idnits 2.17.1 draft-ietf-dkim-deployment-07.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) 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 : ---------------------------------------------------------------------------- ** There are 4 instances of too long lines in the document, the longest one being 2 characters in excess of 72. ** The abstract seems to contain references ([RFC4871]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. ** The document seems to lack a both a reference to RFC 2119 and the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. RFC 2119 keyword, line 277: '...s to a responsible identity. DKIM MAY...' RFC 2119 keyword, line 279: '...ide DKIM verifier MUST communicate the...' RFC 2119 keyword, line 281: '... module and MAY communicate the U...' RFC 2119 keyword, line 291: '...ide DKIM verifier MUST communicate the...' RFC 2119 keyword, line 293: '... module and MAY communicate the U...' (59 more instances...) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'SHOULD not' in this paragraph: If key management hardware support is available, it SHOULD be used. If keys are stored in software, appropriate file control protections MUST be employed, and any location in which the private key is stored in plaintext form SHOULD be excluded from regular backup processes and SHOULD not be accessible through any form of network including private local area networks. Auditing software SHOULD be used periodically to verify that the permissions on the private key files remain secure. == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'SHOULD not' in this paragraph: In the most restrictive case a domain holder maintains full control over the creation of key records and employs appropriate key record restrictions to enforce restrictions on the messages for which the third party signer is able to sign. If such restrictions are impractical, the domain holder SHOULD delegate a DNS subzone for publishing key records to the third party signer. The domain holder SHOULD not allow a third party signer unrestricted access to its DNS service for the purpose of publishing key records. -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (July 11, 2009) is 5396 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC0989' is defined on line 1821, but no explicit reference was found in the text == Unused Reference: 'RFC1034' is defined on line 1825, but no explicit reference was found in the text == Unused Reference: 'RFC1848' is defined on line 1828, but no explicit reference was found in the text == Unused Reference: 'RFC1991' is defined on line 1831, but no explicit reference was found in the text == Unused Reference: 'RFC2440' is defined on line 1834, but no explicit reference was found in the text == Unused Reference: 'RFC3156' is defined on line 1837, but no explicit reference was found in the text == Unused Reference: 'RFC3164' is defined on line 1840, but no explicit reference was found in the text == Unused Reference: 'RFC3851' is defined on line 1843, but no explicit reference was found in the text == Unused Reference: 'RFC4686' is defined on line 1851, but no explicit reference was found in the text == Unused Reference: 'RFC4870' is defined on line 1854, but no explicit reference was found in the text == Unused Reference: 'RFC4880' is defined on line 1862, but no explicit reference was found in the text == Unused Reference: 'RFC5321' is defined on line 1869, but no explicit reference was found in the text -- Obsolete informational reference (is this intentional?): RFC 989 (Obsoleted by RFC 1040, RFC 1113) -- Obsolete informational reference (is this intentional?): RFC 1991 (Obsoleted by RFC 4880) -- Obsolete informational reference (is this intentional?): RFC 2440 (Obsoleted by RFC 4880) -- Obsolete informational reference (is this intentional?): RFC 3164 (Obsoleted by RFC 5424) -- Obsolete informational reference (is this intentional?): RFC 3851 (Obsoleted by RFC 5751) -- 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 5451 (Obsoleted by RFC 7001) Summary: 4 errors (**), 0 flaws (~~), 15 warnings (==), 10 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DomainKeys Identified Mail T. Hansen 3 Internet-Draft AT&T Laboratories 4 Intended status: Informational E. Siegel 5 Expires: January 12, 2010 6 P. Hallam-Baker 7 Default Deny Security, Inc. 8 D. Crocker 9 Brandenburg InternetWorking 10 July 11, 2009 12 DomainKeys Identified Mail (DKIM) Development, Deployment and Operations 13 draft-ietf-dkim-deployment-07 15 Status of this Memo 17 This Internet-Draft is submitted to IETF in full conformance with the 18 provisions of BCP 78 and BCP 79. 20 Internet-Drafts are working documents of the Internet Engineering 21 Task Force (IETF), its areas, and its working groups. Note that 22 other groups may also distribute working documents as Internet- 23 Drafts. 25 Internet-Drafts are draft documents valid for a maximum of six months 26 and may be updated, replaced, or obsoleted by other documents at any 27 time. It is inappropriate to use Internet-Drafts as reference 28 material or to cite them other than as "work in progress." 30 The list of current Internet-Drafts can be accessed at 31 http://www.ietf.org/ietf/1id-abstracts.txt. 33 The list of Internet-Draft Shadow Directories can be accessed at 34 http://www.ietf.org/shadow.html. 36 This Internet-Draft will expire on January 12, 2010. 38 Copyright Notice 40 Copyright (c) 2009 IETF Trust and the persons identified as the 41 document authors. All rights reserved. 43 This document is subject to BCP 78 and the IETF Trust's Legal 44 Provisions Relating to IETF Documents in effect on the date of 45 publication of this document (http://trustee.ietf.org/license-info). 46 Please review these documents carefully, as they describe your rights 47 and restrictions with respect to this document. 49 Abstract 51 DomainKeys Identified Mail (DKIM) allows an organization to claim 52 responsibility for transmitting a message, in a way that can be 53 validated by a recipient. The organization can be the author's, the 54 originating sending site, an intermediary, or one of their agents. A 55 message can contain multiple signatures, from the same or different 56 organizations involved with the message. DKIM defines a domain-level 57 digital signature authentication framework for email, using public 58 key cryptography, using the domain name service as its key server 59 technology [RFC4871]. This permits verification of a responsible 60 organization, as well as the integrity of the message contents. DKIM 61 will also provide a mechanism that permits potential email signers to 62 publish information about their email signing practices; this will 63 permit email receivers to make additional assessments about messages. 64 DKIM's authentication of email identity can assist in the global 65 control of "spam" and "phishing". This document provides 66 implementation, deployment, operational and migration considerations 67 for DKIM. 69 Table of Contents 71 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 72 2. Using DKIM as Part of Trust Assessment . . . . . . . . . . . . 5 73 2.1. A Systems View of Email Trust Assessment . . . . . . . . . 5 74 2.2. Choosing a DKIM Tag for the Assessment Identifier . . . . 7 75 2.3. Choosing the Signing Domain Name . . . . . . . . . . . . . 9 76 2.4. Recipient-based Assessments . . . . . . . . . . . . . . . 11 77 2.5. Filtering . . . . . . . . . . . . . . . . . . . . . . . . 12 78 3. DKIM Key Generation, Storage, and Management . . . . . . . . . 14 79 3.1. Private Key Management: Deployment and Ongoing 80 Operations . . . . . . . . . . . . . . . . . . . . . . . . 15 81 3.2. Storing Public Keys: DNS Server Software Considerations . 16 82 3.3. Per User Signing Key Management Issues . . . . . . . . . . 17 83 3.4. Third Party Signer Key Management and Selector 84 Administration . . . . . . . . . . . . . . . . . . . . . . 17 85 3.5. Key Pair / Selector Lifecycle Management . . . . . . . . . 18 86 4. Signing . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 87 4.1. DNS Records . . . . . . . . . . . . . . . . . . . . . . . 20 88 4.2. Signing Module . . . . . . . . . . . . . . . . . . . . . . 20 89 4.3. Signing Policies and Practices . . . . . . . . . . . . . . 21 90 5. Verifying . . . . . . . . . . . . . . . . . . . . . . . . . . 21 91 5.1. Intended Scope of Use . . . . . . . . . . . . . . . . . . 21 92 5.2. Signature Scope . . . . . . . . . . . . . . . . . . . . . 22 93 5.3. Design Scope of Use . . . . . . . . . . . . . . . . . . . 22 94 5.4. Inbound Mail Filtering . . . . . . . . . . . . . . . . . . 23 95 5.5. Messages sent through Mailing Lists and other 96 Intermediaries . . . . . . . . . . . . . . . . . . . . . . 23 97 5.6. Generation, Transmission and Use of Results Headers . . . 24 98 6. Taxonomy of Signatures . . . . . . . . . . . . . . . . . . . . 24 99 6.1. Single Domain Signature . . . . . . . . . . . . . . . . . 25 100 6.2. Parent Domain Signature . . . . . . . . . . . . . . . . . 25 101 6.3. Third Party Signature . . . . . . . . . . . . . . . . . . 26 102 6.4. Using Trusted Third Party Senders . . . . . . . . . . . . 27 103 6.5. Multiple Signatures . . . . . . . . . . . . . . . . . . . 28 104 7. Example Usage Scenarios . . . . . . . . . . . . . . . . . . . 30 105 7.1. Author's Organization - Simple . . . . . . . . . . . . . . 30 106 7.2. Author's Organization - Differentiated Types of Mail . . . 30 107 7.3. Author Domain Signing Practices . . . . . . . . . . . . . 31 108 7.4. Delegated Signing . . . . . . . . . . . . . . . . . . . . 32 109 7.5. Independent Third Party Service Providers . . . . . . . . 33 110 7.6. Mail Streams Based on Behavioral Assessment . . . . . . . 34 111 7.7. Agent or Mediator Signatures . . . . . . . . . . . . . . . 34 112 8. Usage Considerations . . . . . . . . . . . . . . . . . . . . . 35 113 8.1. Non-standard Submission and Delivery Scenarios . . . . . . 35 114 8.2. Protection of Internal Mail . . . . . . . . . . . . . . . 36 115 8.3. Signature Granularity . . . . . . . . . . . . . . . . . . 36 116 8.4. Email Infrastructure Agents . . . . . . . . . . . . . . . 37 117 8.5. Mail User Agent . . . . . . . . . . . . . . . . . . . . . 39 118 9. Other Considerations . . . . . . . . . . . . . . . . . . . . . 40 119 9.1. Security Considerations . . . . . . . . . . . . . . . . . 40 120 9.2. IANA Considerations . . . . . . . . . . . . . . . . . . . 40 121 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 40 122 11. Informative References . . . . . . . . . . . . . . . . . . . . 40 123 Appendix A. Migration Strategies . . . . . . . . . . . . . . . . 42 124 A.1. Migrating from DomainKeys . . . . . . . . . . . . . . . . 42 125 A.2. Migrating Hash Algorithms . . . . . . . . . . . . . . . . 47 126 A.3. Migrating Signing Algorithms . . . . . . . . . . . . . . . 48 127 Appendix B. General Coding Criteria for Cryptographic 128 Applications . . . . . . . . . . . . . . . . . . . . 49 129 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 49 131 1. Introduction 133 DomainKeys Identified Mail (DKIM) allows an organization to claim 134 responsibility for transmitting a message, in a way that can be 135 validated by a recipient. This document provides practical tips for: 136 those who are developing DKIM software, mailing list managers, 137 filtering strategies based on the output from DKIM verification, and 138 DNS servers; those who are deploying DKIM software, keys, mailing 139 list software, and migrating from DomainKeys; and those who are 140 responsible for the on-going operations of an email infrastructure 141 that has deployed DKIM. 143 The document is organized around the key concepts related to DKIM. 144 Within each section, additional considerations specific to 145 development, deployment, or ongoing operations are highlighted where 146 appropriate. The possibility of use of DKIM results as input to a 147 local reputation database is also discussed. 149 2. Using DKIM as Part of Trust Assessment 151 2.1. A Systems View of Email Trust Assessment 153 DKIM participates in a trust-oriented enhancement to the Internet's 154 email service, to facilitate message handling decisions, such as for 155 delivery and for content display. Trust-oriented message handling 156 has substantial differences from approaches that consider messages in 157 terms of risk and abuse. With trust, there is a collaborative 158 exchange between a willing participant along the sending path and a 159 willing participant at the recipient site. In contrast, the risk 160 model entails independent action by the recipient site, in the face 161 of a potentially unknown, hostile and deceptive sender. This 162 translates into a very basic technical difference: In the face of 163 unilateral action by the recipient and even antagonistic efforts by 164 the sender, risk-oriented mechanisms will be based on heuristics, 165 that is, on guessing. Guessing produces statistical results with 166 some false negatives and some false positives. For trust-based 167 exchanges, the goal is the deterministic exchange of information. 168 For DKIM, that information is the one identifier that represents a 169 stream of mail for which an independent assessment is sought (by the 170 signer.) 172 A trust-based service is built upon a validated Responsible 173 Identifier that labels a stream of mail and is controlled by an 174 identity (role, person or organization). The identity is 175 acknowledging some degree of responsibility for the message stream. 176 Given a basis for believing that an identifier is being used in an 177 authorized manner, the recipient site can make and use an assessment 178 of the associated identity. An identity can use different 179 identifiers, on the assumption that the different streams might 180 produce different assessments. For example, even the best-run 181 marketing campaigns will tend to produce some complaints that can 182 affect the reputation of the associated identifier, whereas a stream 183 of transactional messages is likely to have a more pristine 184 reputation. 186 Determining that the identifier's use is valid is quite different 187 from determining that the content of a message is valid. The former 188 means only that the identifier for the responsible role, person or 189 organization has been legitimately associated with a message. The 190 latter means that the content of the message can be believed and, 191 typically, that the claimed author of the content is correct. DKIM 192 validates only the presence of the identifier used to sign the 193 message. Even when this identifier is validated, DKIM carries no 194 implication that any of the message content, including the 195 RFC5322.From field, is valid. Surprisingly, this limit to the 196 semantics of a DKIM signature applies even when the validated signing 197 identifier is the same domain name as is used in the From: field! 