fool the user if it is also known that the other 2768 From: field is the one checked by arriving message filters. Such is 2769 the case with DKIM; although the From: field must be signed, a 2770 malformed message bearing more than one From: field might only have 2771 the first ("bottom") one signed, in an attempt to show the message 2772 with some "DKIM passed" annotation while also rendering the From: 2773 field that was not authenticated. (This can also be taken as a 2774 demonstration that DKIM is not designed to support author 2775 validation.) 2777 Note that the technique for using "h=...:from:from:...", described in 2778 Section 8.15 below, is of no effect in the case of an attacker that 2779 is also the signer. 2781 The From: field is used above to illustrate this issue, but it is 2782 only one of several fields that Section 3.6 of [RFC5322] constrains 2783 in this way. In reality any agent that forgives such malformations, 2784 or is careless about identifying which parts of a message were 2785 authenticated, is open to exploitation. 2787 8.15. Malformed Inputs 2789 DKIM allows additional header fields to be added to a signed message 2790 without breaking the signature. This tolerance can be abused, for 2791 example in a replay attack or a man-in-the-middle attack. The attack 2792 is accomplished by creating additional instances of header fields to 2793 an already signed message, without breaking the signature. These 2794 then might be displayed to the end user or are used as filtering 2795 input. Salient fields could include From: and Subject:, 2797 The resulting message violates section 3.6 of [RFC5322]. The way 2798 such input will be handled and displayed by an MUA is unpredictable, 2799 but it will commonly display the newly added header fields rather 2800 than those that are part of the originally signed message alongside 2801 some "valid DKIM signature" annotation. This might allow an attacker 2802 to replay a previously sent, signed message with a different 2803 Subject:, From: or To: field. 2805 However, [RFC5322] also tolerates obsolete message syntax, which does 2806 allow things like multiple From: fields on messages. The 2807 implementation of DKIM thus potentially creates a more stringent 2808 layer of expectation regarding the meaning of an identity, while that 2809 additional meaning is either obscured from or incorrectly presented 2810 to an end user in this context. 2812 Implementers need to consider this possibility when designing their 2813 input handling functions. Outright rejection of messages that 2814 violate the relevant standards such as [RFC5322], [RFC2045], etc. 2815 will interfere with delivery of legacy formats. Instead, given such 2816 input, a signing module could return an error rather than generate a 2817 signature; a verifying module might return a syntax error code or 2818 arrange not to return a positive result even if the signature 2819 technically validates. 2821 Senders concerned that their messages might be particularly 2822 vulnerable to this sort of attack and who do not wish to rely on 2823 receiver filtering of invalid messages can ensure that adding 2824 additional header fields will break the DKIM signature by including 2825 two copies of the header fields about which they are concerned in the 2826 signature (e.g. "h= ... from:from:to:to:subject:subject ..."). See 2827 Sections 3.5 and 5.4 for further discussion of this mechanism. 2829 Specific validity rules for all known header fields can be gleaned 2830 from the IANA "Permanent Header Field Registry" and the reference 2831 documents it identifies. 2833 9. References 2835 9.1. Normative References 2837 [FIPS-180-2-2002] 2838 U.S. Department of Commerce, "Secure Hash Standard", FIPS 2839 PUB 180-2, August 2002. 