idnits 2.17.1 draft-ietf-dane-srv-14.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document date (April 23, 2015) is 3290 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-16) exists of draft-ietf-dane-ops-07 == Outdated reference: A later version (-19) exists of draft-ietf-dane-smtp-with-dane-15 ** Obsolete normative reference: RFC 6125 (Obsoleted by RFC 9525) == Outdated reference: A later version (-11) exists of draft-ietf-xmpp-dna-10 Summary: 1 error (**), 0 flaws (~~), 5 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 DNS-Based Authentication of Named Entities (DANE) T. Finch 3 Internet-Draft University of Cambridge 4 Intended status: Standards Track M. Miller 5 Expires: October 25, 2015 Cisco Systems, Inc. 6 P. Saint-Andre 7 &yet 8 April 23, 2015 10 Using DNS-Based Authentication of Named Entities (DANE) TLSA Records 11 with SRV Records 12 draft-ietf-dane-srv-14 14 Abstract 16 The DANE specification (RFC 6698) describes how to use TLSA resource 17 records secured by DNSSEC (RFC 4033) to associate a server's 18 connection endpoint with its TLS certificate (thus enabling 19 administrators of domain names to specify the keys used in that 20 domain's TLS servers). However, application protocols that use SRV 21 records (RFC 2782) to indirectly name the target server connection 22 endpoints for a service domain cannot apply the rules from RFC 6698. 23 Therefore this document provides guidelines that enable such 24 protocols to locate and use TLSA records. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on October 25, 2015. 43 Copyright Notice 45 Copyright (c) 2015 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 61 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 62 3. DNS Checks . . . . . . . . . . . . . . . . . . . . . . . . . 4 63 3.1. SRV Query . . . . . . . . . . . . . . . . . . . . . . . . 4 64 3.2. Address Queries . . . . . . . . . . . . . . . . . . . . . 5 65 3.3. TLSA Queries . . . . . . . . . . . . . . . . . . . . . . 5 66 3.4. Impact on TLS Usage . . . . . . . . . . . . . . . . . . . 6 67 4. TLS Checks . . . . . . . . . . . . . . . . . . . . . . . . . 6 68 4.1. SRV Records Only . . . . . . . . . . . . . . . . . . . . 6 69 4.2. TLSA Records . . . . . . . . . . . . . . . . . . . . . . 7 70 5. Guidance for Protocol Authors . . . . . . . . . . . . . . . . 7 71 6. Guidance for Server Operators . . . . . . . . . . . . . . . . 8 72 7. Guidance for Application Developers . . . . . . . . . . . . . 9 73 8. Internationalization Considerations . . . . . . . . . . . . . 9 74 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 75 10. Security Considerations . . . . . . . . . . . . . . . . . . . 9 76 10.1. Mixed Security Status . . . . . . . . . . . . . . . . . 9 77 10.2. Certificate Subject Name Matching . . . . . . . . . . . 9 78 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 79 11.1. Normative References . . . . . . . . . . . . . . . . . . 10 80 11.2. Informative References . . . . . . . . . . . . . . . . . 11 81 Appendix A. Examples . . . . . . . . . . . . . . . . . . . . . . 11 82 A.1. IMAP . . . . . . . . . . . . . . . . . . . . . . . . . . 12 83 A.2. XMPP . . . . . . . . . . . . . . . . . . . . . . . . . . 12 84 Appendix B. Rationale . . . . . . . . . . . . . . . . . . . . . 13 85 Appendix C. Acknowledgements . . . . . . . . . . . . . . . . . . 14 86 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 88 1. Introduction 90 The base DANE specification [RFC6698] describes how to use TLSA 91 resource records secured by DNSSEC [RFC4033] to associate a target 92 server's connection endpoint with its TLS certificate (thus enabling 93 administrators of domain names to specify the keys used in that 94 domain's TLS servers). Some application protocols locate connection 95 endpoints indirectly via SRV records [RFC2782]. As a result of this 96 indirection, the rules specified in [RFC6698] cannot be directly 97 applied to such application protocols. (Rules for SMTP [RFC5321], 98 which uses MX resource records instead of SRV records, are described 99 in [I-D.ietf-dane-smtp-with-dane].) 101 This document describes how to use DANE TLSA records with SRV 102 records. To summarize: 104 o We rely on DNSSEC to secure SRV records that map the desired 105 service, transport protocol, and service domain to the 106 corresponding target server connection endpoints (i.e., the target 107 server host names and port numbers returned in the SRV records for 108 that service type). 110 o Although in accordance with [RFC2782] a service domain can 111 advertise a number of SRV records (some of which might map to 112 connection endpoints that do not support TLS), the intent of this 113 specification is for a client to securely discover connection 114 endpoints that support TLS. 116 o The TLSA records for each connection endpoint are located using 117 the transport protocol, port number, and host name for the target 118 server (not the service domain). 