198 DKIM's only claim about message content is that the content cited in 199 the DKIM-Signature: field's h= tag has been delivered without 200 modification. That is, it asserts message content integrity, not 201 message content validity. 203 As shown in Figure 1, this enhancement is a communication between a 204 responsible role, person or organization that signs the message and a 205 recipient organization that assesses its trust in the signer and then 206 makes handling decisions based on a collection of assessments, of 207 which the DKIM mechanism is only a part. In this model, validation 208 is an intermediary step, having the sole task of passing a validated 209 Responsible Identifier to the Identity Assessor. The communication 210 is of a single Responsible Identifier that the Responsible Identity 211 wishes to have used by the Identity Assessor. The Identifier is the 212 sole, formal input and output value of DKIM signing. The Identity 213 Assessor uses this single, provided Identifier for consulting 214 whatever assessment data bases are deemed appropriate by the 215 assessing entity. In turn, output from the Identity Assessor is fed 216 into a Handling Filter engine that considers a range of factors, 217 along with this single output value; the range of factors can include 218 ancillary information from the DKIM validation. 220 Identity Assessment covers a range of possible functions. It can be 221 as simple as determining whether the identifier is a member of some 222 list, such as authorized operators or participants in a group that 223 might be of interest for recipient assessment. Equally, it can 224 indicate a degree of trust (reputation) that is to be afforded the 225 actor using that identifier. The extent to which the assessment 226 affects handling of the message is, of course, determined later, by 227 the Handling Filter. 229 +------+------+ +------+------+ 230 | Author | | Recipient | 231 +------+------+ +------+------+ 232 | ^ 233 | | 234 | +------+------+ 235 | -->| Handling |<-- 236 | -->| Filter |<-- 237 | +-------------+ 238 | ^ 239 V Responsible | 240 +-------------+ Identifier +------+------+ 241 | Responsible |. . . . . . . . . . .>| Identity | 242 | Identity | . . | Assessor | 243 +------+------+ . . +-------------+ 244 | . . ^ ^ 245 V . . | | 246 +------------------.-------.--------------------+ | | 247 | +------+------+ . . . . . +-------------+ | | | +-------------+ 248 | | Identifier | | Identifier +--|--+ +--+ Assessment | 249 | | Signer +------------->| Validator | | | Databases | 250 | +-------------+ +-------------+ | +-------------+ 251 | DKIM Service | 252 +-----------------------------------------------+ 254 Figure 1: Actors in a Trust Sequence using DKIM 256 2.2. Choosing a DKIM Tag for the Assessment Identifier 258 The signer of a message needs to be able to provide precise data and 259 know what that data will mean upon delivery to the Assessor. If 260 there is ambiguity in the choice that will be made on the receive 261 side, then the sender cannot know what basis for assessment will be 262 used. DKIM has three values that specify identification information 263 and it is easy to confuse their use, although only one defines the 264 formal input and output of DKIM, with the other two being used for 265 internal protocol functioning and adjunct purposes, such as auditing 266 and debugging. 268 The salient values include the s=, d= and i= parameters in the DKIM- 269 Signature: header field. In order to achieve the end-to-end 270 determinism needed for this collaborative exchange from the signer to 271 the assessor, the core model needs to specify what the signer is 272 required to provide to the assessor. The Update to RFC4871 273 [rfc4871-update]now specifies: 275 DKIM's primary task is to communicate from the Signer to a 276 recipient-side Identity Assessor a single Signing Domain 277 Identifier (SDID) that refers to a responsible identity. DKIM MAY 278 optionally provide a single responsible Agent or User Identifier 279 (AUID)... A receive-side DKIM verifier MUST communicate the 280 Signing Domain Identifier (d=) to a consuming Identity Assessor 281 module and MAY communicate the User Agent Identifier (i=) if 282 present.... To the extent that a receiver attempts to intuit any 283 structured semantics for either of the identifiers, this is a 284 heuristic function that is outside the scope of DKIM's 285 specification and semantics. 287 The single, mandatory value that DKIM supplies as its output is: 289 d= This specifies the "domain of the signing entity." It is a 290 domain name and is combined with the Selector to form a DNS 291 query... A receive-side DKIM verifier MUST communicate the 292 Signing Domain Identifier (d=) to a consuming Identity Assessor 293 module and MAY communicate the User Agent Identifier (i=) if 294 present. 296 The adjunct values are: 298 s= This tag specifies the Selector. It is used to discriminate 299 among different keys that can be used for the same d= domain 300 name. As discussed in Section 4.3 of [RFC5585]: "If verifiers 301 were to employ the selector as part of a name assessment 302 mechanism, then there would be no remaining mechanism for 303 making a transition from an old, or compromised, key to a new 304 one." Consequently, the Selector is not appropriate for use as 305 part or all of the identifier used to make assessments. 307 i= This tag is optional and provides the "[i]dentity of the 308 user or agent (e.g., a mailing list manager) on behalf of which 309 this message is signed." The identity can be in the syntax of 310 an entire email address or only a domain name. The domain name 311 can be the same as for d= or it can be a sub-name of the d= 312 name. 314 NOTE: Although the i= identity has the syntax of an email 315 address, it is not required to have that semantics. That is, 316 "the identity of the user" need not be the same as the user's 317 mailbox. For example the signer might wish to use i= to encode 318 user-related audit information, such as how they were accessing 319 the service at the time of message posting. Therefore it is 320 not possible to conclude anything from the i= string's 321 (dis)similarity to email addresses elsewhere in the header 323 So, i= can have any of these properties: 325 * Be a valid domain when it is the same as d= 327 * Appear to be a sub-domain of d= but might not even exist 329 * Look like a mailbox address but might have different semantics 330 and therefore not function as a valid email address 332 * Be unique for each message, such as indicating access details 333 of the user for the specific posting 335 This underscores why the tag needs to be treated as being opaque, 336 since it can represent any semantics, known only to the signer. 338 Hence, i= serves well as a token that is usable like a Web cookie, 339 for return to the signing ADMD -- such as for auditing and debugging. 340 Of course in some scenarios the i= string might provide a useful 341 adjunct value for additional (heuristic) processing by the Handling 342 Filter. 344 2.3. Choosing the Signing Domain Name 346 A DKIM signing entity can serve different roles, such as author of 347 content, versus operator of the mail service, versus operator of a 348 reputation service. In these different roles, the basis for 349 distinguishing among portions of email traffic can vary. For an 350 entity creating DKIM signatures it is likely that different portions 351 of its mail will warrant different levels of trust. For example: 353 * Mail is sent for different purposes, such as marketing vs. 354 transactional, and recipients demonstrate different patterns of 355 acceptance between these. 357 * For an operator of an email service, there often are distinct 358 sub-populations of users warranting different levels of trust 359 or privilege, such as paid vs. free users, or users engaged in 360 direct correspondence vs. users sending bulk mail. 362 * Mail originating outside an operator's system, such as when it 363 is redistributed by a mailing list service run by the operator, 364 will warrant a different reputation from mail submitted by 365 users authenticated with the operator. 367 It is therefore likely to be useful for a signer to use different d= 368 sub-domain names, for different message traffic streams, so that 369 receivers can make differential assessments. However, too much 370 differentiation -- that is, too fine a granularity of signing domains 371 -- makes it difficult for the receiver to discern a sufficiently 372 stable pattern of traffic for developing an accurate and reliable 373 assessment. So the differentiation needs to achieve a balance. 374 Generally in a trust system, legitimate signers have an incentive to 375 pick a small stable set of identities, so that recipients and others 376 can attribute reputations to them. The set of these identities a 377 receiver trusts is likely to be quite a bit smaller than the set it 378 views as risky. 380 The challenge in using additional layers of sub-domains is whether 381 the extra granularity will be useful for the assessor. In fact, 382 potentially excessive levels invites ambiguity: if the assessor does 383 not take advantage of the added granularity, then what granularity 384 will it use? That ambiguity would move the use of DKIM back to the 385 realm of heuristics, rather than the deterministic processing that is 386 its goal. 388 Hence the challenge is to determine a useful scheme for labeling 389 different traffic streams. The most obvious choices are among 390 different types of content and/or different types of authors. 391 Although stability is essential, it is likely that the choices will 392 change, over time, so the scheme needs to be flexible. 394 For those originating message content, the most likely choice of sub- 395 domain naming scheme will by based upon type of content, which can 396 use content-oriented labels or service-oriented labels. For example: 398 transaction.example.com 399 newsletter.example.com 400 bugreport.example.com 401 support.example.com 402 sales.example.com 403 marketing.example.com 405 where the choices are best dictated by whether they provide the 406 Identity Assessor with the ability to discriminate usefully among 407 streams of mail that demonstrate significantly different degrees of 408 recipient acceptance or safety. Again, the danger in providing too 409 fine a granularity is that related message streams that are labeled 410 separately will not benefit from an aggregate reputation. 412 For those operating messaging services on behalf of a variety of 413 customers, an obvious scheme to use has a different sub-domain label 414 for each customer. For example: 416 widgetco.example.net 417 moviestudio.example.net 418 bigbank.example.net 420 However it can also be appropriate to label by the class of service 421 or class of customer, such as: 423 premier.example.net 424 free.example.net 425 certified.example.net 427 Prior to using domain names for distinguishing among sources of data, 428 IP Addresses have been the basis for distinction. Service operators 429 typically have done this by dedicating specific outbound IP Addresses 430 to specific mail streams -- typically to specific customers. For 431 example, a university might want to distinguish mail from the 432 Administration, versus mail from the student dorms. In order to make 433 adoption of a DKIM-based service easier, it can be reasonable to 434 translate the same partitioning of traffic, using domain names in 435 place of the different IP Addresses. 437 2.4. Recipient-based Assessments 439 DKIM gives the recipient site's Identity Assessor a verifiable 440 identifier to use for analysis. Although the mechanism does not make 441 claims that the signer is a Good Actor or a Bad Actor, it does make 442 it possible to know that use of the identifier is valid. This is in 443 marked contrast with schemes that do not have authentication. 444 Without verification, it is not possible to know whether the 445 identifier -- whether taken from the RFC5322.From field, 446 RFC5321.MailFrom command, or the like -- is being used by an 447 authorized agent. DKIM solves this problem. Hence with DKIM, the 448 Assessor can know that two messages with the same DKIM d= identifier 449 are, in fact, signed by the same person or organization. This 450 permits a far more stable and accurate assessment of mail traffic 451 using that identifier. 453 DKIM is distinctive, in that it provides an identifier which is not 454 necessarily related to any other identifier in the message. Hence, 455 the signer might be the author's ADMD, one of the operators along the 456 transit path, or a reputation service being used by one of those 457 handling services. In fact, a message can have multiple signatures, 458 possibly by any number of these actors. 460 As discussed above, the choice of identifiers needs to be based on 461 differences that the signer thinks will be useful for the recipient 462 Assessor. Over time, industry practices establish norms for these 463 choices. 465 Absent such norms, it is best for signers to distinguish among 466 streams that have significant differences, while consuming the 467 smallest number of identifiers possible. This will limit the 468 burden on recipient Assessors. 470 A common view about a DKIM signature is that it carries a degree of 471 assurance about some or all of the message contents, and in 472 particular that the RFC5322.From field is likely to be valid. In 473 fact, DKIM makes assurances only about the integrity of the data and 474 not about its validity. Still, presumptions of From: field validity 475 remain a concern. Hence a signer using a domain name that is 476 unrelated to the domain name in the From: field can reasonably expect 477 that the disparity will warrant some curiosity, at least until 478 signing by independent operators has produced some established 479 practice among recipient Assessors. 