idnits 2.17.1 draft-ietf-uta-email-deep-09.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document updates RFC5068, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC2595, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC3501, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC1939, but the abstract doesn't seem to mention this, which it should. -- The draft header indicates that this document updates RFC3464, but the abstract doesn't seem to mention this, which it should. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year (Using the creation date from RFC1939, updated by this document, for RFC5378 checks: 1995-05-15) -- 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 (September 12, 2017) is 2416 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) ** Obsolete normative reference: RFC 3501 (Obsoleted by RFC 9051) ** Obsolete normative reference: RFC 5246 (Obsoleted by RFC 8446) ** Obsolete normative reference: RFC 7525 (Obsoleted by RFC 9325) == Outdated reference: A later version (-21) exists of draft-ietf-uta-mta-sts-09 -- Obsolete informational reference (is this intentional?): RFC 4346 (Obsoleted by RFC 5246) -- Obsolete informational reference (is this intentional?): RFC 6125 (Obsoleted by RFC 9525) Summary: 3 errors (**), 0 flaws (~~), 2 warnings (==), 9 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Moore 3 Internet-Draft Windrock, Inc. 4 Updates: 1939, 2595, 3464, 3501, 5068, C. Newman 5 6186, 6409 (if approved) Oracle 6 Intended status: Standards Track September 12, 2017 7 Expires: March 16, 2018 9 Cleartext Considered Obsolete: Use of TLS for Email Submission and 10 Access 11 draft-ietf-uta-email-deep-09 13 Abstract 15 This specification outlines current recommendations for use of 16 Transport Layer Security (TLS) to provide confidentiality of email 17 traffic between a mail user agent (MUA) and a mail submission or mail 18 access server. 20 Status of This Memo 22 This Internet-Draft is submitted in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at https://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on March 16, 2018. 37 Copyright Notice 39 Copyright (c) 2017 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents 44 (https://trustee.ietf.org/license-info) in effect on the date of 45 publication of this document. Please review these documents 46 carefully, as they describe your rights and restrictions with respect 47 to this document. Code Components extracted from this document must 48 include Simplified BSD License text as described in Section 4.e of 49 the Trust Legal Provisions and are provided without warranty as 50 described in the Simplified BSD License. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 55 2. Conventions and Terminology Used in This Document . . . . . . 3 56 3. Implicit TLS . . . . . . . . . . . . . . . . . . . . . . . . 4 57 3.1. Implicit TLS for POP . . . . . . . . . . . . . . . . . . 5 58 3.2. Implicit TLS for IMAP . . . . . . . . . . . . . . . . . . 5 59 3.3. Implicit TLS for SMTP Submission . . . . . . . . . . . . 5 60 3.4. Implicit TLS Connection Closure for POP, IMAP and SMTP 61 Submission . . . . . . . . . . . . . . . . . . . . . . . 6 62 4. Recommendations for Use of TLS by Mail Access 63 Services and Message Submission Services . . . . . . . . . . 6 64 4.1. Deprecation of Services Using Cleartext and TLS Versions 65 < 1.1 . . . . . . . . . . . . . . . . . . . . . . . . . . 8 66 4.2. Mail Server Use of Client Certificate Authentication . . 8 67 4.3. Recording TLS Cipher Suite in Received Header . . . . . . 9 68 4.4. TLS Server Certificate Requirements . . . . . . . . . . . 9 69 4.5. Recommended DNS records for mail protocol servers . . . . 10 70 4.5.1. MX records . . . . . . . . . . . . . . . . . . . . . 10 71 4.5.2. SRV records . . . . . . . . . . . . . . . . . . . . . 10 72 4.5.3. DNSSEC . . . . . . . . . . . . . . . . . . . . . . . 10 73 4.5.4. TLSA records . . . . . . . . . . . . . . . . . . . . 10 74 4.6. Changes to Internet Facing Servers . . . . . . . . . . . 10 75 5. Recommendations for use of TLS by Mail User Agents . . . . . 10 76 5.1. Use of SRV records in Establishing Configuration . . . . 12 77 5.2. Minimum Confidentiality Level . . . . . . . . . . . . . . 13 78 5.3. Certificiate Validation . . . . . . . . . . . . . . . . . 14 79 5.4. Certificate Pinning . . . . . . . . . . . . . . . . . . . 14 80 5.5. Client Certificate Authentication . . . . . . . . . . . . 15 81 6. Considerations related to Anti-Virus/Anti-Spam Software and 82 Services . . . . . . . . . . . . . . . . . . . . . . . . . . 15 83 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 15 84 7.1. POP3S Port Registration Update . . . . . . . . . . . . . 16 85 7.2. IMAPS Port Registration Update . . . . . . . . . . . . . 16 86 7.3. Submissions Port Registration . . . . . . . . . . . . . . 16 87 8. Security Considerations . . . . . . . . . . . . . . . . . . . 17 88 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 17 89 9.1. Normative References . . . . . . . . . . . . . . . . . . 17 90 9.2. Informative References . . . . . . . . . . . . . . . . . 19 91 Appendix A. Design Considerations . . . . . . . . . . . . . . . 20 92 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 21 93 Appendix C. Acknowledgements . . . . . . . . . . . . . . . . . . 27 94 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 27 96 1. Introduction 98 Software that provides email service via Internet Message Access 99 Protocol (IMAP) [RFC3501], Post Office Protocol (POP) [RFC1939] and/ 100 or Simple Mail Transfer Protocol (SMTP) Submission [RFC6409] usually 101 has Transport Layer Security (TLS) [RFC5246] support but often does 102 not use it in a way that maximizes end-user confidentiality. This 103 specification describes current recommendations for use of TLS in 104 interactions between Mail User Agents and Mail Access Services, and 105 between Mail User Agents and Mail Submission Services. 