120 o When DNSSEC-validated TLSA records are published for a given 121 connection endpoint, clients always use TLS when connecting (even 122 if the connection endpoint supports cleartext communication). 124 o If there is at least one usable TLSA record for a given connection 125 endpoint, the connection endpoint's TLS certificate or public key 126 needs to match at least one of those usable TLSA records. 128 o If there are no usable TLSA records for a given connection 129 endpoint, the target server host name is used as one of the 130 acceptable reference identifiers, as described in [RFC6125]. 131 Other reference identifiers might arise through CNAME expansion of 132 either the service domain or target server host name, as detailed 133 in [I-D.ietf-dane-ops]. 135 o If there are no usable TLSA records for any connection endpoint 136 (and thus the client cannot securely discover a connection 137 endpoint that supports TLS), the client's behavior is a matter for 138 the application protocol or client implementation; this might 139 involve a fallback to non-DANE behavior using the public key 140 infrastructure [RFC5280]. 142 2. Terminology 144 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 145 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 146 "OPTIONAL" in this memo are to be interpreted as described in 147 [RFC2119]. 149 This draft uses the definitions for "secure", "insecure", "bogus", 150 and "indeterminate" from Section 4.3 of [RFC4035]. This draft uses 151 the acronyms from [RFC7218] for the values of TLSA fields where 152 appropriate. 154 Additionally, this document uses the following terms: 156 connection endpoint: A tuple of a fully qualified DNS host name, 157 transport protocol, and port number that a client uses to 158 establish a connection to the target server. 160 service domain: The fully qualified DNS domain name that identifies 161 an application service; corresponds to the term "source domain" 162 from [RFC6125]. 164 This document uses the term "target server host name" in place of the 165 term "derived domain" from the CertID specification [RFC6125]. 167 3. DNS Checks 169 3.1. SRV Query 171 When the client makes an SRV query, a successful result will 172 typically be a list of one or more SRV records (or possibly a chain 173 of CNAME / DNAME aliases leading to such a list). 175 NOTE: Implementers need to be aware that unsuccessful results can 176 occur because of various DNS-related errors; guidance on avoiding 177 downgrade attacks can be found in Section 2.1 of 178 [I-D.ietf-dane-smtp-with-dane]. 180 For this specification to apply, the entire chain of DNS RRset(s) 181 returned MUST be "secure" according to DNSSEC validation (Section 5 182 of [RFC4035]). In the case where the answer is obtained via a chain 183 of CNAME and/or DNAME aliases, the whole chain of CNAME and DNAME 184 RRsets MUST also be secure. 186 If the SRV lookup fails because the RRset is "bogus" (or the lookup 187 fails for reasons other than no records), the client MUST abort its 188 attempt to connect to the desired service. If the lookup result is 189 "insecure" (or no SRV records exist), this protocol does not apply 190 and the client SHOULD fall back to its non-DNSSEC, non-DANE (and 191 possibly non-SRV) behavior. 193 When the lookup returns a "secure" RRset (possibly via a chain of 194 "secure" CNAME/DNAME records), the client now has an authentic list 195 of target server connection endpoints with weight and priority 196 values. It performs server ordering and selection using the weight 197 and priority values without regard to the presence or absence of 198 DNSSEC or TLSA records. It also takes note of the DNSSEC validation 199 status of the SRV response for use when checking certificate names 200 (see Section 4). The client can then proceed to making address 201 queries on the target server host names as described in the following 202 section. 204 3.2. Address Queries 206 For each SRV target server connnection endpoint, the client makes A 207 and/or AAAA queries, performs DNSSEC validation on the address (A or 208 AAAA) response, and continues as follows based on the results: 210 o If a returned RRSet is "secure", the client MUST perform a TLSA 211 query for that target server connection endpoint as described in 212 the next section. 214 o If no returned RRsets are "secure", the client MUST NOT perform a 215 TLSA query for that target server connection endpoint; the TLSA 216 query will most likely fail or produce spurious results. 