2841 [ITU-X660-1997] 2842 "Information Technology - ASN.1 encoding rules: 2843 Specification of Basic Encoding Rules (BER), Canonical 2844 Encoding Rules (CER) and Distinguished Encoding Rules 2845 (DER)", 1997. 2847 [RFC1034] Mockapetris, P., "DOMAIN NAMES - CONCEPTS AND FACILITIES", 2848 RFC 1034, November 1987. 2850 [RFC2045] Freed, N. and N. Borenstein, "Multipurpose Internet Mail 2851 Extensions (MIME) Part One: Format of Internet Message 2852 Bodies", RFC 2045, November 1996. 2854 [RFC2049] Freed, N. and N. Borenstein, "Multipurpose Internet Mail 2855 Extensions (MIME) Part Five: Conformance Criteria and 2856 Examples", RFC 2049, November 1996. 2858 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2859 Requirement Levels", BCP 14, RFC 2119, March 1997. 2861 [RFC3447] Jonsson, J. and B. Kaliski, "Public-Key Cryptography 2862 Standards (PKCS) #1: RSA Cryptography Specifications 2863 Version 2.1", RFC 3447, February 2003. 2865 [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax 2866 Specifications: ABNF", RFC 5234, January 2008. 2868 [RFC5321] Klensin, J., "Simple Mail Transfer Protocol", RFC 5321, 2869 October 2008. 2871 [RFC5322] Resnick, P., "Internet Message Format", RFC 5322, 2872 October 2008. 2874 [RFC5598] Crocker, D., "Internet Mail Architecture", RFC 5598, 2875 July 2009. 2877 [RFC5890] Klensin, J., "Internationalizing Domain Names in 2878 Applications (IDNA): Definitions and Document Framework", 2879 RFC 5890, August 2010. 2881 9.2. Informative References 2883 [BONEH03] "Remote Timing Attacks are Practical", Proceedings 12th 2884 USENIX Security Symposium, 2003. 2886 [RFC1847] Galvin, J., Murphy, S., Crocker, S., and N. Freed, 2887 "Security Multiparts for MIME: Multipart/Signed and 2888 Multipart/Encrypted", RFC 1847, October 1995. 2890 [RFC3766] Orman, H. and P. Hoffman, "Determining Strengths For 2891 Public Keys Used For Exchanging Symmetric Keys", BCP 86, 2892 RFC 3766, April 2004. 2894 [RFC3833] Atkins, D. and R. Austein, "Threat Analysis of the Domain 2895 Name System (DNS)", RFC 3833, August 2004. 2897 [RFC3864] Klyne, G., Nottingham, M., and J. Mogul, "Registration 2898 Procedures for Message Header Fields", BCP 90, RFC 3864, 2899 September 2004. 2901 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 2902 Rose, "DNS Security Introduction and Requirements", 2903 RFC 4033, March 2005. 2905 [RFC4409] Gellens, R. and J. Klensin, "Message Submission for Mail", 2906 RFC 4409, April 2006. 2908 [RFC4686] Fenton, J., "Analysis of Threats Motivating DomainKeys 2909 Identified Mail (DKIM)", RFC 4686, September 2006. 2911 [RFC4870] Delany, M., "Domain-Based Email Authentication Using 2912 Public Keys Advertised in the DNS (DomainKeys)", RFC 4870, 2913 May 2007. 2915 [RFC4871] Allman, E., Callas, J., Delany, M., Libbey, M., Fenton, 2916 J., and M. Thomas, "DomainKeys Identified Mail (DKIM) 2917 Signatures", RFC 4871, May 2007. 2919 [RFC4880] Callas, J., Donnerhacke, L., Finney, H., and R. Thayer, 2920 "OpenPGP Message Format", RFC 4880, November 2007. 2922 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2923 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2924 May 2008. 2926 [RFC5451] Kucherawy, M., "Message Header Field for Indicating 2927 Message Authentication Status", RFC 5451, April 2009. 2929 [RFC5585] Hansen, T., Crocker, D., and P. Hallam-Baker, "DomainKeys 2930 Identified Mail (DKIM) Service Overview", RFC 5585, 2931 July 2009. 2933 [RFC5751] Ramsdell, B., "Secure/Multipurpose Internet Mail 2934 Extensions (S/MIME) Version 3.1 Message Specification", 2935 RFC 5751, January 2010. 2937 [RFC5863] Hansen, T., Siegel, E., Hallam-Baker, P., and D. Crocker, 2938 "DomainKeys Identified Mail (DKIM) Development, 2939 Deployment, and Operations", RFC 5863, May 2010. 