107 In brief, this memo now recommends that: 109 o TLS version 1.2 or greater be used for all traffic between mail 110 user agents (MUAs) and mail submission servers, and also between 111 MUAs and mail access servers. 113 o MUAs and mail service providers discourage use of cleartext 114 protocols for mail access and mail submission, and deprecate use 115 of cleartext protocols for these purposes as soon as practicable. 117 o Use of "Implicit TLS" on ports reserved for that purpose, in 118 preference to STARTTLS on a port that otherwise supports 119 cleartext. 121 This memo does not address use of TLS with SMTP for message relay 122 (where Message Submission [RFC6409] does not apply). Improved use of 123 TLS with SMTP for message relay requires a different approach. One 124 approach to address that topic is described in [RFC7672]; another is 125 in [I-D.ietf-uta-mta-sts]. 127 The recommendations in this memo do not replace the functionality of, 128 and are not intended as a substitute for, end-to-end encryption of 129 electronic mail. 131 2. Conventions and Terminology Used in This Document 133 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 134 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 135 document are to be interpreted as described in [RFC2119]. 137 The term "Implicit TLS" refers to the automatic negotiation of TLS 138 whenever a TCP connection is made on a particular TCP port that is 139 used exclusively by that server for TLS connections. The term 140 "Implicit TLS" is intended to contrast with use of STARTTLS and 141 similar commands in POP, IMAP, SMTP message submission, and other 142 protocols, that are used by client and server to explicitly negotiate 143 TLS on an established cleartext TCP connection. 145 The term "Mail Access Services" includes POP, IMAP and any other 146 protocol used to access or modify received messages, or to access or 147 modify a mail user's account configuration. 149 "Mail Submission Service" refers to the use of the protocol specified 150 in [RFC6409] (or one of its predecessors or successors) for 151 submission of outgoing messages for delivery to recipients. 153 The term "Mail Service Provider" (MSP) refers to a provider of Mail 154 Access Services and/or Mail Submission Services. 156 The term "Mail Account" refers to a user's identity with a Mail 157 Service Provider, that user's authentication credentials, any user 158 email that is stored by the MSP, and any other per-user configuration 159 information maintained by the MSP (for example, spam filtering 160 instructions). Most Mail User Agents (MUAs) support the ability to 161 access multiple Mail Accounts. 163 For each account that an MUA accesses on its user's behalf, it must 164 have the server names, ports, authentication credentials, and other 165 configuration information specified by the user. This information 166 which is used by the MUA is referred to as "Mail Account 167 Configuration" 169 This specification expresses syntax using the Augmented Backus-Naur 170 Form (ABNF) as described in [RFC5234], including the core rules in 171 Appendix B and rules from [RFC5322]. 173 3. Implicit TLS 175 Previous standards for use of email protocols with TLS used the 176 STARTTLS mechanism: [RFC2595], [RFC3207], and [RFC3501]. With 177 STARTTLS, the client establishes a cleartext application session and 178 determines whether to issue a STARTTLS command based on server 179 capabilities and client configuration. If the client issues a 180 STARTTLS command, a TLS handshake follows that can upgrade the 181 connection. While this mechanism has been deployed, an alternate 182 mechanism where TLS is negotiated immediately at connection start on 183 a separate port (referred to in this document as "Implicit TLS") has 184 been deployed more successfully. To encourage more widespread use of 185 TLS, and to encourage a greater consistency for how TLS is used, this 186 specification now recommends use of Implicit TLS for POP, IMAP, SMTP 187 Submission, and all other protocols used between a Mail User Agent 188 and a mail service. 190 3.1. Implicit TLS for POP 192 When a TCP connection is established for the "pop3s" service (default 193 port 995), a TLS handshake begins immediately. Clients MUST 194 implement the certificate validation mechanism described in 195 [RFC7817]. Once the TLS session is established, POP3 [RFC1939] 196 protocol messages are exchanged as TLS application data for the 197 remainder of the TCP connection. After the server sends a +OK 198 greeting, the server and client MUST enter AUTHORIZATION state, even 199 if client credentials were supplied during the TLS handshake. 201 See Section 5.5 and Section 4.2 for additional information on client 202 certificate authentication. See Section 7.1 for port registration 203 information. 205 3.2. Implicit TLS for IMAP 207 When a TCP connection is established for the "imaps" service (default 208 port 993), a TLS handshake begins immediately. Clients MUST 209 implement the certificate validation mechanism described in [RFC3501] 210 and SHOULD implement the certificate validation mechanism described 211 in [RFC7817]. Once the TLS session is established, IMAP [RFC3501] 212 protocol messages are exchanged as TLS application data for the 213 remainder of the TCP connection. If client credentials were provided 214 during the TLS handshake that the server finds acceptable, the server 215 MAY issue a PREAUTH greeting in which case both the server and client 216 enter AUTHENTICATED state. If the server issues an OK greeting then 217 both server and client enter NOT AUTHENTICATED state. 