218 o If the address record lookup fails (this a validation status of 219 either "bogus" or "indeterminate"), the client MUST NOT connect to 220 this connection endpoint; instead it uses the next most 221 appropriate SRV target. This mitigates against downgrade attacks. 223 3.3. TLSA Queries 225 The client SHALL construct the TLSA query name as described in 226 Section 3 of [RFC6698], based on the fields from the SRV record: the 227 port number from the SRV RDATA, the transport protocol from the SRV 228 query name, and the TLSA base domain from the SRV target server host 229 name. 231 For example, the following SRV record for IMAP (see [RFC6186]): 233 _imap._tcp.example.com. 86400 IN SRV 10 0 9143 imap.example.net. 235 leads to the TLSA query shown below: 237 _9143._tcp.imap.example.net. IN TLSA ? 239 3.4. Impact on TLS Usage 241 The client SHALL determine if the TLSA records returned in the 242 previous step are usable according to Section 4.1 of [RFC6698]. This 243 affects the use of TLS as follows: 245 o If the TLSA response is "secure" and usable, then the client MUST 246 use TLS when connecting to the target server. The TLSA records 247 are used when validating the server's certificate as described in 248 Section 4. 250 o If the TLSA response is "bogus" or "indeterminate" (or the lookup 251 fails for reasons other than no records), then the client MUST NOT 252 connect to the target server (the client can still use other SRV 253 targets). 255 o If the TLSA response is "insecure" (or no TLSA records exist), 256 then the client SHALL proceed as if the target server had no TLSA 257 records. It MAY connect to the target server with or without TLS, 258 subject to the policies of the application protocol or client 259 implementation. 261 4. TLS Checks 263 When connecting to a server, the client MUST use TLS if the responses 264 to the SRV and TLSA queries were "secure" as described above. The 265 rules described in the next two sections apply to such secure 266 responses; Section 4.2 where there is at least one usable TLSA 267 record, and Section 4.1 otherwise. 269 4.1. SRV Records Only 271 If the client received zero usable TLSA certificate associations, it 272 SHALL validate the server's TLS certificate using the normal PKIX 273 rules [RFC5280] or protocol-specific rules (e.g., following 274 [RFC6125]) without further input from the TLSA records. In this 275 case, the client uses the information in the server certificate and 276 the DNSSEC validation status of the SRV query in its authentication 277 checks. It SHOULD use the Server Name Indication extension (TLS SNI) 278 [RFC6066] or its functional equivalent in the relevant application 279 protocol (e.g., in XMPP [RFC6120] this is the 'to' address of the 280 initial stream header). The preferred name SHALL be chosen as 281 follows, and the client SHALL verify the identity asserted by the 282 server's certificate according to Section 6 of [RFC6125], using a 283 list of reference identifiers constructed as follows (note again that 284 in RFC 6125 the terms "source domain" and "derived domain" to refer 285 to the same things as "service domain" and "target server host name" 286 in this document). The examples below assume a service domain of 287 "im.example.com" and a target server host name of 288 "xmpp23.hosting.example.net". 290 SRV is insecure: The reference identifiers SHALL include the service 291 domain and MUST NOT include the SRV target server host name (e.g., 292 include "im.example.com" but not "xmpp23.hosting.example.net"). 293 The service domain is the preferred name for TLS SNI or its 294 equivalent. 296 SRV is secure: The reference identifiers SHALL include both the 297 service domain and the SRV target server host name (e.g., include 298 both "im.example.com" and "xmpp23.hosting.example.net"). The 299 target server host name is the preferred name for TLS SNI or its 300 equivalent. 302 In the latter case, the client will accept either identity to ensure 303 compatibility with servers that support this specification as well as 304 servers that do not support this specification. 306 4.2. TLSA Records 308 If the client received one or more usable TLSA certificate 309 associations, it SHALL process them as described in Section 2.1 of 310 [RFC6698]. 312 If the TLS server's certificate -- or the public key of the server's 313 certificate -- matches a usable TLSA record with Certificate Usage 314 "DANE-EE", the client MUST ignore validation checks from [RFC5280] 315 and reference identifier checks from [RFC6125]. The information in 316 such a TLSA record supersedes the non-key information in the 317 certificate. 