2941 Appendix A. Example of Use (INFORMATIVE) 2943 This section shows the complete flow of an email from submission to 2944 final delivery, demonstrating how the various components fit 2945 together. The key used in this example is shown in Appendix C. 2947 A.1. The User Composes an Email 2948 From: Joe SixPack 2949 To: Suzie Q 2950 Subject: Is dinner ready? 2951 Date: Fri, 11 Jul 2003 21:00:37 -0700 (PDT) 2952 Message-ID: <;20030712040037.46341.5F8J@football.example.com> 2954 Hi. 2956 We lost the game. Are you hungry yet? 2958 Joe. 2960 Figure 1: The User Composes an Email 2962 A.2. The Email is Signed 2964 This email is signed by the example.com outbound email server and now 2965 looks like this: 2966 DKIM-Signature: v=1; a=rsa-sha256; s=brisbane; d=example.com; 2967 c=simple/simple; q=dns/txt; i=joe@football.example.com; 2968 h=Received : From : To : Subject : Date : Message-ID; 2969 bh=2jUSOH9NhtVGCQWNr9BrIAPreKQjO6Sn7XIkfJVOzv8=; 2970 b=AuUoFEfDxTDkHlLXSZEpZj79LICEps6eda7W3deTVFOk4yAUoqOB 2971 4nujc7YopdG5dWLSdNg6xNAZpOPr+kHxt1IrE+NahM6L/LbvaHut 2972 KVdkLLkpVaVVQPzeRDI009SO2Il5Lu7rDNH6mZckBdrIx0orEtZV 2973 4bmp/YzhwvcubU4=; 2974 Received: from client1.football.example.com [192.0.2.1] 2975 by submitserver.example.com with SUBMISSION; 2976 Fri, 11 Jul 2003 21:01:54 -0700 (PDT) 2977 From: Joe SixPack 2978 To: Suzie Q 2979 Subject: Is dinner ready? 2980 Date: Fri, 11 Jul 2003 21:00:37 -0700 (PDT) 2981 Message-ID: <20030712040037.46341.5F8J@football.example.com> 2983 Hi. 2985 We lost the game. Are you hungry yet? 2987 Joe. 2989 The Email is Signed 2991 The signing email server requires access to the private key 2992 associated with the "brisbane" selector to generate this signature. 2994 A.3. The Email Signature is Verified 2996 The signature is normally verified by an inbound SMTP server or 2997 possibly the final delivery agent. However, intervening MTAs can 2998 also perform this verification if they choose to do so. The 2999 verification process uses the domain "example.com" extracted from the 3000 "d=" tag and the selector "brisbane" from the "s=" tag in the DKIM- 3001 Signature header field to form the DNS DKIM query for: 3002 brisbane._domainkey.example.com 3004 Signature verification starts with the physically last Received 3005 header field, the From header field, and so forth, in the order 3006 listed in the "h=" tag. Verification follows with a single CRLF 3007 followed by the body (starting with "Hi."). The email is canonically 3008 prepared for verifying with the "simple" method. The result of the 3009 query and subsequent verification of the signature is stored (in this 3010 example) in the X-Authentication-Results header field line. After 3011 successful verification, the email looks like this: 3012 X-Authentication-Results: shopping.example.net 3013 header.from=joe@football.example.com; dkim=pass 3014 Received: from mout23.football.example.com (192.168.1.1) 3015 by shopping.example.net with SMTP; 3016 Fri, 11 Jul 2003 21:01:59 -0700 (PDT) 3017 DKIM-Signature: v=1; a=rsa-sha256; s=brisbane; d=example.com; 3018 c=simple/simple; q=dns/txt; i=joe@football.example.com; 3019 h=Received : From : To : Subject : Date : Message-ID; 3020 bh=2jUSOH9NhtVGCQWNr9BrIAPreKQjO6Sn7XIkfJVOzv8=; 3021 b=AuUoFEfDxTDkHlLXSZEpZj79LICEps6eda7W3deTVFOk4yAUoqOB 3022 4nujc7YopdG5dWLSdNg6xNAZpOPr+kHxt1IrE+NahM6L/LbvaHut 3023 KVdkLLkpVaVVQPzeRDI009SO2Il5Lu7rDNH6mZckBdrIx0orEtZV 3024 4bmp/YzhwvcubU4=; 3025 Received: from client1.football.example.com [192.0.2.1] 3026 by submitserver.example.com with SUBMISSION; 3027 Fri, 11 Jul 2003 21:01:54 -0700 (PDT) 3028 From: Joe SixPack 3029 To: Suzie Q 3030 Subject: Is dinner ready? 