219 See Section 5.5 and Section 4.2 for additional information on client 220 certificate authentication. See Section 7.1 and Section 7.2 for port 221 registration information. 223 3.3. Implicit TLS for SMTP Submission 225 When a TCP connection is established for the "submissions" service 226 (default port 465), a TLS handshake begins immediately. Clients MUST 227 implement the certificate validation mechanism described in 228 [RFC7817]. Once a TLS session is established, message submission 229 protocol data [RFC6409] is exchanged as TLS application data for the 230 remainder of the TCP connection. (Note: the "submissions" service 231 name is defined in section 10.3 of this document, and follows the 232 usual convention that the name of a service layered on top of 233 Implicit TLS consists of the name of the service as used without TLS, 234 with an "s" appended.) 236 The STARTTLS mechanism on port 587 is relatively widely deployed due 237 to the situation with port 465 (discussed in Section 7.3). This 238 differs from IMAP and POP services where Implicit TLS is more widely 239 deployed on servers than STARTTLS. It is desirable to migrate core 240 protocols used by MUA software to Implicit TLS over time for 241 consistency as well as the additional reasons discussed in 242 Appendix A. However, to maximize use of encryption for submission it 243 is desirable to support both mechanisms for Message Submission over 244 TLS for a transition period of several years. As a result, clients 245 and servers SHOULD implement both STARTTLS on port 587 and Implicit 246 TLS on port 465 for this transition period. Note that there is no 247 significant difference between the security properties of STARTTLS on 248 port 587 and Implicit TLS on port 465 if the implementations are 249 correct and both client and server are configured to require 250 successful negotiation of TLS prior to message submission. 252 Note that the "submissions" port provides access to a Mail Submission 253 Agent (MSA) as defined in [RFC6409] so requirements and 254 recommendations for MSAs in that document apply to the submissions 255 port, including the requirement to implement SMTP AUTH [RFC4954]. 257 See Section 5.5 and Section 4.2 for additional information on client 258 certificate authentication. See Section 7.3 for port registration 259 information. 261 3.4. Implicit TLS Connection Closure for POP, IMAP and SMTP Submission 263 When a client or server wishes to close the connection, it SHOULD 264 initiate the exchange of TLS close alerts before TCP connection 265 termination. The client MAY, after sending a TLS close alert, 266 gracefully close the TCP connection without waiting for a TLS 267 response from the server. 269 4. Recommendations for Use of TLS by Mail Access Services and Message 270 Submission Services 272 The following practices are recommended for Mail Access Services and 273 Mail Submission Services: 275 o Mail Service Providers (MSPs) which support POP, IMAP, and/or 276 Message Submission, MUST support TLS access for those services. 278 o Other services than POP, IMAP and/or Message Submission provided 279 by MSPs SHOULD support TLS access, and MUST support TLS access for 280 those services which support authentication via username and 281 password. 283 o MSPs which support POP, IMAP, and/or Message Submission, SHOULD 284 provide and support instances of those services which use Implicit 285 TLS. (See Section 3.) 287 o For compatibility with existing MUAs and existing MUA 288 configurations, MSPs SHOULD also, in the near term, provide 289 instances of these services which support STARTTLS. This will 290 permit legacy MUAs to discover new availability of TLS capability 291 on servers, and may increase use of TLS by such MUAs. However, 292 servers SHOULD NOT advertise STARTTLS if use of the STARTTLS 293 command by a client is likely to fail (for example, if the server 294 has no server certificate configured.) 296 o MSPs SHOULD advertise their Mail Access Services and Mail 297 Submission Services using DNS SRV records according to [RFC6186]. 298 (In addition to making correct configuration easier for MUAs, this 299 provides a way by which MUAs can discover when an MSP begins to 300 offer TLS-based services.) Services supporting TLS SHOULD be 301 advertised in preference to cleartext services (if offered). In 302 addition, services using Implicit TLS SHOULD be advertised in 303 preference to services supporting STARTTLS (if offered). (See 304 also Section 4.5.) 306 o MSPs SHOULD deprecate use of cleartext Mail Access Services and 307 Mail Submission Services as soon as practicable. (See 308 Section 4.1.) 310 o MSPs currently supporting such use of cleartext SMTP (on port 25) 311 as a means of message submission by their users (whether or not 312 requiring authentication) SHOULD transition their users to using 313 TLS (either Implicit TLS or STARTTLS) as soon as practicable. 315 o Mail services MUST support TLS 1.2 or later. 317 o All Mail services SHOULD implement the recommended TLS cipher 318 suites described in [RFC7525] or a future BCP or standards track 319 revision of that document. 321 o Mail services currently supporting SSL 2.x, SSL 3.0, or TLS 1.0 322 SHOULD transition their users to later versions of TLS, and 323 discontinue support for those versions of SSL and TLS, as soon as 324 practicable. 326 o Mail Submission Servers accepting mail using TLS SHOULD include 327 the TLS ciphersuite of the session in which the mail was received, 328 in the Received field of the outgoing message. (See Section 4.3.) 330 o All Mail services implementing TLS SHOULD log TLS cipher 331 information along with any connection or authentication logs that 332 they maintain. 334 Additional considerations and details appear below. 