481 With the identifier(s) supplied by DKIM, the Assessor can consult an 482 independent assessment service about the entity associated with the 483 identifier(s). Another possibility is that the Assessor can develop 484 its own reputation rating for the identifier(s). That is, over time, 485 the Assessor can observe the stream of messages associated with the 486 identifier(s) developing a reaction to associated content. For 487 example, if there is a high percentage of user complaints regarding 488 signed mail with a "d=" value of "widgetco.example.net", the Assessor 489 might include that fact in the vector of data it provides to the 490 Handling Filter. This is also discussed briefly in Section 5.4. 492 2.5. Filtering 494 After assessing the signer of a message, each receiving site creates 495 and tunes its own Handling Filter according to criteria specific for 496 that site. Still, there are commonalities across sites, and this 497 section offers a discussion, rather than a specification, of some 498 types of input to that process and how they can be used. 500 The discussion focuses on variations in Organizational Trust versus 501 Message Risk, that is, the degree of positive assessment of a DKIM- 502 signing organization, and the potential danger present in the message 503 stream signed by that organization. While it might seem that higher 504 trust automatically means lower risk, the experience with real-world 505 operations provides examples of every combination of the two factors, 506 as shown in Table 1. Only three levels of granularity are listed, in 507 order to keep discussion manageable. This also ensures extensive 508 flexibility for each site's detailed choices. 510 +---+---------------------+--------------------+--------------------+ 511 | | Low | Medium | High | 512 | | | | | 513 | | | | | 514 | | | | | 515 | | | | | 516 | O | | | | 517 | R | | | | 518 | G | | | | 519 | | | | | 520 | T | | | | 521 | R | | | | 522 | U | | | | 523 | S | | | | 524 | T | | | | 525 | | | | | 526 | M | | | | 527 +---+---------------------+--------------------+--------------------+ 528 | * | Unknown org, | Registered org, | Good Org, | 529 | L | Few msgs: | New Identifier: | Good msgs: | 530 | o | _Mild filtering_ | _Medium filtering_ | _Avoid FP(!)_ | 531 | w | | | | 532 | * | Unknown org, | Registered org, | Good org, Bad msg | 533 | M | New Identifier: | Mixed msgs: | burst: | 534 | e | _Default filtering_ | _Medium filtering_ | _Accept & Contact_ | 535 | d | | | | 536 | i | | | | 537 | u | | | | 538 | * | Black-Listed org, | Registered org, | Good org, | 539 | H | Bad msgs: | Bad msgs: | Compromised: | 540 | i | _Avoid FN(!)_ | _Strong filtering_ | _Fully blocked_ | 541 | g | | | | 542 | h | | | | 543 +---+---------------------+--------------------+--------------------+ 545 Table 1: Organizational Trust vs. Message Risk 547 The table indicates preferences for different handling of different 548 combinations, such as tuning filtering to avoid False Positives (FP) 549 or avoiding False Negatives (FN). Perhaps unexpectedly, it also 550 lists a case in which the receiving site might wish to deliver 551 problematic mail, rather than redirecting it, but also of course 552 contacting the signing organization, seeking resolution of the 553 problem. 555 3. DKIM Key Generation, Storage, and Management 557 By itself, verification of a digital signature only allows the 558 verifier to conclude with a very high degree of certainty that the 559 signature was created by a party with access to the corresponding 560 private signing key. It follows that a verifier requires means to 561 (1) obtain the public key for the purpose of verification and (2) 562 infer useful attributes of the key holder. 564 In a traditional Public Key Infrastructure (PKI), the functions of 565 key distribution and key accreditation are separated. In DKIM 566 [RFC4871], these functions are both performed through the DNS. 568 In either case, the ability to infer semantics from a digital 569 signature depends on the assumption that the corresponding private 570 key is only accessible to a party with a particular set of 571 attributes. In traditional PKI, a Trusted Third Party (TTP) vouches 572 that the key holder has been validated with respect to a specified 573 set of attributes. The range of attributes that may be attested in 574 such a scheme is thus limited only to the type of attributes that a 575 TTP can establish effective processes for validating. In DKIM, 576 Trusted Third parties are not employed and the functions of key 577 distribution and accreditation are combined. 579 Consequently there are only two types of inference that a signer may 580 make from a key published in a DKIM Key Record: 582 1. That a party with the ability to control DNS records within a DNS 583 zone intends to claim responsibility for messages signed using 584 the corresponding private signature key. 586 2. That use of a specific key is restricted to the particular subset 587 of messages identified by the selector. 589 The ability to draw any useful conclusion from verification of a 590 digital signature relies on the assumption that the corresponding 591 private key is only accessible to a party with a particular set of 592 attributes. In the case of DKIM, this means that the party that 593 created the corresponding DKIM key record in the specific zone 594 intended to claim responsibility for the signed message. 596 Ideally we would like to draw a stronger conclusion, that if we 597 obtain a DKIM key record from the DNS zone example.com, that the 598 legitimate holder of the DNS zone example.com claims responsibility 599 for the signed message. In order for this conclusion to be drawn it 600 is necessary for the verifier to assume that the operational security 601 of the DNS zone and corresponding private key are adequate. 603 3.1. Private Key Management: Deployment and Ongoing Operations 605 Access to signing keys MUST be carefully managed to prevent use by 606 unauthorized parties and to minimize the consequences if a compromise 607 were to occur. 609 While a DKIM signing key is used to sign messages on behalf of many 610 mail users, the signing key itself SHOULD be under direct control of 611 as few key holders as possible. If a key holder were to leave the 612 organization, all signing keys held by that key holder SHOULD be 613 withdrawn from service and if appropriate, replaced. 615 If key management hardware support is available, it SHOULD be used. 616 If keys are stored in software, appropriate file control protections 617 MUST be employed, and any location in which the private key is stored 618 in plaintext form SHOULD be excluded from regular backup processes 619 and SHOULD not be accessible through any form of network including 620 private local area networks. Auditing software SHOULD be used 621 periodically to verify that the permissions on the private key files 622 remain secure. 624 Wherever possible a signature key SHOULD exist in exactly one 625 location and be erased when no longer used. Ideally a signature key 626 pair SHOULD be generated as close to the signing point as possible 627 and only the public key component transferred to another party. If 628 this is not possible, the private key MUST be transported in an 629 encrypted format that protects the confidentiality of the signing 630 key. A shared directory on a local file system does not provide 631 adequate security for distribution of signing keys in plaintext form. 633 Key escrow schemes are not necessary and SHOULD NOT be used. In the 634 unlikely event of a signing key becomming lost, a new signature key 635 pair may be generated as easily as recovery from a key escrow scheme. 637 To enable accountability and auditing: 639 o Responsibility for the security of a signing key SHOULD ultimately 640 vest in a single named individual. 642 o Where multiple parties are authorized to sign messages, each 643 signer SHOULD use a different key to enable accountability and 644 auditing. 646 Best practices for management of cryptographic keying material 647 require keying material to be refreshed at regular intervals, 648 particularly where key management is achieved through software. 649 While this practice is highly desirable it is of considerably less 650 importance than the requirement to maintain the secrecy of the 651 corresponding private key. An operational practice in which the 652 private key is stored in tamper proof hardware and changed once a 653 year is considerably more desirable than one in which the signature 654 key is changed on an hourly basis but maintained in software. 656 3.2. Storing Public Keys: DNS Server Software Considerations 658 In order to use DKIM a DNS domain holder requires (1) the ability to 659 create the necessary DKIM DNS records and (2) sufficient operational 660 security controls to prevent insertion of spurious DNS records by an 661 attacker. 663 DNS record management is often operated by an administrative staff 664 that is different from those who operate an organization's email 665 service. In order to ensure that DKIM DNS records are accurate, this 666 imposes a requirement for careful coordination between the two 667 operations groups. If the best practices for private key management 668 described above are observed, such deployment is not a one time 669 event; DNS DKIM selectors will be changed over time signing keys are 670 terminated and replaced. 672 At a minimum, a DNS server that handles queries for DKIM key records 673 MUST allow the server administrators to add free-form TXT records. 674 It would be better if the the DKIM records could be entered using a 675 structured form, supporting the DKIM-specific fields. 677 Ideally DNSSEC [RFC4034] SHOULD be employed in a configuration that 678 provides protection against record insertion attacks and zone 679 enumeration. In the case that NSEC3 [RFC5155] records are employed 680 to prevent insertion attack, the OPT-OUT flag MUST be set clear. 682 3.2.1. Assignment of Selectors 684 Selectors are assigned according to the administrative needs of the 685 signing domain, such as for rolling over to a new key or for 686 delegating of the right to authenticate a portion of the namespace to 687 a trusted third party. Examples include: 689 jun2005.eng._domainkey.example.com 691 widget.promotion._domainkey.example.com 693 It is intended that assessments of DKIM identities be based on the 694 domain name, and not include the selector. While past practice of a 695 signer may permit a verifier to infer additional properties of 696 particular messages from the structure DKIM key selector, unannounced 697 administrative changes such as a change of signing softeware may 698 cause such heuristics to fail at any time. 700 3.3. Per User Signing Key Management Issues 702 While a signer may establish business rules, such as issue of 703 individual signature keys for each end-user, DKIM makes no provision 704 for communicating these to other parties. Out of band distribution 705 of such business rules is outside the scope of DKIM. Consequently 706 there is no means by which external parties may make use of such keys 707 to attribute messages with any greater granularity than a DNS domain. 709 If per-user signing keys are assigned for internal purposes (e.g. 710 authenticating messages sent to an MTA for distribution), the 711 following issues need to be considered before using such signatures 712 as an alternative to traditional edge signing at the outbound MTA: 714 External verifiers will be unable to make use of the additional 715 signature granularity without access to additional information 716 passed out of band with respect to [RFC4871]. 718 If the number of user keys is large, the efficiency of local 719 caching of key records by verifiers will be lower. 721 A large number of end users may be less likely to be able to 722 manage private key data securely on their personal computer than 723 an administrator running an edge MTA. 725 3.4. Third Party Signer Key Management and Selector Administration 727 A DKIM key record only asserts that the holder of the corresponding 728 domain name makes a claim of responsibility for messages signed under 729 the corresponding key. In some applications, such as bulk mail 730 delivery, it is desirable to delegate the ability to make a claim of 731 responsibility to another party. In this case the trust relationship 732 is established between the domain holder and the verifier but the 733 private signature key is held by a third party. 735 Signature keys used by a third party signer SHOULD be kept entirely 736 separate from those used by the domain holder and other third party 737 signers. To limit potential exposure of the private key, the 738 signature key pair SHOULD be generated by the third party signer and 739 the public component of the key transmitted to the domain holder, 740 rather than have the domain holder generate the key pair and transmit 741 the private component to the third party signer. 743 Domain holders SHOULD adopt a least privilege approach and grant 744 third party signers the minimum access necessary to perform the 745 desired function. Limiting the access granted to Third Party Signers 746 serves to protect the interests of both parties. The domain holder 747 minimizes its security risk and the Trusted Third Party Signer avoids 748 unnecessary liability. 750 In the most restrictive case a domain holder maintains full control 751 over the creation of key records and employs appropriate key record 752 restrictions to enforce restrictions on the messages for which the 753 third party signer is able to sign. If such restrictions are 754 impractical, the domain holder SHOULD delegate a DNS subzone for 755 publishing key records to the third party signer. The domain holder 756 SHOULD not allow a third party signer unrestricted access to its DNS 757 service for the purpose of publishing key records. 759 3.5. Key Pair / Selector Lifecycle Management 761 Deployments SHOULD establish, document and observe processes for 762 managing the entire lifecycle of a public key pair. 764 3.5.1. Example Key Deployment Process 766 When it is determined that a new key pair is required: 768 1. A Key Pair is generated by the signing device. 770 2. A proposed key selector record is generated and transmitted to 771 the DNS administration infrasrtructure. 773 3. The DNS administration infrastructure verifies the authenticity 774 of the key selector registration request. If accepted 776 1. A key selector is assigned. 778 2. The corresponding key record published in the DNS. 780 3. Wait for DNS updates to propagate (if necessary). 