3031 Date: Fri, 11 Jul 2003 21:00:37 -0700 (PDT) 3032 Message-ID: <20030712040037.46341.5F8J@football.example.com> 3034 Hi. 3036 We lost the game. Are you hungry yet? 3038 Joe. 3040 Successful verification 3042 Appendix B. Usage Examples (INFORMATIVE) 3044 DKIM signing and validating can be used in different ways, for 3045 different operational scenarios. This Appendix discusses some common 3046 examples. 3048 NOTE: Descriptions in this Appendix are for informational purposes 3049 only. They describe various ways that DKIM can be used, given 3050 particular constraints and needs. In no case are these examples 3051 intended to be taken as providing explanation or guidance 3052 concerning DKIM specification details, when creating an 3053 implementation. 3055 B.1. Alternate Submission Scenarios 3057 In the most simple scenario, a user's MUA, MSA, and Internet 3058 (boundary) MTA are all within the same administrative environment, 3059 using the same domain name. Therefore, all of the components 3060 involved in submission and initial transfer are related. However, it 3061 is common for two or more of the components to be under independent 3062 administrative control. This creates challenges for choosing and 3063 administering the domain name to use for signing, and for its 3064 relationship to common email identity header fields. 3066 B.1.1. Delegated Business Functions 3068 Some organizations assign specific business functions to discrete 3069 groups, inside or outside the organization. The goal, then, is to 3070 authorize that group to sign some mail, but to constrain what 3071 signatures they can generate. DKIM selectors (the "s=" signature 3072 tag) facilitate this kind of restricted authorization. Examples of 3073 these outsourced business functions are legitimate email marketing 3074 providers and corporate benefits providers. 3076 Here, the delegated group needs to be able to send messages that are 3077 signed, using the email domain of the client company. At the same 3078 time, the client often is reluctant to register a key for the 3079 provider that grants the ability to send messages for arbitrary 3080 addresses in the domain. 3082 There are multiple ways to administer these usage scenarios. In one 3083 case, the client organization provides all of the public query 3084 service (for example, DNS) administration, and in another it uses DNS 3085 delegation to enable all ongoing administration of the DKIM key 3086 record by the delegated group. 3088 If the client organization retains responsibility for all of the DNS 3089 administration, the outsourcing company can generate a key pair, 3090 supplying the public key to the client company, which then registers 3091 it in the query service, using a unique selector. The client company 3092 retains control over the use of the delegated key because it retains 3093 the ability to revoke the key at any time. 3095 If the client wants the delegated group to do the DNS administration, 3096 it can have the domain name that is specified with the selector point 3097 to the provider's DNS server. The provider then creates and 3098 maintains all of the DKIM signature information for that selector. 3099 Hence, the client cannot provide constraints on the Local-part of 3100 addresses that get signed, but it can revoke the provider's signing 3101 rights by removing the DNS delegation record. 3103 B.1.2. PDAs and Similar Devices 3105 PDAs demonstrate the need for using multiple keys per domain. 3106 Suppose that John Doe wanted to be able to send messages using his 3107 corporate email address, jdoe@example.com, and his email device did 3108 not have the ability to make a Virtual Private Network (VPN) 3109 connection to the corporate network, either because the device is 3110 limited or because there are restrictions enforced by his Internet 3111 access provider. If the device was equipped with a private key 3112 registered for jdoe@example.com by the administrator of the 3113 example.com domain, and appropriate software to sign messages, John 3114 could sign the message on the device itself before transmission 3115 through the outgoing network of the access service provider. 