336 4.1. Deprecation of Services Using Cleartext and TLS Versions < 1.1 338 The specific means employed for deprecation of cleartext Mail Access 339 Services and Mail Submission Services MAY vary from one MSP to the 340 next in light of their user communities' needs and constraints. For 341 example, an MSP MAY implement a gradual transition in which, over 342 time, more and more users are forbidden to authenticate to cleartext 343 instances of these services, thus encouraging those users to migrate 344 to Implicit TLS. Access to cleartext services should eventually be 345 either disabled, or limited strictly for use by legacy systems which 346 cannot be upgraded. 348 After a user's ability to authenticate to a service using cleartext 349 is revoked, the server denying such access MUST NOT provide any 350 indication over a cleartext channel of whether the user's 351 authentication credentials were valid. An attempt to authenticate as 352 such a user using either invalid credentials or valid credentials 353 MUST both result in the same indication of access being denied. 355 Also, users previously authenticating with passwords sent as 356 cleartext SHOULD be required to change those passwords when migrating 357 to TLS, since the old passwords were likely to have been compromised. 359 Transition of users from SSL or TLS 1.0 to later versions of TLS MAY 360 be accomplished by a means similar to that described above. There 361 are multiple ways to accomplish this. One way is for the server to 362 refuse a ClientHello message from any client sending a protocol 363 version number corresponding to any version of SSL or TLS 1.0. 364 Another way is for the server to accept ClientHello messages from 365 some client versions that it does not wish to support, but later 366 refuse to allow the user to authenticate. The latter method may 367 provide a better indication to the user of the reason for the failure 368 but (depending on the protocol and method of authentication used) may 369 also risk exposure of the user's password over an channel which is 370 known to not provide adequate confidentiality. 372 It is RECOMMENDED that new users be required to use TLS version 1.1 373 or greater from the start. However an MSP may find it necessary to 374 make exceptions to accommodate some legacy systems which support only 375 earlier versions of TLS, or only cleartext. 377 4.2. Mail Server Use of Client Certificate Authentication 379 Mail servers MAY implement client certificate authentication on the 380 Implicit TLS port. Servers MUST NOT request a client certificate 381 during the TLS handshake unless the server is configured to accept 382 some client certificates as sufficient for authentication and the 383 server has the ability to determine a mail server authorization 384 identity matching such certificates. How to make this determination 385 is presently implementation specific. 387 If the server accepts the client's certificate as sufficient for 388 authorization, it MUST enable the SASL EXTERNAL [RFC4422] mechanism. 389 An IMAPS server MAY issue a PREAUTH greeting instead of enabling SASL 390 EXTERNAL. 392 4.3. Recording TLS Cipher Suite in Received Header 394 The ESMTPS transmission type [RFC3848] provides trace information 395 that can indicate TLS was used when transferring mail. However, TLS 396 usage by itself is not a guarantee of confidentiality or security. 397 The TLS cipher suite provides additional information about the level 398 of security made available for a connection. This defines a new SMTP 399 "tls" Received header additional-registered-clause that is used to 400 record the TLS cipher suite that was negotiated for the connection. 401 The value included in this additional clause SHOULD be the registered 402 cipher suite name (e.g., TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) 403 included in the TLS cipher suite registry. In the event the 404 implementation does not know the name of the cipher suite (a 405 situation that should be remedied promptly), a four-digit hexadecimal 406 cipher suite identifier MAY be used. The ABNF for the field follows: 408 tls-cipher-clause = CFWS "tls" FWS tls-cipher 410 tls-cipher = tls-cipher-suite-name / tls-cipher-suite-hex 412 tls-cipher-name = ALPHA *(ALPHA / DIGIT / "_") 413 ; as registered in IANA cipher suite registry 415 tls-cipher-hex = "0x" 4HEXDIG 417 4.4. TLS Server Certificate Requirements 419 MSPs MUST maintain valid server certificates for all servers. See 420 [RFC7817] for the recommendations and requirements necessary to 421 achieve this. 423 If a protocol server provides service for more than one mail domain, 424 it MAY use a separate IP address for each domain and/or a server 425 certificate that advertises multiple domains. This will generally be 426 necessary unless and until it is acceptable to impose the constraint 427 that the server and all clients support the Server Name Indication 428 extension to TLS [RFC6066]. For more discussion of this problem, see 429 section 5.1 of [RFC7817]. 431 4.5. Recommended DNS records for mail protocol servers 433 This section discusses not only the DNS records that are recommended, 434 but also implications of DNS records for server configuration and TLS 435 server certificates. 437 4.5.1. MX records 439 It is recommended that MSPs advertise MX records for handling of 440 inbound mail (instead of relying entirely on A or AAAA records), and 441 that those MX records be signed using DNSSEC. This is mentioned here 442 only for completeness, as handling of inbound mail is out of scope 443 for this document. 445 4.5.2. SRV records 447 MSPs SHOULD advertise SRV records to aid MUAs in determination of 448 proper configuration of servers, per the instructions in [RFC6186]. 450 MSPs SHOULD advertise servers that support Implicit TLS in preference 451 to those which support cleartext and/or STARTTLS operation. 