782 4. Report assigned key selector to signing device. 784 4. Signer verifies correct registration of the key record. 786 5. Signer begins generating signatures using the new key pair. 788 6. Signer terminates any private keys that are no longer required 789 due to issue of replacement. 791 3.5.2. Example Key Termination Process 793 When it is determined that a private signature key is no longer 794 required: 796 1. Signer stops using the private key for signature operations. 798 2. Signer deletes all records of the private key, including in- 799 memory copies at the signing device. 801 3. Signer notifies the DNS administration infrasrtructure that the 802 signing key is withdrawn from service and that the corresponding 803 key records may be withdrawn from service at a specified future 804 date. 806 4. The DNS administration infrastructure verifies the authenticity 807 of the key selector termination request. If accepted, 809 1. The key selector is scheduled for deletion at a future time 810 determined by site policy. 812 2. Wait for deletion time to arrive. 814 3. The signer either publishes a revocation key selector with an 815 empty "p=" field, or deletes the key selector record 816 entirely. 818 5. As far as the verifier is concerned, there is no functional 819 difference between verifying against a key selector with an empty 820 "p=" field, and verifying against a missing key selector: both 821 result in a failed signature and the signature should be treated 822 as if it had not been there. However, there is a minor semantic 823 difference: with the empty "p=" field, the signer is explicitly 824 stating that the key has been revoked. The empty "p=" record 825 provides a gravestone for an old selector, making it less likely 826 that the selector might be accidently reused with a different 827 public key. 829 4. Signing 831 Creating messages that have one or more DKIM signatures, requires 832 support in only two outbound email service components: 834 o A DNS Administrative interface that can create and maintain the 835 relevant DNS names -- including names with underscores -- and 836 resource records (RR). 838 o A trusted module, called the Signing Module, which is within the 839 organization's outbound email handling service and which creates 840 and adds the DKIM-Signature: header field(s) to the message. 842 If the module creates more than one signature, there needs to be the 843 appropriate means of telling it which one(s) to use. If a large 844 number of names is used for signing, it will help to have the 845 administrative tool support a batch processing mode. 847 4.1. DNS Records 849 A receiver attempting to verify a DKIM signature obtains the public 850 key that is associated with the signature for that message. The 851 DKIM-Signature: header in the message contains the d= tag with the 852 basic domain name doing the signing and serving as output to the 853 Identity Assessor, and the s= tag with the selector that is added to 854 the name, for finding the specific public key. Hence, the relevant 855 ._domainkey. DNS record needs to contain a 856 DKIM-related RR that provides the public key information. 858 The administrator of the zone containing the relevant domain name 859 adds this information. Initial DKIM DNS information is contained 860 within TXT RRs. DNS administrative software varies considerably in 861 its abilities to support DKIM names, such as with underscores, and to 862 add new types of DNS information. 864 4.2. Signing Module 866 The module doing signing can be placed anywhere within an 867 organization's trusted Administrative Management Domain (ADMD); 868 obvious choices include department-level posting agents, as well as 869 outbound boundary MTAs to the open Internet. However any other 870 module, including the author's MUA, is potentially acceptable, as 871 long as the signature survives any remaining handling within the 872 ADMD. Hence the choice among the modules depends upon software 873 development, administrative overhead, security exposures and transit 874 handling tradeoffs. One perspective that helps to resolve this 875 choice is the difference between the increased flexibility, from 876 placement at (or close to) the MUA, versus the streamlined 877 administration and operation, that is more easily obtained by 878 implementing the mechanism "deeper" into the organization's email 879 infrastructure, such as at its boundary MTA. 881 Note the discussion in Section 2.2, concerning use of the i= tag. 883 The signing module uses the appropriate private key to create one or 884 more signatures. The means by which the signing module obtains the 885 private key(s) is not specified by DKIM. Given that DKIM is intended 886 for use during email transit, rather than for long-term storage, it 887 is expected that keys will be changed regularly. For administrative 888 convenience, key information SHOULD NOT be hard-coded into software. 890 4.3. Signing Policies and Practices 892 Every organization (ADMD) will have its own policies and practices 893 for deciding when to sign messages (message stream) and with what 894 domain name, selector and key. Examples of particular message 895 streams include all mail sent from the ADMD, versus mail from 896 particular types of user accounts, versus mail having particular 897 types of content. Given this variability, and the likelihood that 898 signing practices will change over time, it will be useful to have 899 these decisions represented through run-time configuration 900 information, rather than being hard-coded into the signing software. 902 As noted in Section 2.3, the choice of signing name granularity 903 requires balancing administrative convenience and utility for 904 recipients. Too much granularity is higher administrative overhead 905 and well might attempt to impose more differential analysis on the 906 recipient than they wish to support. In such cases, they are likely 907 to use only a super-name -- right-hand substring -- of the signing 908 name. When this occurs, the signer will not know what portion is 909 being used; this then moves DKIM back to the non-deterministic world 910 of heuristics, rather than the mechanistic world of signer/recipient 911 collaboration that DKIM seeks. 913 5. Verifying 915 A message recipient may verify a DKIM signature to determine if a 916 claim of responsibility has been made for the message by a trusted 917 domain. 919 Access control requires two components: authentication and 920 authorization. By design, verification of a DKIM signature only 921 provides the authentication component of an access control decision 922 and MUST be combined with additional sources of information such as 923 reputation data to arrive at an access control decision. 925 5.1. Intended Scope of Use 927 DKIM requires that a message with a signature that is found to be 928 invalid is to be treated as if the message had not been signed at 929 all. 931 If a DKIM signature fails to verify, it is entirely possible that the 932 message is valid and that either there is a configuration error in 933 the signer's system (e.g. a missing key record) or that the message 934 was inadvertently modified in transit. It is thus undesirable for 935 mail infrastructure to treat messages with invalid signatures less 936 favorably than those with no signatures whatsoever. Contrariwise, 937 creation of an invalid signature requires a trivial amount of effort 938 on the part of an attacker. If messages with invalid signatures were 939 to be treated preferentially to messages with no signatures 940 whatsoever, attackers will simply add invalid signature blocks to 941 gain the preferential treatment. It follows that messages with 942 invalid signatures SHOULD be treated no better and no worse than 943 those with no signature at all. 945 5.2. Signature Scope 947 As with any other digital signature scheme, verifiers MUST only 948 consider the part of the message that is inside the scope of the 949 message as being authenticated by the signature. 951 For example, if the l= option is employed to specify a content length 952 for the scope of the signature, only the part of the message that is 953 within the scope of the content signature would be considered 954 authentic. 956 5.3. Design Scope of Use 958 Public Key cryptography provides an exceptionally high degree of 959 assurance, bordering on absolute certainty, that the party that 960 created a valid digital signature had access to the private key 961 corresponding to the public key indicated in the signature. 963 In order to make useful conclusions from the verification of a valid 964 digital signature, the verifier is obliged to make assumptions that 965 fall far short of absolute certainty. Consequently, mere validation 966 of a DKIM signature does not represent proof positive that a valid 967 claim of responsibility was made for it by the indicated party, that 968 the message is authentic, or that the message is not abusive. In 969 particular: 971 o The legitimate private key holder may have lost control of its 972 private key. 974 o The legitimate domain holder may have lost control of the DNS 975 server for the zone from which the key record was retrieved. 977 o The key record may not have been delivered from the legitimate DNS 978 server for the zone from which the key record was retrieved. 980 o Ownership of the DNS zone may have changed. 982 In practice these limitations have little or no impact on the field 983 of use for which DKIM is designed but may have a bearing if use is 984 made of the DKIM message signature format or key retrieval mechanism 985 in other specifications. 987 In particular the DKIM key retrieval mechanism is designed for ease 988 of use and deployment rather than to provide a high assurance Public 989 Key Infrastructure suitable for purposes that require robust non- 990 repudiation such as establishing legally binding contracts. 991 Developers seeking to extend DKIM beyond its design application 992 SHOULD consider replacing or supplementing the DNS key retreival 993 mechanism with one that is designed to meet the intended purposes. 995 5.4. Inbound Mail Filtering 997 DKIM is frequently employed in a mail filtering strategy to avoid 998 performing content analysis on email originating from trusted 999 sources. Messages that carry a valid DKIM signature from a trusted 1000 source may be whitelisted, avoiding the need to perform computation 1001 and hence energy intensive content analysis to determine the 1002 disposition of the message. 1004 Mail sources may be determined to be trusted by means of previously 1005 observed behavior and/or reference to external reputation or 1006 accreditation services. The precise means by which this is 1007 acomplished is outside the scope of DKIM. 1009 5.4.1. Non-Verifying Adaptive Spam Filtering Systems 1011 Adaptive (or learning) spam filtering mechanisms that are not capable 1012 of verifying DKIM signatures SHOULD at minimum be configured to 1013 ignore DKIM header data entirely. 1015 5.5. Messages sent through Mailing Lists and other Intermediaries 1017 Intermediaries such as mailing lists pose a particular challenge for 1018 DKIM implementations as the message processing steps performed by the 1019 intermediary may cause the message content to change in ways that 1020 prevent the signature passing verification. 1022 Such intermediaries are strongly encouraged to deploy DKIM signing so 1023 that a verifiable claim of responsibility remains available to 1024 parties attempting to verify the modified message. 1026 5.6. Generation, Transmission and Use of Results Headers 1028 In many deployments it is desirable to separate signature 1029 verification from the application relying on the verification. A 1030 system may choose to relay information indicating the results of its 1031 message authentication efforts using various means; adding a "results 1032 header" to the message is one such mechanism. [RFC5451] For example, 1033 consider the cases where: 1035 o The application relying on DKIM signature verification is not 1036 capable of performing the verification. 1038 o The message may be modified after the signature verification is 1039 performed. 1041 o The signature key may not be available by the time that the 1042 message is read. 1044 In such cases it is important that the communication link between the 1045 signature verifier and the relying application be sufficiently secure 1046 to prevent insertion of a message that carries a bogus results 1047 header. 1049 An intermediary that generates results headers SHOULD ensure that 1050 relying applications are able to distinguish valid results headers 1051 issued by the intermediary from those introduced by an attacker. For 1052 example, this can be accomplished by signing the results header. At 1053 a minimum, results headers on incoming messages SHOULD be removed if 1054 they purport to have been issued by the intermediary but cannot be 1055 verified as authentic. 1057 Further discussion on trusting the results as relayed from a verifier 1058 to something downstream can be found in [RFC5451] 1060 6. Taxonomy of Signatures 1062 As described in section Section 2.1, a DKIM signature tells the 1063 signature verifier that the owner of a particular domain name accepts 1064 some responsibility for the message. It does not, in and of itself, 1065 provide any information about the trustworthiness or behavior of that 1066 identity. What it does provide is a verified identity to which such 1067 behavioral information can be associated, so that those who collect 1068 and use such information can be assured that it truly pertains to the 1069 identity in question. 1071 This section lays out a taxonomy of some of the different identities, 1072 or combinations of identities, that might usefully be represented by 1073 a DKIM signature. 1075 6.1. Single Domain Signature 1077 Perhaps the simplest case is when an organization signs its own 1078 outbound email using its own domain in the SDID [rfc4871-update] of 1079 the signature. For example, Company A would sign the outbound mail 1080 from its employees with d=companyA.example. 1082 In the most straightforward configuration, the addresses in the 1083 RFC5322.From would also be in the companyA.example domain, but that 1084 direct correlation is not required. 1086 A special case of the Single Domain Signature is an Author Signature 1087 as defined by the Author Domain Signing Practices specification 1088 [I-D.ietf-dkim-ssp]. Author signatures are signatures from an 1089 author's organization that have an SDID value that matches that of an 1090 RFC5322.From address of the signed message. 