319 5. Guidance for Protocol Authors 321 This document describes how to use DANE with application protocols in 322 which target servers are discovered via SRV records. Although this 323 document attempts to provide generic guidance applying to all such 324 protocols, additional documents for particular application protocols 325 could cover related topics, such as: 327 o Fallback logic in the event that a client is unable to connect 328 securely to a target server by following the procedures defined in 329 this document. 331 o How clients ought to behave if they do not support SRV lookups, or 332 if clients that support SRV lookups encounter service domains that 333 do not offer SRV records. 335 o Whether the application protocol has a functional equivalent for 336 TLS SNI that is preferred within that protocol. 338 o Use of SRV records with additional discovery technologies, such as 339 the use of both SRV records and NAPTR records [RFC3403] for 340 transport selection in the Session Initiation Protocol (SIP). 342 For example, [I-D.ietf-xmpp-dna] covers such topics for the 343 Extensible Messaging and Presence Protocol (XMPP). 345 6. Guidance for Server Operators 347 To conform to this specification, the published SRV records and 348 subsequent address (A and AAAA) records MUST be secured with DNSSEC. 349 There SHOULD also be at least one TLSA record published that 350 authenticates the server's certificate. 352 When using TLSA records with Certificate Usage "DANE-EE", it is not 353 necessary for the deployed certificate to contain an identifier for 354 either the source domain or target server host name. However, 355 operators need to be aware that servers relying solely on validation 356 using Certificate Usage "DANE-EE" TLSA records might prevent clients 357 that do not support this specification from successfully connecting 358 with TLS. 360 For TLSA records with Certificate Usage types other than "DANE-EE", 361 the certificate(s) MUST contain an identifier that matches: 363 o the service domain name (the "source domain" in [RFC6125] terms, 364 which is the SRV query domain); and/or 366 o the target server host name (the "derived domain" in [RFC6125] 367 terms, which is the SRV target host name). 369 Servers that support multiple service domains (i.e., so-called 370 "multi-tenanted environments") can implement the Transport Layer 371 Security Server Name Indication (TLS SNI) [RFC6066] or its functional 372 equivalent to determine which certificate to offer. Clients that do 373 not support this specification will indicate a preference for the 374 service domain name, while clients that support this specification 375 will indicate the target server host name. However, the server 376 determines what certificate to present in the TLS handshake; e.g., 377 the presented certificate might only authenticate the target server 378 host name. 380 7. Guidance for Application Developers 382 Developers of application clients that depend on DANE-SRV often would 383 like to prepare as quickly as possible for making a connection to the 384 intended service, thus reducing the wait time for end users. To make 385 this optimization possible, a DNS library might perform the SRV 386 queries, address queries, and TLSA queries in parallel. (Because a 387 TLSA record can be ignored if it turns out that the address record on 388 which it depends is not secure, performing the TLSA queries in 389 parallel with the SRV queries and address queries is not harmful from 390 a security perspective and can yield some operational benefits.) 392 8. Internationalization Considerations 394 If any of the DNS queries are for an internationalized domain name, 395 then they need to use the A-label form [RFC5890]. 397 9. IANA Considerations 399 No IANA action is required. 401 10. Security Considerations 403 10.1. Mixed Security Status 405 We do not specify that all of the target server connection endpoints 406 for a service domain need to be consistent in whether they have or do 407 not have TLSA records. This is so that partial or incremental 408 deployment does not break the service. Different levels of 409 deployment are likely if a service domain has a third-party fallback 410 server, for example. 412 The SRV sorting rules are unchanged; in particular they have not been 413 altered in order to prioritize secure connection endpoints over 414 insecure connection endpoints. If a site wants to be secure it needs 415 to deploy this protocol completely; a partial deployment is not 416 secure and we make no special effort to support it. 418 10.2. Certificate Subject Name Matching 420 Section 4 of the TLSA specification [RFC6698] leaves the details of 421 checking names in certificates to higher level application protocols, 422 though it suggests the use of [RFC6125]. 424 Name checks are not necessary if the matching TLSA record is of 425 Certificate Usage "DANE-EE". Because such a record identifies the 426 specific certificate (or public key of the certificate), additional 427 checks are superfluous and potentially conflicting. 