453 4.5.3. DNSSEC 455 All DNS records advertised by an MSP as a means of aiding clients in 456 communicating with the MSP's servers, SHOULD be signed using DNSSEC. 458 4.5.4. TLSA records 460 MSPs SHOULD advertise TLSA records to provide an additional trust 461 anchor for public keys used in TLS server certificates. However, 462 TLSA records MUST NOT be advertised unless they are signed using 463 DNSSEC. 465 4.6. Changes to Internet Facing Servers 467 When an MSP changes the Internet Facing Servers providing mail access 468 and mail submission services, including SMTP-based spam/virus 469 filters, it is generally necessary to support the same and/or a newer 470 version of TLS and the same security directives that were previously 471 advertised. 473 5. Recommendations for use of TLS by Mail User Agents 475 It is recommended that Mail User Agents implement the following 476 practices: 478 o MUAs SHOULD be capable of using DNS SRV records to discover Mail 479 Access Services and Mail Submission Services that are advertised 480 by a MSP for an account being configured. Other means of 481 discovering server configuration information (e.g. a database 482 maintained by the MUA vendor) MAY also be supported. (See 483 Section 5.1 for more information.) 485 o MUAs SHOULD be configurable to require a minimum level of 486 confidentiality for any particular Mail Account, and refuse to 487 exchange information via any service associated with that Mail 488 Account if the session does not provide that minimum level of 489 confidentiality. (See Section 5.2.) 491 o MUAs MUST NOT consider a session as meeting a minimum level of 492 confidentiality if the server's TLS certificate cannot be 493 validated. (See Section 5.3.) 495 o MUAs MAY impose other minimum confidentiality requirements in the 496 future, e.g. in order to discourage use of TLS versions or 497 cryptographic algorithms in which weaknesses have been discovered. 499 o MUAs SHOULD provide a prominent visual indication of the level of 500 confidentiality associated with an account configuration (for 501 example, indications such as "lock" icons or changed background 502 colors similar to those used by some browsers), at appropriate 503 times and locations in order to inform the user of the 504 confidentiality of the communications associated with that 505 account. For example, this might be done whenever (a) prompting 506 the user for authentication credentials, (b) the user is composing 507 mail that will be sent to a particular submission server, (c) a 508 list of accounts is displayed (particularly if the user can select 509 from that list to read mail), or (d) the user is requesting to 510 view or update any configuration data that will be stored on a 511 remote server. 513 o MUAs MUST implement TLS 1.2 [RFC5246] or later. Earlier TLS and 514 SSL versions MAY also be supported so long as the MUA requires at 515 least TLS 1.1 [RFC4346] when accessing accounts that are 516 configured to impose minimum confidentiality requirements. Per 517 [RFC7525], TLS 1.1 (or earlier) SHOULD NOT be used unless no 518 higher version is available during TLS protocol negotiation. 520 o All MUAs SHOULD implement the recommended TLS cipher suites 521 described in [RFC7525] or a future BCP or standards track revision 522 of that document. 524 o MUAs that are configured to not require minimum confidentiality 525 for one or more accounts SHOULD detect when TLS becomes available 526 on those accounts (using [RFC6186] or other means), and offer to 527 upgrade the account to require TLS. 529 Additional considerations and details appear below. 531 5.1. Use of SRV records in Establishing Configuration 533 This section updates [RFC6186] by changing the preference rules and 534 adding a new SRV service label _submissions._tcp to refer to Message 535 Submission with Implicit TLS. 537 User-configurable MUAs SHOULD support use of [RFC6186] for account 538 setup. However, when using configuration information obtained by 539 this method, MUAs SHOULD ignore advertised services that do not 540 satisfy minimum confidentiality requirements, unless the user has 541 explicitly requested reduced confidentiality. This will have the 542 effect of causing the MUA to default to ignoring advertised 543 configurations that do not support TLS, even when those advertised 544 configurations have a higher priority than other advertised 545 configurations. 547 When using [RFC6186] configuration information, Mail User Agents 548 SHOULD NOT automatically establish new configurations that do not 549 require TLS for all servers, unless there are no advertised 550 configurations using TLS. If such a configuration is chosen, prior 551 to attempting to authenticate to the server or use the server for 552 message submission, the MUA SHOULD warn the user that traffic to that 553 server will not be encrypted and that it will therefore likely be 554 intercepted by unauthorized parties. The specific wording is to be 555 determined by the implementation, but it should adequately capture 556 the sense of risk given the widespread incidence of mass surveillance 557 of email traffic. 559 Similarly, a MUA MUST NOT attempt to "test" a particular mail account 560 configuration by submitting the user's authentication credentials to 561 a server, unless a TLS session meeting minimum confidentiality levels 562 has been established with that server. If minimum confidentiality 563 requirements have not been satisfied, the MUA must explicitly warn 564 the user that his password may be exposed to attackers before testing 565 the new configuration. 