1092 Although an author signature might in some cases be proof against 1093 spoofing the domain name of the RFC5322.From address, it is important 1094 to note that the DKIM and ADSP validation apply only to the exact 1095 address string and not to look-alike addresses nor to the human- 1096 friendly "display-name" or names and addresses used within the body 1097 of the message. That is, it protects only against the misuse of a 1098 precise address string within the RFC5322.From field and nothing 1099 else. For example, a message from bob@domain.example with a valid 1100 signature where d=d0main.example would fail an ADSP check because the 1101 signature domain, however similar, is distinct; however a message 1102 from bob@d0main.example with a valid signature where d=d0main.example 1103 would pass an ADSP check, even though to a human it might be obvious 1104 that d0main.example is likely a malicious attempt to spoof the domain 1105 domain.example. This example highlights that ADSP, like DKIM, is 1106 only able to validate a signing identifier: it still requires some 1107 external process to attach a meaningful reputation to that 1108 identifier. 1110 6.2. Parent Domain Signature 1112 Another approach that might be taken by an organization with multiple 1113 active subdomains is to apply the same (single) signature domain to 1114 mail from all subdomains. In this case, the signature chosen would 1115 usually be the signature of a parent domain common to all subdomains. 1116 For example, mail from marketing.domain.example, 1117 sales.domain.example, and engineering.domain.example might all use a 1118 signature where d=domain.example. 1120 This approach has the virtue of simplicity, but it is important to 1121 consider the implications of such a choice. As discussed in 1122 Section 2.3, if the type of mail sent from the different subdomains 1123 is significantly different or if there is reason to believe that the 1124 reputation of the subdomains would differ, then it may be a good idea 1125 to acknowledge this and provide distinct signatures for each of the 1126 subdomains (d=marketing.domain.example, sales.domain.example, etc.). 1127 However, if the mail and reputations are likely to be similar, then 1128 the simpler approach of using a single common parent domain in the 1129 signature may work well. 1131 Another approach to distinguishing the streams using a single DKIM 1132 key would be to leverage the AUID [rfc4871-update] (i= tag) in the 1133 DKIM signature to differentiate the mail streams. For example, 1134 marketing email would be signed with i=marketing.domain.example and 1135 d=domain.example. 1137 It's important to remember, however, that under core DKIM semantics 1138 the AUID is opaque to receivers. That means that it will only be an 1139 effective differentiator if there is an out of band agreement about 1140 the i= semantics. 1142 6.3. Third Party Signature 1144 A signature whose domain does not match the domain of the 1145 RFC5322.From address is sometimes referred to as a third party 1146 signature. In certain cases even the parent domain signature 1147 described above would be considered a third party signature because 1148 it would not be an exact match for the domain in the From: address. 1150 Although there is often heated debate about the value of third party 1151 signatures, it is important to note that the DKIM specification 1152 attaches no particular significance to the identity in a DKIM 1153 signature. The identity specified within the signature is the 1154 identity that is taking responsibility for the message, and it is 1155 only the interpretation of a given receiver that gives one identity 1156 more or less significance than another. In particular, most 1157 independent reputation services assign trust based on the specific 1158 identifier string, not its "role": in general they make no 1159 distinction between, for example, an author signature and a third 1160 party signature. 1162 For some, a signature unrelated to the author domain (the domain in 1163 the RFC5322.From address) is less valuable because there is an 1164 assumption that the presence of an author signature guarantees that 1165 the use of the address in the From: header is authorized. 1167 For others, that relevance is tied strictly to the recorded 1168 behavioral data assigned to the identity in question, i.e. its trust 1169 assessment or reputation. The reasoning here is that an identity 1170 with a good reputation is unlikely to maintain that good reputation 1171 if it is in the habit of vouching for messages that are unwanted or 1172 abusive; in fact, doing so will rapidly degrade its reputation so 1173 that future messages will no longer benefit from it. It is therefore 1174 low risk to facilitate the delivery of messages that contain a valid 1175 signature of a domain with a strong positive reputation, independent 1176 of whether or not that domain is associated with the address in the 1177 RFC5322.From header field of the message. 1179 Third party signatures encompass a wide range of identities. Some of 1180 the more common are: 1182 Service Provider: In cases where email is outsourced to an Email 1183 Service Provider (ESP), Internet Service Provider (ISP), or other 1184 type of service provider, that service provider may choose to DKIM 1185 sign outbound mail with either its own identifier -- relying on 1186 its own, aggregate reputation -- or with a subdomain of the 1187 provider that is unique to the message author but still part of 1188 the provider's aggregate reputation. Such service providers may 1189 also encompass delegated business functions such as benefit 1190 management, although these will more often be treated as trusted 1191 third party senders (see below). 1193 Parent Domain. As discussed above, organizations choosing to apply a 1194 parent domain signature to mail originating from subdomains may 1195 have their signatures treated as third party by some verifiers, 1196 depending on whether or not the "t=s" tag is used to constrain the 1197 parent signature to apply only to its own specific domain. The 1198 default is to consider a parent domain signature valid for its 1199 subdomains. 1201 Reputation Provider: Another possible category of third party 1202 signature would be the identity of a third party reputation 1203 provider. Such a signature would indicate to receivers that the 1204 message was being vouched for by that third party. 1206 6.4. Using Trusted Third Party Senders 1208 For most of the cases described so far, there has been an assumption 1209 that the signing agent was responsible for creating and maintaining 1210 its own DKIM signing infrastructure, including its own keys, and 1211 signing with its own identity. 1213 A different model arises when an organization uses a trusted third 1214 party sender for certain key business functions, but still wants that 1215 email to benefit from the organization's own identity and reputation: 1216 in other words, the mail would come out of the trusted third party's 1217 mail servers, but the signature applied would be that of the 1218 controlling organization. 1220 This can be done by having the third party generate a key pair that 1221 is designated uniquely for use by that trusted third party and 1222 publishing the public key in the controlling organization's DNS 1223 domain, thus enabling the third party to sign mail using the 1224 signature of the controlling organization. For example, if Company A 1225 outsources its employee benefits to a third party, it can use a 1226 special key pair that enables the benefits company to sign mail as 1227 "companyA.example". Because the key pair is unique to that trusted 1228 third party, it is easy for Company A to revoke the authorization if 1229 necessary by simply removing the public key from the companyA.example 1230 DNS. 1232 A more cautious approach would be to create a dedicated subdomain 1233 (e.g. benefits.companyA.example) to segment the outsourced mail 1234 stream, and to publish the public key there; the signature would then 1235 use d=benefits.companyA.example. 1237 6.4.1. DNS Delegation 1239 Another possibility for configuring trusted third party access, as 1240 discussed in section 3.4, is to have Company A use DNS delegation and 1241 have the designated subdomain managed directly by the trusted third 1242 party. In this case, Company A would create a subdomain 1243 benefits.companya.example, and delegate the DNS management of that 1244 subdomain to the benefits company so it could maintain its own key 1245 records. Should revocation become necessary, Company A could simply 1246 remove the DNS delegation record. 1248 6.5. Multiple Signatures 1250 A simple configuration for DKIM-signed mail is to have a single 1251 signature on a given message. This works well for domains that 1252 manage and send all of their own email from single sources, or for 1253 cases where multiple email streams exist but each has its own unique 1254 key pair. It also represents the case in which only one of the 1255 participants in an email sequence is able to sign, no matter whether 1256 it represents the author or one of the operators. 1258 The examples thus far have considered the implications of using 1259 different identities in DKIM signatures, but have used only one such 1260 identity for any given message. In some cases, it may make sense to 1261 have more than one identity claiming responsibility for the same 1262 message. 1264 There are a number of situations where applying more than one DKIM 1265 signature to the same message might make sense. A few examples are: 1267 Companies with multiple subdomain identities: A company that has 1268 multiple subdomains sending distinct categories of mail might 1269 choose to sign with distinct subdomain identities to enable each 1270 subdomain to manage its own identity. However, it might also want 1271 to provide a common identity that cuts across all of the distinct 1272 subdomains. For example, Company A may sign mail for its sales 1273 department with a signature where d=sales.companya.example, and a 1274 second signature where d=companya.example 1276 Service Providers: A service providers may, as described above, 1277 choose to sign outbound messages with either its own identity or 1278 with an identity unique to each of its clients (possibly 1279 delegated). However, it may also do both: sign each outbound 1280 message with its own identity as well as with the identity of each 1281 individual client. For example, ESP A might sign mail for its 1282 client Company B with its service provider signature 1283 d=espa.example, and a second client-specific signature where d= 1284 either companyb.example, or companyb.espa.example. The existence 1285 of the service provider signature could, for example, help cover a 1286 new client while it establishes its own reputation, or help a very 1287 small volume client who might never reach a volume threshold 1288 sufficient to establish an individual reputation. 1290 Forwarders Forwarded mail poses a number of challenges to email 1291 authentication. DKIM is relatively robust in the presence of 1292 forwarders as long as the signature is designed to avoid message 1293 parts that are likely to be modified; however, some forwarders do 1294 make modifications that can invalidate a DKIM signature. 1296 Some forwarders such as mailing lists or "forward article to a 1297 friend" services might choose to add their own signatures to 1298 outbound messages to vouch for them having legitimately originated 1299 from the designated service. In this case, the signature would be 1300 added even in the presence of a preexisting signature, and both 1301 signatures would be relevant to the verifier. 1303 Any forwarder that modifies messages in ways that will break 1304 preexisting DKIM signatures SHOULD always sign its forwarded 1305 messages. 1307 Reputation Providers: Although third party reputation providers 1308 today use a variety of protocols to communicate their information 1309 to receivers, it is possible that they, or other organizations 1310 willing to put their "seal of approval" on an email stream might 1311 choose to use a DKIM signature to do it. In nearly all cases, 1312 this "reputation" signature would be in addition to the author or 1313 originator signature. 1315 One important caveat to the use of multiple signatures is that there 1316 is currently no clear consensus among receivers on how they plan to 1317 handle them. The opinions range from ignoring all but one signature 1318 (and the specification of which of them is verified differs from 1319 receiver to receiver), to verifying all signatures present and 1320 applying a weighted blend of the trust assessments for those 1321 identifiers, to verifying all signatures present and simply using the 1322 identifier that represents the most positive trust assessment. It is 1323 likely that the industry will evolve to accept multiple signatures 1324 using either the second or third of these, but it may take some time 1325 before one approach becomes pervasive. 1327 7. Example Usage Scenarios 1329 Signatures are created by different types of email actors, based on 1330 different criteria, such as where the actor operates in the sequence 1331 from author to recipient, whether they want different messages to be 1332 evaluated under the same reputation or a different one, and so on. 1333 This section provides some examples of usage scenarios for DKIM 1334 deployments; the selection is not intended to be exhaustive, but to 1335 illustrate a set of key deployment considerations. 1337 7.1. Author's Organization - Simple 1339 The simplest DKIM configuration is to have some mail from a given 1340 organization (Company A) be signed with the same d= value (e.g. 1341 d=companya.example). If there is a desire to associate additional 1342 information, the AUID [rfc4871-update] value can become 1343 uniqueID@companya.example, or @uniqueID.companya.example. 