429 Otherwise, while DNSSEC provides a secure binding between the server 430 name and the TLSA record, and the TLSA record provides a binding to a 431 certificate, this latter step can be indirect via a chain of 432 certificates. For example, a Certificate Usage "PKIX-TA" TLSA record 433 only authenticates the CA that issued the certificate, and third 434 parties can obtain certificates from the same CA. Therefore, clients 435 need to check whether the server's certificate matches one of the 436 expected reference identifiers to ensure that the certificate was 437 issued by the CA to the server the client expects (naturally, this is 438 in addition to standard certificate-related checks as specified in 439 [RFC5280], including but not limited to certificate syntax, 440 certificate extensions such as name constraints and extended key 441 usage, and handling of certification paths). 443 11. References 445 11.1. Normative References 447 [I-D.ietf-dane-ops] 448 Dukhovni, V. and W. Hardaker, "Updates to and Operational 449 Guidance for the DANE Protocol", draft-ietf-dane-ops-07 450 (work in progress), October 2014. 452 [I-D.ietf-dane-smtp-with-dane] 453 Dukhovni, V. and W. Hardaker, "SMTP security via 454 opportunistic DANE TLS", draft-ietf-dane-smtp-with-dane-15 455 (work in progress), March 2015. 457 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 458 Requirement Levels", BCP 14, RFC 2119, March 1997. 460 [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for 461 specifying the location of services (DNS SRV)", RFC 2782, 462 February 2000. 464 [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. 465 Rose, "DNS Security Introduction and Requirements", RFC 466 4033, March 2005. 468 [RFC4035] Arends, R., Austein, R., Larson, M., Massey, D., and S. 469 Rose, "Protocol Modifications for the DNS Security 470 Extensions", RFC 4035, March 2005. 472 [RFC5280] Cooper, D., Santesson, S., Farrell, S., Boeyen, S., 473 Housley, R., and W. Polk, "Internet X.509 Public Key 474 Infrastructure Certificate and Certificate Revocation List 475 (CRL) Profile", RFC 5280, May 2008. 477 [RFC5890] Klensin, J., "Internationalized Domain Names for 478 Applications (IDNA): Definitions and Document Framework", 479 RFC 5890, August 2010. 481 [RFC6066] Eastlake, D., "Transport Layer Security (TLS) Extensions: 482 Extension Definitions", RFC 6066, January 2011. 484 [RFC6125] Saint-Andre, P. and J. Hodges, "Representation and 485 Verification of Domain-Based Application Service Identity 486 within Internet Public Key Infrastructure Using X.509 487 (PKIX) Certificates in the Context of Transport Layer 488 Security (TLS)", RFC 6125, March 2011. 490 [RFC6698] Hoffman, P. and J. Schlyter, "The DNS-Based Authentication 491 of Named Entities (DANE) Transport Layer Security (TLS) 492 Protocol: TLSA", RFC 6698, August 2012. 494 [RFC7218] Gudmundsson, O., "Adding Acronyms to Simplify 495 Conversations about DNS-Based Authentication of Named 496 Entities (DANE)", RFC 7218, April 2014. 498 11.2. Informative References 500 [I-D.ietf-xmpp-dna] 501 Saint-Andre, P., Miller, M., and P. Hancke, "Domain Name 502 Associations (DNA) in the Extensible Messaging and 503 Presence Protocol (XMPP)", draft-ietf-xmpp-dna-10 (work in 504 progress), March 2015. 506 [RFC3403] Mealling, M., "Dynamic Delegation Discovery System (DDDS) 507 Part Three: The Domain Name System (DNS) Database", RFC 508 3403, October 2002. 510 [RFC5321] Klensin, J., "Simple Mail Transfer Protocol", RFC 5321, 511 October 2008. 513 [RFC6120] Saint-Andre, P., "Extensible Messaging and Presence 514 Protocol (XMPP): Core", RFC 6120, March 2011. 516 [RFC6186] Daboo, C., "Use of SRV Records for Locating Email 517 Submission/Access Services", RFC 6186, March 2011. 519 Appendix A. Examples 521 In the following, most of the DNS resource data is elided for 522 simplicity. 524 A.1. IMAP 526 ; mail domain 527 _imap._tcp.example.com. SRV 10 0 9143 imap.example.net. 528 example.com. RRSIG SRV ... 530 ; target server host name 531 imap.example.net. A 192.0.2.1 532 imap.example.net. RRSIG A ... 534 imap.example.net. AAAA 2001:db8:212:8::e:1 535 imap.example.net. RRSIG ... 537 ; TLSA resource record 538 _9143._tcp.imap.example.net. TLSA ... 539 _9143._tcp.imap.example.net. RRSIG TLSA ... 541 Mail messages received for addresses at example.com are retrieved via 542 IMAP at imap.example.net. Connections to imap.example.net port 9143 543 that use STARTTLS will get a server certificate that authenticates 544 the name imap.example.net. 546 A.2. XMPP 548 ; XMPP domain 549 _xmpp-client._tcp.example.com. SRV 1 0 5222 im.example.net. 550 _xmpp-client._tcp.example.com. RRSIG SRV ... 