567 When establishing a new configuration for connecting to an IMAP, POP, 568 or SMTP submission server, based on SRV records, an MUA SHOULD either 569 verify that the SRV records are verifiably signed using DNSSEC, or 570 that the target FQDN of the SRV record matches the original server 571 FQDN for which the SRV queries were made. If the target FQDN is not 572 in the queried domain, the MUA SHOULD verify with the user that the 573 SRV target FQDN is suitable for use, before executing any connections 574 to the host. (See [RFC6186] section 6). 576 An MUA MUST NOT consult SRV records to determine which servers to use 577 on every connection attempt, unless those SRV records are signed by 578 DNSSEC and have a valid signature. However, an MUA MAY consult SRV 579 records from time to time to determine if an MSP's server 580 configuration has changed, and alert the user if it appears that this 581 has happened. This can also serve as a means to encourage users to 582 upgrade their configurations to require TLS if and when their MSPs 583 support it. 585 5.2. Minimum Confidentiality Level 587 MUAs SHOULD, by default, require a minimum level of confidentiality 588 for services accessed by each account. For MUAs supporting the 589 ability to access multiple mail accounts, this requirement SHOULD be 590 configurable on a per-account basis. 592 The default minimum expected level of confidentiality for all new 593 accounts SHOULD be at least use of TLS version 1.1 or greater, and 594 successful validation of the server's certificate. (Future revisions 595 to this specification may raise these requirements or impose 596 additional requirements to address newly-discovered weaknesses in 597 protocols or cryptographic algorithms.) 599 MUAs MAY permit the user to disable this minimum confidentiality 600 requirement during initial account configuration, or subsequently 601 editing an account configuration, but MUST warn users that such a 602 configuration will not assure privacy for either passwords or 603 messages. 605 An MUA which is configured to require a minimum level of 606 confidentiality for a mail account MUST NOT attempt to perform any 607 operation other than capability discovery, or STARTTLS for servers 608 not using Implicit TLS, unless the minimum level of confidentiality 609 is provided by that connection. 611 MUAs SHOULD NOT allow users to easily access or send mail via an 612 connection, or authenticate to any service using a password, if that 613 account is configured to impose minimum confidentiality requirements 614 and that connection does not meet all of those requirements. An 615 example of "easily access" would be to display a dialog informing the 616 user that the security requirements of the account were not met by 617 the connecting, but allowing the user to "click through" to send mail 618 or access the service anyway. Experience indicates that users 619 presented with such an option often "click through" without 620 understanding the risks that they're accepting by doing so. 622 Furthermore, users who frequently find the need to "click through" to 623 use an insecure connection may become conditioned to do so as a 624 matter of habit, before considering whether the risks are reasonable 625 in each specific instance. 627 An MUA which is not configured to require a minimum level of 628 confidentiality for a mail account SHOULD still attempt to connect to 629 the services associated with that account using the most secure means 630 available, e.g. by using Implicit TLS or STARTTLS. 632 5.3. Certificiate Validation 634 MUAs MUST validate TLS server certificates according to [RFC7817] and 635 PKIX [RFC5280]. 637 MUAs MAY also support DANE [RFC6698] as a means of validating server 638 certificates in order to meet minimum confidentiality requirements. 640 MUAs MAY support use of certificate pinning but MUST NOT consider a 641 connection in which the server's authenticity relies on certificate 642 pinning, as providing the minimum level of confidentiality. (See 643 Section 5.4.) 645 5.4. Certificate Pinning 647 During account setup, the MUA will identify servers that provide 648 account services such as mail access and mail submission (the 649 previous section describes one way to do this). The certificates for 650 these servers are verified using the rules described in [RFC7817] and 651 PKIX [RFC5280]. In the event the certificate does not validate due 652 to an expired certificate, lack of appropriate chain of trust, or 653 lack of identifier match, the MUA MAY offer to create a persistent 654 binding between that certificate and the saved host name for the 655 server, for use when accessing that account's servers. This is 656 called certificate pinning. 658 Certificate pinning is only appropriate during mail account setup and 659 MUST NOT be offered as an option in response to a failed certificate 660 validation for an existing mail account. An MUA that allows 661 certificate pinning MUST NOT allow a certificate pinned for one 662 account to validate connections for other accounts. 664 A pinned certificate is subject to a man-in-the-middle attack at 665 account setup time, and typically lacks a mechanism to automatically 666 revoke or securely refresh the certificate. Note also that a man-in- 667 the-middle attack at account setup time will expose the user's 668 password to the attacker (if a password is used). Therefore use of a 669 pinned certificate does not meet the requirement for a minimum 670 confidentiality level, and an MUA MUST NOT indicate to the user that 671 the such confidentiality is provided. Additional advice on 672 certificate pinning is present in [RFC6125]. 674 5.5. Client Certificate Authentication 676 MUAs MAY implement client certificate authentication on the Implicit 677 TLS port. An MUA MUST NOT provide a client certificate during the 678 TLS handshake unless the server requests one and the client has 679 determined the certificate can be safely used with that specific 680 server, OR the client has been explicitly configured by the user to 681 use that particular certificate with that server. How to make this 682 determination is presently implementation specific. 