3117 B.1.3. Roaming Users 3119 Roaming users often find themselves in circumstances where it is 3120 convenient or necessary to use an SMTP server other than their home 3121 server; examples are conferences and many hotels. In such 3122 circumstances, a signature that is added by the submission service 3123 will use an identity that is different from the user's home system. 3125 Ideally, roaming users would connect back to their home server using 3126 either a VPN or a SUBMISSION server running with SMTP AUTHentication 3127 on port 587. If the signing can be performed on the roaming user's 3128 laptop, then they can sign before submission, although the risk of 3129 further modification is high. If neither of these are possible, 3130 these roaming users will not be able to send mail signed using their 3131 own domain key. 3133 B.1.4. Independent (Kiosk) Message Submission 3135 Stand-alone services, such as walk-up kiosks and web-based 3136 information services, have no enduring email service relationship 3137 with the user, but users occasionally request that mail be sent on 3138 their behalf. For example, a website providing news often allows the 3139 reader to forward a copy of the article to a friend. This is 3140 typically done using the reader's own email address, to indicate who 3141 the author is. This is sometimes referred to as the "Evite problem", 3142 named after the website of the same name that allows a user to send 3143 invitations to friends. 3145 A common way this is handled is to continue to put the reader's email 3146 address in the From header field of the message, but put an address 3147 owned by the email posting site into the Sender header field. The 3148 posting site can then sign the message, using the domain that is in 3149 the Sender field. This provides useful information to the receiving 3150 email site, which is able to correlate the signing domain with the 3151 initial submission email role. 3153 Receiving sites often wish to provide their end users with 3154 information about mail that is mediated in this fashion. Although 3155 the real efficacy of different approaches is a subject for human 3156 factors usability research, one technique that is used is for the 3157 verifying system to rewrite the From header field, to indicate the 3158 address that was verified. For example: From: John Doe via 3159 news@news-site.com . (Note that such rewriting 3160 will break a signature, unless it is done after the verification pass 3161 is complete.) 3163 B.2. Alternate Delivery Scenarios 3165 Email is often received at a mailbox that has an address different 3166 from the one used during initial submission. In these cases, an 3167 intermediary mechanism operates at the address originally used and it 3168 then passes the message on to the final destination. This mediation 3169 process presents some challenges for DKIM signatures. 3171 B.2.1. Affinity Addresses 3173 "Affinity addresses" allow a user to have an email address that 3174 remains stable, even as the user moves among different email 3175 providers. They are typically associated with college alumni 3176 associations, professional organizations, and recreational 3177 organizations with which they expect to have a long-term 3178 relationship. These domains usually provide forwarding of incoming 3179 email, and they often have an associated Web application that 3180 authenticates the user and allows the forwarding address to be 3181 changed. However, these services usually depend on users sending 3182 outgoing messages through their own service providers' MTAs. Hence, 3183 mail that is signed with the domain of the affinity address is not 3184 signed by an entity that is administered by the organization owning 3185 that domain. 