1345 In this scenario, Company A need only generate a single signing key 1346 and publish it under their top level domain (companya.example); the 1347 signing module would then tailor the AUID value as needed at signing 1348 time. 1350 7.2. Author's Organization - Differentiated Types of Mail 1352 A slight variation of the one signature case is where Company A signs 1353 some of its mail, but it wants to differentiate different categories 1354 of its outbound mail by using different identifiers. For example, it 1355 might choose to distinguish marketing mail, billing or transactional 1356 mail, and individual corporate email into marketing.companya.example, 1357 billing.companya.example, and companya.example, where each category 1358 is assigned a unique subdomain and unique signing keys. 1360 7.3. Author Domain Signing Practices 1362 7.3.1. Introduction 1364 Some domains may decide to sign all of their outgoing mail. If all 1365 of the legitimate mail for a domain is signed, recipients can be more 1366 aggressive in their filtering of mail that uses the domain but does 1367 not have a valid signature from the domain; in such a configuration, 1368 the absence of a signature would be more significant than for the 1369 general case. It might be desirable for such domains to be able to 1370 advertise their intent to other receivers: this is the topic of 1371 Author Domain Signing Practices (ADSP). 1373 Note that ADSP is not for everyone. Sending domains that do not 1374 control all legitimate outbound mail purporting to be from their 1375 domain (i.e., with a RFC5322.From address in their domain) are likely 1376 to experience delivery problems with some percentage of that mail. 1377 Administrators evaluating ADSP for their domains SHOULD carefully 1378 weigh the risk of phishing attacks against the likelihood of 1379 undelivered mail. 1381 This section covers some examples of ADSP usage: for the complete 1382 specification, see [I-D.ietf-dkim-ssp] 1384 7.3.2. A Few Definitions 1386 In the ADSP specification, an address in the From header field of a 1387 message [RFC5322] is defined as an "Author Address", and an "Author 1388 Domain" is defined as anything to the right of the '@' in an Author 1389 Address. 1391 An "Author Signature" is thus any valid signature where the value of 1392 the SDID matches an Author Domain in the message. 1394 It is important to note that unlike the DKIM specification which 1395 makes no correlation between the signature domain and any message 1396 headers, the ADSP specification applies only to the author domain. 1397 In essence, under ADSP, any non-author signatures are ignored 1398 (treated as if they are not present). 1400 Signers wishing to publish an Author Domain Signing Practices (ADSP) 1401 [I-D.ietf-dkim-ssp] record describing their signing practices will 1402 thus want to include an author signature on their outbound mail to 1403 avoid ADSP verification failures. For example, if the address in the 1404 RFC5322.From is bob@company.example, the SDID value of the author 1405 signature must be company.example. 1407 7.3.3. Some ADSP Examples 1409 An organization (Company A) may specify its signing practices by 1410 publishing an ADSP record with "dkim=all" or "dkim=discardable". In 1411 order to avoid misdelivery of its mail at receivers that are 1412 validating ADSP, Company A MUST first have done an exhaustive 1413 analysis to determine all sources of outbound mail from its domain 1414 (companyA.example) and ensure that they all have valid author 1415 signatures from that domain. 1417 For example, email with an RFC5322.From address of bob@ 1418 companyA.example MUST have an author signature where the SDID value 1419 is "companyA.example" or it will fail an ADSP validation. 1421 Note that once an organization publishes an ADSP record using 1422 dkim=all or dkim=discardable, any email with a RFC5322.From address 1423 that uses the domain where the ADSP record is published that does not 1424 have a valid author signature is at risk of being misdelivered or 1425 discarded. For example, if a message with an RFC5322.From address of 1426 newsletter@companyA.example has a signature with 1427 d=marketing.companyA.example, that message will fail the ADSP check 1428 because the signature would not be considered a valid author 1429 signature. 1431 Because the semantics of an ADSP author signature are more 1432 constrained than the semantics of a "pure" DKIM signature, it is 1433 important to make sure the nuances are well understood before 1434 deploying an ADSP record. The ADSP specification [I-D.ietf-dkim-ssp] 1435 provides some fairly extensive lookup examples (in Appendix A) and 1436 usage examples (in Appendix B). 1438 In particular, in order to prevent mail from being negatively 1439 impacted or even discarded at the receiver, it is essential to 1440 perform a thorough survey of outbound mail from a domain before 1441 publishing an ADSP policy of anything stronger than "unknown". This 1442 includes mail that might be sent from external sources that may not 1443 be authorized to use the domain signature, as well as mail that risks 1444 modification in transit that might invalidate an otherwise valid 1445 author signature (e.g. mailing lists, courtesy forwarders, and other 1446 paths that could add or modify headers, or modify the message body). 1448 7.4. Delegated Signing 1450 An organization may choose to outsource certain key services to an 1451 independent company. For example, Company A might outsource its 1452 benefits management, or Organization B might outsource its marketing 1453 email. 1455 If Company A wants to ensure that all of the mail sent on its behalf 1456 through the benefits providers email servers shares the Company A 1457 reputation, as discussed in Section 6.4 it can either publish keys 1458 designated for the use of the benefits provider under 1459 companyA.example (preferably under a designated subdomain of 1460 companyA.example), or it can delegate a subdomain (e.g. 1461 benefits.companyA.example) to the provider and enable the provider to 1462 generate the keys and manage the DNS for the designated subdomain. 1464 In both of these cases, mail would be physically going out of the 1465 benefit provider's mail servers with a signature of e.g. 1466 d=benefits.companya.example. Note that the From: address is not 1467 constrained: it could either be affiliated with the benefits company 1468 (e.g. benefits-admin@benefitprovider.example, or 1469 benefits-provider@benefits.companya.example), or with the companyA 1470 domain. 1472 Note that in both of the above scenarios, as discussed in 1473 Section 3.4, security concerns dictate that the keys be generated by 1474 the organization that plans to do the signing so that there is no 1475 need to transfer the private key. In other words, the benefits 1476 provider would generate keys for both of the above scenarios. 1478 7.5. Independent Third Party Service Providers 1480 Another way to manage the service provider configuration would be to 1481 have the service provider sign the outgoing mail on behalf of its 1482 client Company A with its own (provider) identifier. For example, an 1483 Email Service Provider (ESP A) might want to share its own mailing 1484 reputation with its clients, and may sign all outgoing mail from its 1485 clients with its own d= domain (e.g. d=espa.example). 1487 Should the ESP want to distinguish among its clients, it has two 1488 options: 1490 o Share the SDID domain, and use the AUID value to distinguish among 1491 the clients: e.g. a signature on behalf of client A would have 1492 d=espa.example and i=clienta.espa.example (or 1493 i=clienta@espa.example) 1495 o Extend the SDID domain, so there is a unique value (and subdomain) 1496 for each client: e.g. a signature on behalf of client A would have 1497 d=clienta.espa.example. 1499 Note that this scenario and the delegation scenario are not mutually 1500 exclusive: in some cases, it may be desirable to sign the same 1501 message with both the ESP and the ESP client identities. 1503 7.6. Mail Streams Based on Behavioral Assessment 1505 An ISP (ISP A) might want to assign signatures to outbound mail from 1506 its users according to each user's past sending behavior 1507 (reputation). In other words, the ISP would segment its outbound 1508 traffic according to its own assessment of message quality, to aid 1509 recipients in differentiating among these different streams. Since 1510 the semantics of behavioral assessments are not valid AUID values, 1511 ISP A (ispa.example) may configure subdomains corresponding to the 1512 assessment categories (e.g. good.ispa.example, neutral.ispa.example, 1513 bad.ispa.example), and use these subdomains in the d= value of the 1514 signature. 1516 The signing module may also set the AUID value to have a unique user 1517 id (distinct from the local-part of the user's email address), for 1518 example user3456@neutral.domain.example. Using a userid that is 1519 distinct from a given email alias is useful in environments where a 1520 single user might register multiple email aliases. 1522 Note that in this case the AUID values are only partially stable. 1523 They are stable in the sense that a given i= value will always 1524 represent the same identity, but they are unstable in the sense that 1525 a given user can migrate among the assessment subdomains depending on 1526 their sending behavior (i.e., the same user might have multiple AUID 1527 values over the lifetime of a single account). 1529 In this scenario, ISP A may generate as many keys as there are 1530 assessment subdomains (SDID values), so that each assessment 1531 subdomain has its own key. The signing module would then choose its 1532 signing key based on the assessment of the user whose mail was being 1533 signed, and if desired include the user id in the AUID of the 1534 signature. As discussed earlier, the per-user granularity of the 1535 AUID may be ignored by many verifiers, so organizations choosing to 1536 use it should not rely on its use for receiver side filtering 1537 results; however, some organizations may also find the information 1538 useful for their own purposes in processing bounces or abuse reports. 1540 7.7. Agent or Mediator Signatures 1542 Another scenario is that of an agent, usually a re-mailer of some 1543 kind, that signs on behalf of the service or organization that it 1544 represents. Some examples of agents might be a mailing list manager, 1545 or the "forward article to a friend" service that many online 1546 publications offer. In most of these cases, the signature is 1547 asserting that the message originated with, or was relayed by, the 1548 service asserting responsibility. In general, if the service is 1549 configured in such a way that its forwarding would break existing 1550 DKIM signatures, it SHOULD always add its own signature. 1552 8. Usage Considerations 1554 8.1. Non-standard Submission and Delivery Scenarios 1556 The robustness of DKIM's verification mechanism is based on the fact 1557 that only authorized signing modules have access to the designated 1558 private key. This has the side effect that email submission and 1559 delivery scenarios that originate or relay messages from outside the 1560 domain of the authorized signing module will not have access to that 1561 protected private key, and thus will be unable to attach the expected 1562 domain signature to those messages. Such scenarios include mailing 1563 lists, courtesy forwarders, MTAs at hotels, hotspot networks used by 1564 travelling users, and other paths that could add or modify headers, 1565 or modify the message body. 1567 For example, assume Joe works for Company A and has an email address 1568 joe@companya.example. Joe also has a ISP-1 account 1569 joe@isp1.example.com, and he uses ISP-1's multiple address feature to 1570 attach his work email joe@companya.example to his ISP-1 account. 1571 When Joe sends email from his ISP-1 account and uses 1572 joe@companya.example as his designated From: address, that email 1573 cannot have a signature with d=companya.example because the ISP-1 1574 servers have no access to Company A's private key. In ISP-1's case 1575 it will have a ISP-1 signature, but for some other mail clients 1576 offering the same multiple address feature there may be no signature 1577 at all on the message. 1579 Another example might be the use of a forward article to a friend 1580 service. Most instances of these services today allow someone to 1581 send an article with their email address in the RFC5322.From to their 1582 designated recipient. If Joe used either of his two addresses 1583 (joe@companya.example or joe@isp1.example.com), the forwarder would 1584 be equally unable to sign with a corresponding domain . As in the 1585 mail client case, the forwarder may either sign as its own domain, or 1586 may put no signature on the message. 1588 A third example is the use of privately configured forwarding. 1589 Assume that Joe has another account at ISP-2, joe@isp-2.example.com, 1590 but he'd prefer to read his ISP-2 mail from his ISP-1 account. He 1591 sets up his ISP-2 account to forward all incoming mail to 1592 joe@isp1.example.com. Assume alice@companyb.example sends 1593 joe@isp-2.example.com an email. Depending on how companyb.example 1594 configured its signature, and depending on whether or not ISP-2 1595 modifies messages that it forwards, it is possible that when Alice's 1596 message is received in Joe's ISP-1 account the original signature 1597 fails verification. 1599 8.2. Protection of Internal Mail 1601 One identity is particularly amenable to easy and accurate 1602 assessment: the organization's own identity. Members of an 1603 organization tend to trust messages that purport to be from within 1604 that organization. However Internet Mail does not provide a 1605 straightforward means of determining whether such mail is, in fact, 1606 from within the organization. DKIM can be used to remedy this 1607 exposure. If the organization signs all of its mail, then its 1608 boundary MTAs can look for mail purporting to be from the 1609 organization that does not contain a verifiable signature. 1611 Such mail can in most cases be presumed to be spurious. However, 1612 domain managers are advised to consider the ways that mail processing 1613 can modify messages in ways that will invalidate an existing DKIM 1614 signature: mailing lists, courtesy forwarders, and other paths that 1615 could add or modify headers or modify the message body (e.g. MTAs at 1616 hotels, hotspot networks used by travelling users, and other 1617 scenarios described in the previous section). Such breakage is 1618 particularly relevant in the presence of Author Domain Signing 1619 Practices. 