684 A client supporting client certificate authentication with Implicit 685 TLS MUST implement the SASL EXTERNAL [RFC4422] mechanism using the 686 appropriate authentication command (AUTH for POP3 [RFC5034], AUTH for 687 SMTP Submission [RFC4954], AUTHENTICATE for IMAP [RFC3501]). 689 6. Considerations related to Anti-Virus/Anti-Spam Software and Services 691 There are multiple ways to connect an Anti-Virus and/or Anti-Spam 692 (AVAS) service to a mail server. Some mechanisms, such as the de- 693 facto milter protocol, are out of scope for this specification. 694 However, some services use an SMTP relay proxy that intercepts mail 695 at the application layer to perform a scan and proxy or forward to 696 another MTA. Deploying AVAS services in this way can cause many 697 problems [RFC2979] including direct interference with this 698 specification, and other forms of confidentiality or security 699 reduction. An AVAS product or service is considered compatible with 700 this specification if all IMAP, POP and SMTP-related software 701 (including proxies) it includes are compliant with this 702 specification. 704 Note that end-to-end email encryption prevents AVAS software and 705 services from using email content as part of a spam or virus 706 assessment. Furthermore, while a minimum confidentiality level can 707 prevent a man-in-the-middle from introducing spam or virus content 708 between the MUA and Submission server, it does not prevent other 709 forms of client or account compromise. Use of AVAS services for 710 submitted email therefore remains necessary. 712 7. IANA Considerations 713 7.1. POP3S Port Registration Update 715 IANA is asked to update the registration of the TCP well-known port 716 995 using the following template ([RFC6335]): 718 Service Name: pop3s 719 Transport Protocol: TCP 720 Assignee: IETF 721 Contact: IESG 722 Description: POP3 over TLS protocol 723 Reference: RFC XXXX (this document once published) 724 Port Number: 995 726 7.2. IMAPS Port Registration Update 728 IANA is asked to update the registration of the TCP well-known port 729 993 using the following template ([RFC6335]): 731 Service Name: imaps 732 Transport Protocol: TCP 733 Assignee: IETF 734 Contact: IESG 735 Description: IMAP over TLS protocol 736 Reference: RFC XXXX (this document once published) 737 Port Number: 993 739 7.3. Submissions Port Registration 741 IANA is asked to assign an alternate usage of port 465 in addition to 742 the current assignment using the following template ([RFC6335]): 744 Service Name: submissions 745 Transport Protocol: TCP 746 Assignee: IETF 747 Contact: IESG 748 Description: Message Submission over TLS protocol 749 Reference: RFC XXXX (this document once published) 750 Port Number: 465 752 This is a one-time procedural exception to the rules in RFC 6335. 753 This requires explicit IESG approval and does not set a precedent. 755 Historically, port 465 was briefly registered as the "smtps" port. 756 This registration made no sense as the SMTP transport MX 757 infrastructure has no way to specify a port, so port 25 is always 758 used. As a result, the registration was revoked and was subsequently 759 reassigned to a different service. In hindsight, the "smtps" 760 registration should have been renamed or reserved rather than 761 revoked. Unfortunately, some widely deployed mail software 762 interpreted "smtps" as "submissions" [RFC6409] and used that port for 763 email submission by default when an end-user requests security during 764 account setup. If a new port is assigned for the submissions 765 service, email software will either continue with unregistered use of 766 port 465 (leaving the port registry inaccurate relative to de-facto 767 practice and wasting a well-known port), or confusion between the de- 768 facto and registered ports will cause harmful interoperability 769 problems that will deter use of TLS for message submission. The 770 authors believe both of these outcomes are less desirable than a wart 771 in the registry documenting real-world usage of a port for two 772 purposes. Although STARTTLS-on-port-587 has deployed, it has not 773 replaced deployed use of Implicit TLS submission on port 465. 775 8. Security Considerations 777 This entire document is about security considerations. In general, 778 this is targeted to improve mail confidentiality and to mitigate 779 threats external to the email system such as network-level snooping 780 or interception; this is not intended to mitigate active attackers 781 who have compromised service provider systems. 783 9. References 785 9.1. Normative References 787 [RFC1939] Myers, J. and M. Rose, "Post Office Protocol - Version 3", 788 STD 53, RFC 1939, DOI 10.17487/RFC1939, May 1996, 789 . 791 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 792 Requirement Levels", BCP 14, RFC 2119, 793 DOI 10.17487/RFC2119, March 1997, 794 . 796 [RFC3207] Hoffman, P., "SMTP Service Extension for Secure SMTP over 797 Transport Layer Security", RFC 3207, DOI 10.17487/RFC3207, 798 February 2002, . 800 [RFC3501] Crispin, M., "INTERNET MESSAGE ACCESS PROTOCOL - VERSION 801 4rev1", RFC 3501, DOI 10.17487/RFC3501, March 2003, 802 . 804 [RFC5034] Siemborski, R. and A. Menon-Sen, "The Post Office Protocol 805 (POP3) Simple Authentication and Security Layer (SASL) 806 Authentication Mechanism", RFC 5034, DOI 10.17487/RFC5034, 807 July 2007, . 809 [RFC5068] Hutzler, C., Crocker, D., Resnick, P., Allman, E., and T. 810 Finch, "Email Submission Operations: Access and 811 Accountability Requirements", BCP 134, RFC 5068, 812 DOI 10.17487/RFC5068, November 2007, 813 . 