3187 With DKIM, affinity domains could use the Web application to allow 3188 users to register per-user keys to be used to sign messages on behalf 3189 of their affinity address. The user would take away the secret half 3190 of the key pair for signing, and the affinity domain would publish 3191 the public half in DNS for access by verifiers. 3193 This is another application that takes advantage of user-level 3194 keying, and domains used for affinity addresses would typically have 3195 a very large number of user-level keys. Alternatively, the affinity 3196 domain could handle outgoing mail, operating a mail submission agent 3197 that authenticates users before accepting and signing messages for 3198 them. This is of course dependent on the user's service provider not 3199 blocking the relevant TCP ports used for mail submission. 3201 B.2.2. Simple Address Aliasing (.forward) 3203 In some cases a recipient is allowed to configure an email address to 3204 cause automatic redirection of email messages from the original 3205 address to another, such as through the use of a Unix .forward file. 3206 In this case, messages are typically redirected by the mail handling 3207 service of the recipient's domain, without modification, except for 3208 the addition of a Received header field to the message and a change 3209 in the envelope recipient address. In this case, the recipient at 3210 the final address' mailbox is likely to be able to verify the 3211 original signature since the signed content has not changed, and DKIM 3212 is able to validate the message signature. 3214 B.2.3. Mailing Lists and Re-Posters 3216 There is a wide range of behaviors in services that take delivery of 3217 a message and then resubmit it. A primary example is with mailing 3218 lists (collectively called "forwarders" below), ranging from those 3219 that make no modification to the message itself, other than to add a 3220 Received header field and change the envelope information, to those 3221 that add header fields, change the Subject header field, add content 3222 to the body (typically at the end), or reformat the body in some 3223 manner. The simple ones produce messages that are quite similar to 3224 the automated alias services. More elaborate systems essentially 3225 create a new message. 3227 A Forwarder that does not modify the body or signed header fields of 3228 a message is likely to maintain the validity of the existing 3229 signature. It also could choose to add its own signature to the 3230 message. 3232 Forwarders which modify a message in a way that could make an 3233 existing signature invalid are particularly good candidates for 3234 adding their own signatures (e.g., mailing-list-name@example.net). 3236 Since (re-)signing is taking responsibility for the content of the 3237 message, these signing forwarders are likely to be selective, and 3238 forward or re-sign a message only if it is received with a valid 3239 signature or if they have some other basis for knowing that the 3240 message is not spoofed. 3242 A common practice among systems that are primarily redistributors of 3243 mail is to add a Sender header field to the message, to identify the 3244 address being used to sign the message. This practice will remove 3245 any preexisting Sender header field as required by [RFC5322]. The 3246 forwarder applies a new DKIM-Signature header field with the 3247 signature, public key, and related information of the forwarder. 