1621 8.3. Signature Granularity 1623 Although DKIM's use of domain names is optimized for a scope of 1624 organization-level signing, it is possible to administer sub-domains 1625 or otherwise adjust signatures in a way that supports per-user 1626 identification. This user level granularity can be specified in two 1627 ways: either by sharing the signing identity and specifying an 1628 extension to the i= value that has a per-user granularity, or by 1629 creating and signing with unique per-user keys. 1631 A subdomain or local part in the i= tag SHOULD be treated as an 1632 opaque identifier and thus need not correspond directly to a DNS 1633 subdomain or be a specific user address. 1635 The primary way to sign with per-user keys requires each user to have 1636 a distinct DNS (sub)domain, where each distinct d= value has a key 1637 published. (It is possible, although not recommended, to publish the 1638 same key in more than one distinct domain.) 1640 It is technically possible to publish per-user keys within a single 1641 domain or subdomain by utilizing different selector values. This is 1642 not recommended and is unlikely to be treated uniquely by Assessors: 1643 the primary purpose of selectors is to facilitate key management, and 1644 the DKIM specification recommends against using them in determining 1645 or assessing identies. 1647 In most cases, it would be impractical to sign email on a per-user 1648 granularity. Such an approach would be 1650 likely to be ignored: In most cases today, if receivers are 1651 verifying DKIM signatures they are in general taking the simplest 1652 possible approach. In many cases maintaining reputation 1653 information at a per user granularity is not interesting to them, 1654 in large part because the per user volume is too small to be 1655 useful or interesting. So even if senders take on the complexity 1656 necessary to support per user signatures, receivers are unlikely 1657 to retain anything more than the base domain reputation. 1659 difficult to manage: Any scheme that involves maintenance of a 1660 significant number of public keys may require infrastructure 1661 enhancements or extensive administrative expertise. For domains 1662 of any size, maintaining a valid per-user keypair, knowing when 1663 keys need to be revoked or added due to user attrition or 1664 onboarding, and the overhead of having the signing engine 1665 constantly swapping keys can create significant and often 1666 unnecessary managment complexity. It is also important to note 1667 that there is no way within the scope of the DKIM specification 1668 for a receiver to infer that a sender intends a per-user 1669 granularity. 1671 As mentioned before, what may make sense, however, is to use the 1672 infrastructure that enables finer granularity in signatures to 1673 identify segments smaller than a domain but much larger than a per- 1674 user segmentation. For example, a university might want to segment 1675 student, staff, and faculty mail into three distinct streams with 1676 differing reputations. This can be done by creating seperate sub- 1677 domains for the desired segments, and either specifying the 1678 subdomains in the i= tag of the DKIM Signature or by adding 1679 subdomains to the d= tag and assigning and signing with different 1680 keys for each subdomain. 1682 For those who choose to represent user level granularity in 1683 signatures, the performance and management considerations above 1684 suggest that it would be more effective to do it by specifying a 1685 local part or subdomain extension in the i= tag rather than by 1686 extending the d= domain and publishing individual keys. 1688 8.4. Email Infrastructure Agents 1690 It is expected that the most common venue for a DKIM implementation 1691 will be within the infrastructure of an organization's email service, 1692 such as a department or a boundary MTA. What follows are some 1693 general recommendations for the Email Infrastructure. 1695 Outbound: An MSA or an Outbound MTA used for mail submission 1696 SHOULD ensure that the message sent is in compliance with the 1697 advertised email sending policy. It SHOULD also be able to 1698 generate an operator alert if it determines that the email 1699 messages do not comply with the published DKIM sending policy. 1701 An MSA SHOULD be aware that some MUAs may add their own 1702 signatures. If the MSA needs to perform operations on a 1703 message to make it comply with its email sending policy, if at 1704 all possible, it SHOULD do so in a way that would not break 1705 those signatures. 1707 MUAs equipped with the ability to sign SHOULD NOT be 1708 encouraged. In terms of security, MUAs are generally not under 1709 the direct control of those in responsible roles within an 1710 organization and are thus more vulnerable to attack and 1711 compromise, which would expose private signing keys to 1712 intruders and thus jeopardize the integrity and reputation of 1713 the organization. 1715 Inbound: When an organization deploys DKIM, it needs to make 1716 sure that its email infrastructure components that do not have 1717 primary roles in DKIM handling do not modify message in ways 1718 that prevent subsequent verification. 1720 An inbound MTA or an MDA may incorporate an indication of the 1721 verification results into the message, such as using an 1722 Authentication-Results header field. [RFC5451] 1724 Intermediaries: An email intermediary is both an inbound and 1725 outbound MTA. Each of the requirements outlined in the 1726 sections relating to MTAs apply. If the intermediary modifies 1727 a message in a way that breaks the signature, the intermediary 1729 + SHOULD deploy abuse filtering measures on the inbound mail, 1730 and 1732 + MAY remove all signatures that will be broken 1734 In addition the intermediary MAY: 1736 + Verify the message signature prior to modification. 1738 + Incorporate an indication of the verification results into 1739 the message, such as using an Authentication-Results header 1740 field. [RFC5451] 1742 + Sign the modified message including the verification results 1743 (e.g., the Authentication-Results header field). 1745 8.5. Mail User Agent 1747 The DKIM specification is expected to be used primarily between 1748 Boundary MTAs, or other infrastructure components of the originating 1749 and receiving ADMDs. However there is nothing in DKIM that is 1750 specific to those venues. In particular, MUAs MAY also support DKIM 1751 signing and verifying directly. 1753 Outbound: An MUA MAY support signing even if mail is to be 1754 relayed through an outbound MSA. In this case the signature 1755 applied by the MUA will be in addition to any signature added 1756 by the MSA. However, the warnings in the previous section 1757 should be taken into consideration. 1759 Some user software goes beyond simple user functionality and 1760 also perform MSA and MTA functions. When this is employed for 1761 sending directly to a receiving ADMD, the user software SHOULD 1762 be considered an outbound MTA. 1764 Inbound: An MUA MAY rely on a report of a DKIM signature 1765 verification that took place at some point in the inbound MTA/ 1766 MDA path (e.g., an Authentication-Results header field), or an 1767 MUA MAY perform DKIM signature verification directly. A 1768 verifying MUA SHOULD allow for the case where mail has modified 1769 in the inbound MTA path; if a signature fails, the message 1770 SHOULD NOT be treated any different than if it did not have a 1771 signature. 1773 An MUA that looks for an Authentication-Results header field 1774 MUST be configurable to choose which Authentication-Results are 1775 considered trustable. The MUA developer is encouraged to re- 1776 read the Security Considerations of [RFC5451]. 1778 DKIM requires that all verifiers treat messages with signatures 1779 that do not verify as if they are unsigned. 1781 If verification in the client is to be acceptable to users, it 1782 is essential that successful verification of a signature not 1783 result in a less than satisfactory user experience compared to 1784 leaving the message unsigned. The mere presence of a verified 1785 DKIM signature MUST NOT by itself be used by an MUA to indicate 1786 that a message is to be treated better than a message without a 1787 verified DKIM signature. However, the fact that a DKIM 1788 signature was verified MAY be used as input into a reputation 1789 system (i.e., a whitelist of domains and users) for 1790 presentation of such indicators. 1792 It is common for components of an ADMD's email infrastructure to do 1793 violence to a message, such that a DKIM signature might be rendered 1794 invalid. Hence, users of MUAs that support DKIM signing and/or 1795 verifying need a basis for knowing that their associated email 1796 infrastructure will not break a signature. 1798 9. Other Considerations 1800 9.1. Security Considerations 1802 The security considerations of the DKIM protocol are described in the 1803 DKIM base specification [RFC4871]. 1805 9.2. IANA Considerations 1807 This document has no considerations for IANA. 1809 10. Acknowledgements 1811 TBD 1813 11. Informative References 1815 [I-D.ietf-dkim-ssp] 1816 field, h., Domain, A., error, r., Allman, E., Fenton, J., 1817 Delany, M., and J. Levine, "DomainKeys Identified Mail 1818 (DKIM) Author Domain Signing Practices (ADSP)", 1819 draft-ietf-dkim-ssp-10 (work in progress), May 2009. 1821 [RFC0989] Linn, J. and IAB Privacy Task Force, "Privacy enhancement 1822 for Internet electronic mail: Part I: Message encipherment 1823 and authentication procedures", RFC 989, February 1987. 1825 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 1826 STD 13, RFC 1034, November 1987. 1828 [RFC1848] Crocker, S., Galvin, J., Murphy, S., and N. Freed, "MIME 1829 Object Security Services", RFC 1848, October 1995. 1831 [RFC1991] Atkins, D., Stallings, W., and P. Zimmermann, "PGP Message 1832 Exchange Formats", RFC 1991, August 1996. 1834 [RFC2440] Callas, J., Donnerhacke, L., Finney, H., and R. Thayer, 1835 "OpenPGP Message Format", RFC 2440, November 1998. 1837 [RFC3156] Elkins, M., Del Torto, D., Levien, R., and T. Roessler, 1838 "MIME Security with OpenPGP", RFC 3156, August 2001. 1840 [RFC3164] Lonvick, C., "The BSD Syslog Protocol", RFC 3164, 1841 August 2001. 1843 [RFC3851] Ramsdell, B., "Secure/Multipurpose Internet Mail 1844 Extensions (S/MIME) Version 3.1 Message Specification", 1845 RFC 3851, July 2004. 1847 [RFC4034] Arends, R., Austein, R., Larson, M., Massey, D., and S. 1848 Rose, "Resource Records for the DNS Security Extensions", 1849 RFC 4034, March 2005. 1851 [RFC4686] Fenton, J., "Analysis of Threats Motivating DomainKeys 1852 Identified Mail (DKIM)", RFC 4686, September 2006. 1854 [RFC4870] Delany, M., "Domain-Based Email Authentication Using 1855 Public Keys Advertised in the DNS (DomainKeys)", RFC 4870, 1856 May 2007. 1858 [RFC4871] Allman, E., Callas, J., Delany, M., Libbey, M., Fenton, 1859 J., and M. Thomas, "DomainKeys Identified Mail (DKIM) 1860 Signatures", RFC 4871, May 2007. 1862 [RFC4880] Callas, J., Donnerhacke, L., Finney, H., Shaw, D., and R. 1863 Thayer, "OpenPGP Message Format", RFC 4880, November 2007. 1865 [RFC5155] Laurie, B., Sisson, G., Arends, R., and D. Blacka, "DNS 1866 Security (DNSSEC) Hashed Authenticated Denial of 1867 Existence", RFC 5155, March 2008. 1869 [RFC5321] Klensin, J., "Simple Mail Transfer Protocol", RFC 5321, 1870 October 2008. 1872 [RFC5322] Resnick, P., Ed., "Internet Message Format", RFC 5322, 1873 October 2008. 1875 [RFC5451] Kucherawy, M., "Message Header Field for Indicating 1876 Message Authentication Status", RFC 5451, April 2009. 1878 [RFC5585] Hansen, T., Crocker, D., and P. Hallam-Baker, "DomainKeys 1879 Identified Mail (DKIM) Service Overview", RFC 5585, 1880 July 2009. 1882 [rfc4871-update] 1883 Crocker, D., Ed., "RFC 4871 DomainKeys Identified Mail 1884 (DKIM) Signatures -- Update", 1885 I-D draft-ietf-dkim-rfc4871-errata-03, April 2009. 1887 Appendix A. Migration Strategies 1889 There are three migration occassions worth noting in particular for 1890 DKIM: 1892 1. Migrating from Domain Keys to DKIM. 1894 2. Migrating from a current hash algorithm to a new standardized 1895 hash algorithm. 1897 3. Migrating from a current signing algorithm to a new standardized 1898 signing algorithm. 1900 The case of deploying a new key selector record is described 1901 elsewhere (Section 3.5). 1903 As with any migration, the steps required will be determined by who 1904 is doing the migration and their assessment of 1906 o the users of what they are generating, or 1908 o the providers of what they are consuming. 1910 Signers and verifiers have different considerations. 1912 A.1. Migrating from DomainKeys 1914 DKIM replaces the earlier DomainKeys (DK) specification. Selector 1915 files are mostly compatible between the two specifications. 1917 A.1.1. Signers 1919 A signer that currently signs with DK will go through various stages 1920 as it migrates to using DKIM, not all of which are required for all 1921 signers. The real questions that a signer must ask are: 1923 1. how many receivers or what types of receivers are *only* looking 1924 at the DK signatures and not the DKIM signatures, and 1926 2. how much does the signer care about those receivers? 1928 If no one is looking at the DK signature any more, then it's no 1929 longer necessary to sign with DK. Or if all "large players" are 1930 looking at DKIM in addition to or instead of DK, a signer MAY choose 1931 to stop signing with DK. 1933 With respect to signing policies, a reasonable, initial approach is 1934 to use DKIM signatures in the same way as DomainKeys signatures are 1935 already being used. In particular, the same selectors and DNS Key 1936 Records may be used for both, after verifying that they are 1937 compatible as discussed below. 1939 Each secondary step in all of the following scenarios is to be 1940 prefaced with the gating factor "test, then when comfortable with the 1941 previous step's results, continue". 