552 ; target server host name 553 im.example.net. A 192.0.2.3 554 im.example.net. RRSIG A ... 556 im.example.net. AAAA 2001:db8:212:8::e:4 557 im.example.net. RRSIG AAAA ... 559 ; TLSA resource record 560 _5222._tcp.im.example.net. TLSA ... 561 _5222._tcp.im.example.net. RRSIG TLSA ... 563 XMPP sessions for addresses at example.com are established at 564 im.example.net. Connections to im.example.net port 5222 that use 565 STARTTLS will get a server certificate that authenticates the name 566 im.example.net. 568 Appendix B. Rationale 570 The long-term goal of this specification is to settle on TLS 571 certificates that verify the target server host name rather than the 572 service domain, since this is more convenient for servers hosting 573 multiple domains (so-called "multi-tenanted environments") and scales 574 up more easily to larger numbers of service domains. 576 There are a number of other reasons for doing it this way: 578 o The certificate is part of the server configuration, so it makes 579 sense to associate it with the server host name rather than the 580 service domain. 582 o In the absence of TLS SNI, if the certificate identifies the 583 target server host name then it does not need to list all the 584 possible service domains. 586 o When the server certificate is replaced it is much easier if there 587 is one part of the DNS that needs updating to match, instead of an 588 unbounded number of hosted service domains. 590 o The same TLSA records work with this specification, and with 591 direct connections to the connection endpoint in the style of 592 [RFC6698]. 594 o Some application protocols, such as SMTP, allow a client to 595 perform transactions with multiple service domains in the same 596 connection. It is not in general feasible for the client to 597 specify the service domain using TLS SNI when the connection is 598 established, and the server might not be able to present a 599 certificate that authenticates all possible service domains. See 600 [I-D.ietf-dane-smtp-with-dane] for details. 602 o It is common for SMTP servers to act in multiple roles, for 603 example as outgoing relays or as incoming MX servers, depending on 604 the client identity. It is simpler if the server can present the 605 same certificate regardless of the role in which it is to act. 606 Sometimes the server does not know its role until the client has 607 authenticated, which usually occurs after TLS has been 608 established. See [I-D.ietf-dane-smtp-with-dane] for details. 610 This specification does not provide an option to put TLSA records 611 under the service domain because that would add complexity without 612 providing any benefit, and security protocols are best kept simple. 613 As described above, there are real-world cases where authenticating 614 the service domain cannot be made to work, so there would be 615 complicated criteria for when service domain TLSA records might be 616 used and when they cannot. This is all avoided by putting the TLSA 617 records under the target server host name. 619 The disadvantage is that clients which do not complete DNSSEC 620 validation must, according to [RFC6125] rules, check the server 621 certificate against the service domain, since they have no other way 622 to authenticate the server. This means that SNI support or its 623 functional equivalent is necessary for backward compatibility. 625 Appendix C. Acknowledgements 627 Thanks to Mark Andrews for arguing that authenticating the target 628 server host name is the right thing, and that we ought to rely on 629 DNSSEC to secure the SRV lookup. Thanks to Stephane Bortzmeyer, 630 James Cloos, Viktor Dukhovni, Ned Freed, Olafur Gudmundsson, Paul 631 Hoffman, Phil Pennock, Hector Santos, Jonas Schneider, and Alessandro 632 Vesely for helpful suggestions. 634 Carl Wallace completed an insightful review on behalf of the Security 635 Directorate. 637 Ben Campbell, Brian Haberman, and Alvaro Retana provided helpful 638 feedback during IESG review. 640 The authors gratefully acknowledge the assistance of Olafur 641 Gudmundsson and Warren Kumari as the working group chairs and Stephen 642 Farrell as the sponsoring Area Director. 644 Peter Saint-Andre wishes to acknowledge Cisco Systems, Inc., for 645 employing him during his work on earlier versions of this document. 647 Authors' Addresses 649 Tony Finch 650 University of Cambridge Computing Service 651 New Museums Site 652 Pembroke Street 653 Cambridge CB2 3QH 654 ENGLAND 656 Phone: +44 797 040 1426 657 Email: dot@dotat.at 658 URI: http://dotat.at/ 659 Matthew Miller 660 Cisco Systems, Inc. 661 1899 Wynkoop Street, Suite 600 662 Denver, CO 80202 663 USA 665 Email: mamille2@cisco.com 667 Peter Saint-Andre 668 &yet 670 Email: peter@andyet.com 671 URI: https://andyet.com/