3249 Appendix C. Creating a Public Key (INFORMATIVE) 3251 The default signature is an RSA signed SHA256 digest of the complete 3252 email. For ease of explanation, the openssl command is used to 3253 describe the mechanism by which keys and signatures are managed. One 3254 way to generate a 1024-bit, unencrypted private key suitable for DKIM 3255 is to use openssl like this: 3256 $ openssl genrsa -out rsa.private 1024 3257 For increased security, the "-passin" parameter can also be added to 3258 encrypt the private key. Use of this parameter will require entering 3259 a password for several of the following steps. Servers may prefer to 3260 use hardware cryptographic support. 3262 The "genrsa" step results in the file rsa.private containing the key 3263 information similar to this: 3264 -----BEGIN RSA PRIVATE KEY----- 3265 MIICXwIBAAKBgQDwIRP/UC3SBsEmGqZ9ZJW3/DkMoGeLnQg1fWn7/zYtIxN2SnFC 3266 jxOCKG9v3b4jYfcTNh5ijSsq631uBItLa7od+v/RtdC2UzJ1lWT947qR+Rcac2gb 3267 to/NMqJ0fzfVjH4OuKhitdY9tf6mcwGjaNBcWToIMmPSPDdQPNUYckcQ2QIDAQAB 3268 AoGBALmn+XwWk7akvkUlqb+dOxyLB9i5VBVfje89Teolwc9YJT36BGN/l4e0l6QX 3269 /1//6DWUTB3KI6wFcm7TWJcxbS0tcKZX7FsJvUz1SbQnkS54DJck1EZO/BLa5ckJ 3270 gAYIaqlA9C0ZwM6i58lLlPadX/rtHb7pWzeNcZHjKrjM461ZAkEA+itss2nRlmyO 3271 n1/5yDyCluST4dQfO8kAB3toSEVc7DeFeDhnC1mZdjASZNvdHS4gbLIA1hUGEF9m 3272 3hKsGUMMPwJBAPW5v/U+AWTADFCS22t72NUurgzeAbzb1HWMqO4y4+9Hpjk5wvL/ 3273 eVYizyuce3/fGke7aRYw/ADKygMJdW8H/OcCQQDz5OQb4j2QDpPZc0Nc4QlbvMsj 3274 7p7otWRO5xRa6SzXqqV3+F0VpqvDmshEBkoCydaYwc2o6WQ5EBmExeV8124XAkEA 3275 qZzGsIxVP+sEVRWZmW6KNFSdVUpk3qzK0Tz/WjQMe5z0UunY9Ax9/4PVhp/j61bf 3276 eAYXunajbBSOLlx4D+TunwJBANkPI5S9iylsbLs6NkaMHV6k5ioHBBmgCak95JGX 3277 GMot/L2x0IYyMLAz6oLWh2hm7zwtb0CgOrPo1ke44hFYnfc= 3278 -----END RSA PRIVATE KEY----- 3279 To extract the public-key component from the private key, use openssl 3280 like this: 3281 $ openssl rsa -in rsa.private -out rsa.public -pubout -outform PEM 3283 This results in the file rsa.public containing the key information 3284 similar to this: 3285 -----BEGIN PUBLIC KEY----- 3286 MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQDwIRP/UC3SBsEmGqZ9ZJW3/DkM 3287 oGeLnQg1fWn7/zYtIxN2SnFCjxOCKG9v3b4jYfcTNh5ijSsq631uBItLa7od+v/R 3288 tdC2UzJ1lWT947qR+Rcac2gbto/NMqJ0fzfVjH4OuKhitdY9tf6mcwGjaNBcWToI 3289 MmPSPDdQPNUYckcQ2QIDAQAB 3290 -----END PUBLIC KEY----- 3292 This public-key data (without the BEGIN and END tags) is placed in 3293 the DNS: 3294 $ORIGIN _domainkey.example.org. 3295 brisbane IN TXT ("v=DKIM1; p=MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQ" 3296 "KBgQDwIRP/UC3SBsEmGqZ9ZJW3/DkMoGeLnQg1fWn7/zYt" 3297 "IxN2SnFCjxOCKG9v3b4jYfcTNh5ijSsq631uBItLa7od+v" 3298 "/RtdC2UzJ1lWT947qR+Rcac2gbto/NMqJ0fzfVjH4OuKhi" 3299 "tdY9tf6mcwGjaNBcWToIMmPSPDdQPNUYckcQ2QIDAQAB") 3301 C.1. Compatibility with DomainKeys Key Records 3303 DKIM key records were designed to be backwards-compatible in many 3304 cases with key records used by DomainKeys [RFC4870] (sometimes 3305 referred to as "selector records" in the DomainKeys context). One 3306 area of incompatibility warrants particular attention. The "g=" tag/ 3307 value may be used in DomainKeys and [RFC4871] key records to provide 3308 finer granularity of the validity of the key record to a specific 3309 local-part. A null "g=" value in DomainKeys is valid for all 3310 addresses in the domain. This differs from the usage in the original 3311 DKIM specification, where a null "g=" value is not valid for any 3312 address. In particular, the example public key record in Section 3313 3.2.3 of [RFC4870] with DKIM. 3315 Although the "g=" tag has been deprecated in this version of the DKIM 3316 specification, signers are advised not to include the "g=" tag in key 3317 records because some [RFC4871]-compliant verifiers will be in use for 3318 a considerable period to come. 3320 Appendix D. MUA Considerations 3322 When a DKIM signature is verified, the processing system sometimes 3323 makes