1943 One migration strategy is to: 1945 o ensure that the current selector DNS key record is compatible with 1946 both DK and DKIM 1948 o sign messages with both DK and DKIM signatures 1950 o when it's decided that DK signatures are no longer necessary, stop 1951 signing with DK 1953 Another migration strategy is to: 1955 o add a new selector DNS key record only for DKIM signatures 1957 o sign messages with both DK (using the old DNS key record) and DKIM 1958 signatures (using the new DNS key record) 1960 o when it's decided that DK signatures are no longer necessary, stop 1961 signing with DK 1963 o eventually remove the old DK selector DNS record 1965 A combined migration strategy is to: 1967 o ensure that the current selector DNS key record is compatible with 1968 both DK and DKIM 1970 o start signing messages with both DK and DKIM signatures 1972 o add a new selector DNS key record for DKIM signatures 1974 o switch the DKIM signatures to use the new selector 1975 o when it's decided that DK signatures are no longer necessary, stop 1976 signing with DK 1978 o eventually remove the old DK selector DNS record 1980 Another migration strategy is to: 1982 o add a new selector DNS key record for DKIM signatures 1984 o do a flash cut and replace the DK signatures with DKIM signatures 1986 o eventually remove the old DK selector DNS record 1988 Another migration strategy is to: 1990 o ensure that the current selector DNS key record is compatible with 1991 both DK and DKIM 1993 o do a flash cut and replace the DK signatures with DKIM signatures 1995 Note that when you have separate key records for DK and DKIM, you can 1996 use the same public key for both. 1998 A.1.1.1. DNS Selector Key Records 2000 The first step in some of the above scenarios is ensuring that the 2001 selector DNS key records are compatible for both DK and DKIM. The 2002 format of the DNS key record was intentionally meant to be backwardly 2003 compatible between the two systems, but not necessarily upwardly 2004 compatible. DKIM has enhanced the DK DNS key record format by adding 2005 several optional parameters, which DK must ignore. However, there is 2006 one critical difference between DK and DKIM DNS key records: the 2007 definitions of the "g" fields: 2009 g= granularity of the key In both DK and DKIM, this is an optional 2010 field that is used to constrain which sending address(es) can 2011 legitimately use this selector. Unfortunately, the treatment of 2012 an empty field ("g=;") is different. DKIM allows wildcards where 2013 DK does not. For DK, an empty field is the same as a missing 2014 value, and is treated as allowing any sending address. For DKIM, 2015 an empty field only matches an empty local part. In DKIM, both a 2016 missing value and "g=*;" mean to allow any sending address. 2018 If your DK DNS key record has an empty "g" field in it ("g=;"), 2019 your best course of action is to modify the record to remove the 2020 empty field. In that way, the DK semantics will remain the same, 2021 and the DKIM semantics will match. 2023 If your DNS key record does not have an empty "g" field in it 2024 ("g=;"), it's probable that the record can be left alone. But your 2025 best course of action would still be to make sure it has a "v" field. 2026 When the decision is made to stop supporting DomainKeys and to only 2027 support DKIM, you MUST verify that the "g" field is compatible with 2028 DKIM, and it SHOULD have "v=DKIM1;" in it. It is highly RECOMMENDED 2029 that if you want to use an empty "g" field in your DKIM selector, you 2030 also include the "v" field. 2032 A.1.1.2. Removing DomainKeys Signatures 2034 The principal use of DomainKeys is at Boundary MTAs. Because no 2035 operational transition is ever instantaneous, it is advisable to 2036 continue performing DomainKeys signing until it is determined that 2037 DomainKeys receive-side support is no longer used, or is sufficiently 2038 reduced. That is, a signer SHOULD add a DKIM signature to a message 2039 that also has a DomainKeys signature and keep it there until you 2040 decide it is deemed no longer useful. The signer may do its 2041 transitions in a straightforward manner, or more gradually. Note 2042 that because digital signatures are not free, there is a cost to 2043 performing both signing algorithms, so signing with both algorithms 2044 should not be needlessly prolonged. 2046 The tricky part is deciding when DK signatures are no longer 2047 necessary. The real questions are: how many DomainKeys verifiers are 2048 there that do *not* also do DKIM verification, which of those are 2049 important, and how can you track their usage? Most of the early 2050 adopters of DK verification have added DKIM verification, but not all 2051 yet. If a verifier finds a message with both DK and DKIM, it may 2052 choose to verify both signatures, or just one or the other. 2054 Many DNS services offer tracking statistics so it can be determined 2055 how often a DNS record has been accessed. By using separate DNS 2056 selector key records for your signatures, you can chart the usage of 2057 your records over time, and watch the trends. An additional 2058 distinguishing factor to track would take into account the verifiers 2059 that verify both the DK and DKIM signatures, and discount those from 2060 counts of DK selector usage. When the number for DK selector access 2061 reaches a low-enough level, that's the time to consider discontinuing 2062 signing with DK. 2064 Note, this level of rigor is not required. It is perfectly 2065 reasonable for a DK signer to decide to follow the "flash cut" 2066 scenario described above. 2068 A.1.2. Verifiers 2070 As a verifier, several issues must be considered: 2072 A.1.2.1. Should DK signature verification be performed? 2074 At the time of writing, there is still a significant number of sites 2075 that are only producing DK signatures. Over time, it is expected 2076 that this number will go to zero, but it may take several years. So 2077 it would be prudent for the foreseeable future for a verifier to look 2078 for and verify both DKIM and DK signatures. 2080 A.1.2.2. Should both DK and DKIM signatures be evaluated on a single 2081 message? 2083 For a period of time, there will be sites that sign with both DK and 2084 DKIM. A verifier receiving a message that has both types of 2085 signatures may verify both signatures, or just one. One disadvantage 2086 of verifying both signatures is that signers will have a more 2087 difficult time deciding how many verifiers are still using their DK 2088 selectors. One transition strategy is to verify the DKIM signature, 2089 then only verify the DK signature if the DKIM verification fails. 2091 A.1.2.3. DNS Selector Key Records 2093 The format of the DNS key record was intentionally meant to be 2094 backwardly compatible between DK and DKIM, but not necessarily 2095 upwardly compatible. DKIM has enhanced the DK DNS key record format 2096 by adding several optional parameters, which DK must ignore. 2097 However, there is one key difference between DK and DKIM DNS key 2098 records: the definitions of the g fields: 2100 g= granularity of the key In both DK and DKIM, this is an optional 2101 field that is used to constrain which sending address(es) can 2102 legitimately use this selector. Unfortunately, the treatment of 2103 an empty field ("g=;") is different. For DK, an empty field is 2104 the same as a missing value, and is treated as allowing any 2105 sending address. For DKIM, an empty field only matches an empty 2106 local part. 2108 v= version of the selector It is recommended that a DKIM selector 2109 have "v=DKIM1;" at its beginning, but it is not required. 2111 If a DKIM verifier finds a selector record that has an empty "g" 2112 field ("g=;") and it does not have a "v" field ("v=DKIM1;") at its 2113 beginning, it is faced with deciding if this record was 2114 1. from a DK signer that transitioned to supporting DKIM but forgot 2115 to remove the "g" field (so that it could be used by both DK and 2116 DKIM verifiers), or 2118 2. from a DKIM signer that truly meant to use the empty "g" field 2119 but forgot to put in the "v" field. It is RECOMMENDED that you 2120 treat such records using the first interpretation, and treat such 2121 records as if the signer did not have a "g" field in the record. 2123 A.2. Migrating Hash Algorithms 2125 [RFC4871] defines the use of two hash algorithms, SHA-1 and SHA-256. 2126 The security of all hash algorithms is constantly under attack, and 2127 SHA-1 has already shown weaknesses as of this writing. Migrating 2128 from SHA-1 to SHA-256 is not an issue, because all verifiers are 2129 already required to support SHA-256. But when it becomes necessary 2130 to replace SHA-256 with a more secure algorithm, there will be a 2131 migratory period. In the following, "NEWHASH" is used to represent a 2132 new hash algorithm. Section 4.1 of [RFC4871] briefly discusses this 2133 scenario. 2135 A.2.1. Signers 2137 As with migrating from DK to DKIM, migrating hash algorithms is 2138 dependent on the signer's best guess as to the utility of continuing 2139 to sign with the older algorithms and the expected support for the 2140 newer algorithm by verifiers. The utility of continuing to sign with 2141 the older algorithms is also based on how broken the existing hash 2142 algorithms are considered and how important that is to the signers. 2144 One strategy is to wait until it's determined that there is a large 2145 enough base of verifiers available that support NEWHASH, and then 2146 flash cut to the new algorithm. 2148 Another strategy is to sign with both the old and new hash algorithms 2149 for a period of time. This is particularly useful for testing the 2150 new code to support the new hash algorithm, as verifiers will 2151 continue to accept the signature for the older hash algorithm and 2152 should ignore any signature that fails because the code is slightly 2153 wrong. Once the signer has determined that the new code is correct 2154 AND it's determined that there is a large enough base of verifiers 2155 available that support NEWHASH, the signer can flash cut to the new 2156 algorithm. 2158 One advantage migrating hash algorithms has is that the selector can 2159 be completely compatible for all hash algorithms. The key selector 2160 has an optional "h=" field that may be used to list the hash 2161 algorithms being used; it also is used to limit the algorithms that a 2162 verifier will accept. If the signer is not currently using the key- 2163 selector "h=" field, no change is required. If the signer is 2164 currently using the key-selector "h=" field, NEWHASH will need to be 2165 added to the list, as in "h=sha256:NEWHASH;". (When the signer is no 2166 longer using sha256, it can be removed from the "h=" list.) 2168 A.2.2. Verifiers 2170 When a new hash algorithm becomes standardized, it is best for a 2171 verifier to start supporting it as quickly as possible. 2173 A.3. Migrating Signing Algorithms 2175 [RFC4871] defines the use of the RSA signing algorithm. Similar to 2176 hashes, signing algorithms are constantly under attack, and when it 2177 becomes necessary to replace RSA with a newer signing algorithm, 2178 there will be a migratory period. In the following, "NEWALG" is used 2179 to represent a new signing algorithm. 2181 A.3.1. Signers 2183 As with the other migration issues discussed above, migrating signing 2184 algorithms is dependent on the signer's best guess as to the utility 2185 of continuing to sign with the older algorithms and the expected 2186 support for the newer algorithm by verifiers. The utility of 2187 continuing to sign with the older algorithms is also based on how 2188 broken the existing signing algorithms are considered and how 2189 important that is to the signers. 2191 As before, the two basic strategies are to 1) wait until there is 2192 sufficient base of verifiers available that support NEWALG and then 2193 do a flash cut to NEWALG, and 2) using a phased approach by signing 2194 with both the old and new algorithms before removing support for the 2195 old algorithm. 2197 It is unlikely that a new algorithm would be able to use the same 2198 public key as "rsa", so using the same selector DNS record for both 2199 algorithms' keys is ruled out. Therefore, in order to use the new 2200 algorithm, a new DNS selector record would need to be deployed in 2201 parallel with the existing DNS selector record for the existing 2202 algorithm. The new DNS selector record would specify a different 2203 "k=" value to reflect the use of NEWALG. 2205 A.3.2. Verifiers 2207 When a new hash algorithm becomes standardized, it is best for a 2208 verifier to start supporting it as quickly as possible. 2210 Appendix B. General Coding Criteria for Cryptographic Applications 2212 NOTE: This section could possibly be changed into a reference to 2213 something else, such as another rfc. 2215 Correct implementation of a cryptographic algorithm is a necessary 2216 but not a sufficient condition for the coding of cryptographic 2217 applications. Coding of cryptographic libraries requires close 2218 attention to security considerations that are unique to cryptographic 2219 applications. 2221 In addition to the usual security coding considerations, such as 2222 avoiding buffer or integer overflow and underflow, implementers 2223 should pay close attention to management of cryptographic private 2224 keys and session keys, ensuring that these are correctly initialized 2225 and disposed of. 2227 Operating system mechanisms that permit the confidentiality of 2228 private keys to be protected against other processes should be used 2229 when available. In particular, great care must be taken when 2230 releasing memory pages to the operating system to ensure that private 2231 key information is not disclosed to other processes. 2233 Certain implementations of public key algorithms such as RSA may be 2234 vulnerable to a timing analysis attack. 2236 Support for cryptographic hardware providing key management 2237 capabilities is strongly encouraged. In addition to offering 2238 performance benefits, many cryptographic hardware devices provide 2239 robust and verifiable management of private keys. 2241 Fortunately appropriately designed and coded cryptographic libraries 2242 are available for most operating system platforms under license terms 2243 compatible with commercial, open source and free software license 2244 terms. Use of standard cryptographic libraries is strongly 2245 encouraged. These have been extensively tested, reduce development 2246 time and support a wide range of cryptographic hardware. 2248 Authors' Addresses 2250 Tony Hansen 2251 AT&T Laboratories 2252 200 Laurel Ave. South 2253 Middletown, NJ 07748 2254 USA 2256 Email: tony+dkimov@maillennium.att.com 2258 Ellen Siegel 2260 Email: dkim@esiegel.net 2262 Phillip Hallam-Baker 2263 Default Deny Security, Inc. 2265 Email: phillip@hallambaker.com 2267 Dave Crocker 2268 Brandenburg InternetWorking 2269 675 Spruce Dr. 2270 Sunnyvale, CA 94086 2271 USA 2273 Email: dcrocker@bbiw.net