the result available to the recipient user's MUA. How to 3324 present this information to the user in a way that helps them is a 3325 matter of continuing human factors usability research. The tendency 3326 is to have the MUA highlight the SDID, in an attempt to show the user 3327 the identity that is claiming responsibility for the message. An MUA 3328 might do this with visual cues such as graphics, or it might include 3329 the address in an alternate view, or it might even rewrite the 3330 original From address using the verified information. Some MUAs 3331 might indicate which header fields were protected by the validated 3332 DKIM signature. This could be done with a positive indication on the 3333 signed header fields, with a negative indication on the unsigned 3334 header fields, by visually hiding the unsigned header fields, or some 3335 combination of these. If an MUA uses visual indications for signed 3336 header fields, the MUA probably needs to be careful not to display 3337 unsigned header fields in a way that might be construed by the end 3338 user as having been signed. If the message has an l= tag whose value 3339 does not extend to the end of the message, the MUA might also hide or 3340 mark the portion of the message body that was not signed. 3342 The aforementioned information is not intended to be exhaustive. The 3343 MUA may choose to highlight, accentuate, hide, or otherwise display 3344 any other information that may, in the opinion of the MUA author, be 3345 deemed important to the end user. 3347 Appendix E. Acknowledgements 3349 The previous IETF version of DKIM [RFC4871] was edited by: Eric 3350 Allman, Jon Callas, Mark Delany, Miles Libbey, Jim Fenton and Michael 3351 Thomas. 3353 That specification was the result of an extended, collaborative 3354 effort, including participation by: Russ Allbery, Edwin Aoki, Claus 3355 Assmann, Steve Atkins, Rob Austein, Fred Baker, Mark Baugher, Steve 3356 Bellovin, Nathaniel Borenstein, Dave Crocker, Michael Cudahy, Dennis 3357 Dayman, Jutta Degener, Frank Ellermann, Patrik Faeltstroem, Mark 3358 Fanto, Stephen Farrell, Duncan Findlay, Elliot Gillum, Olafur 3359 Gu[eth]mundsson, Phillip Hallam-Baker, Tony Hansen, Sam Hartman, 3360 Arvel Hathcock, Amir Herzberg, Paul Hoffman, Russ Housley, Craig 3361 Hughes, Cullen Jennings, Don Johnsen, Harry Katz, Murray S. 3362 Kucherawy, Barry Leiba, John Levine, Charles Lindsey, Simon 3363 Longsdale, David Margrave, Justin Mason, David Mayne, Thierry Moreau, 3364 Steve Murphy, Russell Nelson, Dave Oran, Doug Otis, Shamim Pirzada, 3365 Juan Altmayer Pizzorno, Sanjay Pol, Blake Ramsdell, Christian Renaud, 3366 Scott Renfro, Neil Rerup, Eric Rescorla, Dave Rossetti, Hector 3367 Santos, Jim Schaad, the Spamhaus.org team, Malte S. Stretz, Robert 3368 Sanders, Rand Wacker, Sam Weiler, and Dan Wing. 3370 The earlier DomainKeys was a primary source from which DKIM was 3371 derived. Further information about DomainKeys is at [RFC4870]. 3373 This revision received contributions from: Steve Atkins, Mark Delany, 3374 J.D. Falk, Jim Fenton, Michael Hammer, Barry Leiba, John Levine, 3375 Charles Lindsey, Jeff Macdonald, Franck Martin, Brett McDowell, Doug 3376 Otis, Bill Oxley, Hector Santos, Rolf Sonneveld, Michael Thomas, and 3377 Alessandro Vesely. 3379 Authors' Addresses 3381 D. Crocker (editor) 3382 Brandenburg InternetWorking 3383 675 Spruce Dr. 3384 Sunnyvale 3385 USA 3387 Phone: +1.408.246.8253 3388 Email: dcrocker@bbiw.net 3389 URI: http://bbiw.net 3391 Tony Hansen (editor) 3392 AT&T Laboratories 3393 200 Laurel Ave. South 3394 Middletown, NJ 07748 3395 USA 3397 Email: tony+dkimov@maillennium.att.com 3399 M. Kucherawy (editor) 3400 Cloudmark 3401 128 King St., 2nd Floor 3402 San Francisco, CA 94107 3403 USA 3405 Email: msk@cloudmark.com