815 [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax 816 Specifications: ABNF", STD 68, RFC 5234, 817 DOI 10.17487/RFC5234, January 2008, 818 . 820 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 821 (TLS) Protocol Version 1.2", RFC 5246, 822 DOI 10.17487/RFC5246, August 2008, 823 . 825 [RFC5280] Cooper, D., Santesson, S., Farrell, S., Boeyen, S., 826 Housley, R., and W. Polk, "Internet X.509 Public Key 827 Infrastructure Certificate and Certificate Revocation List 828 (CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, May 2008, 829 . 831 [RFC5322] Resnick, P., Ed., "Internet Message Format", RFC 5322, 832 DOI 10.17487/RFC5322, October 2008, 833 . 835 [RFC6186] Daboo, C., "Use of SRV Records for Locating Email 836 Submission/Access Services", RFC 6186, 837 DOI 10.17487/RFC6186, March 2011, 838 . 840 [RFC6409] Gellens, R. and J. Klensin, "Message Submission for Mail", 841 STD 72, RFC 6409, DOI 10.17487/RFC6409, November 2011, 842 . 844 [RFC7525] Sheffer, Y., Holz, R., and P. Saint-Andre, 845 "Recommendations for Secure Use of Transport Layer 846 Security (TLS) and Datagram Transport Layer Security 847 (DTLS)", BCP 195, RFC 7525, DOI 10.17487/RFC7525, May 848 2015, . 850 [RFC7672] Dukhovni, V. and W. Hardaker, "SMTP Security via 851 Opportunistic DNS-Based Authentication of Named Entities 852 (DANE) Transport Layer Security (TLS)", RFC 7672, 853 DOI 10.17487/RFC7672, October 2015, 854 . 856 [RFC7817] Melnikov, A., "Updated Transport Layer Security (TLS) 857 Server Identity Check Procedure for Email-Related 858 Protocols", RFC 7817, DOI 10.17487/RFC7817, March 2016, 859 . 861 9.2. Informative References 863 [I-D.ietf-uta-mta-sts] 864 Margolis, D., Risher, M., Ramakrishnan, B., Brotman, A., 865 and J. Jones, "SMTP MTA Strict Transport Security (MTA- 866 STS)", draft-ietf-uta-mta-sts-09 (work in progress), 867 September 2017. 869 [RFC2595] Newman, C., "Using TLS with IMAP, POP3 and ACAP", 870 RFC 2595, DOI 10.17487/RFC2595, June 1999, 871 . 873 [RFC2979] Freed, N., "Behavior of and Requirements for Internet 874 Firewalls", RFC 2979, DOI 10.17487/RFC2979, October 2000, 875 . 877 [RFC3848] Newman, C., "ESMTP and LMTP Transmission Types 878 Registration", RFC 3848, DOI 10.17487/RFC3848, July 2004, 879 . 881 [RFC4346] Dierks, T. and E. Rescorla, "The Transport Layer Security 882 (TLS) Protocol Version 1.1", RFC 4346, 883 DOI 10.17487/RFC4346, April 2006, 884 . 886 [RFC4422] Melnikov, A., Ed. and K. Zeilenga, Ed., "Simple 887 Authentication and Security Layer (SASL)", RFC 4422, 888 DOI 10.17487/RFC4422, June 2006, 889 . 891 [RFC4954] Siemborski, R., Ed. and A. Melnikov, Ed., "SMTP Service 892 Extension for Authentication", RFC 4954, 893 DOI 10.17487/RFC4954, July 2007, 894 . 896 [RFC6066] Eastlake 3rd, D., "Transport Layer Security (TLS) 897 Extensions: Extension Definitions", RFC 6066, 898 DOI 10.17487/RFC6066, January 2011, 899 . 901 [RFC6125] Saint-Andre, P. and J. Hodges, "Representation and 902 Verification of Domain-Based Application Service Identity 903 within Internet Public Key Infrastructure Using X.509 904 (PKIX) Certificates in the Context of Transport Layer 905 Security (TLS)", RFC 6125, DOI 10.17487/RFC6125, March 906 2011, . 908 [RFC6335] Cotton, M., Eggert, L., Touch, J., Westerlund, M., and S. 909 Cheshire, "Internet Assigned Numbers Authority (IANA) 910 Procedures for the Management of the Service Name and 911 Transport Protocol Port Number Registry", BCP 165, 912 RFC 6335, DOI 10.17487/RFC6335, August 2011, 913 . 915 [RFC6698] Hoffman, P. and J. Schlyter, "The DNS-Based Authentication 916 of Named Entities (DANE) Transport Layer Security (TLS) 917 Protocol: TLSA", RFC 6698, DOI 10.17487/RFC6698, August 918 2012, . 920 Appendix A. Design Considerations 922 This section is not normative. 924 The first version of this was written independently from draft-moore- 925 email-tls-00.txt; subsequent versions merge ideas from both drafts. 927 One author of this document was also the author of RFC 2595 that 928 became the standard for TLS usage with POP and IMAP, and the other 929 author was perhaps the first to propose that idea. In hindsight both 930 authors now believe that that approach was a mistake. At this point 931 the authors believe that while anything that makes it easier to 932 deploy TLS is good, the desirable end state is that these protocols 933 always use TLS, leaving no need for a separate port for cleartext 934 operation except to support legacy clients while they continue to be 935 used. The separate port model for TLS is inherently simpler to 936 implement, debug and deploy. It also enables a "generic TLS load- 937 balancer" that accepts secure client connections for arbitrary foo- 938 over-TLS protocols and forwards them to a server that may or may not 939 support TLS. Such load-balancers cause many problems because they 940 violate the end-to-end principle and the server loses the ability to 941 log security-relevant information about the client unless the 942 protocol is designed to forward that information (as this 943 specification does for the cipher suite). However, they can result 944 in TLS deployment where it would not otherwise happen which is a 945 sufficiently important goal that it overrides the problems. 947 Although STARTTLS appears only slightly more complex than separate- 948 port TLS, we again learned the lesson that complexity is the enemy of 949 security in the form of the STARTTLS command injection vulnerability 950 (CERT vulnerability ID #555316). Although there's nothing inherently 951 wrong with STARTTLS, the fact it resulted in a common implementation 952 error (made independently by multiple implementers) suggests it is a 953 less secure architecture than Implicit TLS. 955 Section 7 of RFC 2595 critiques the separate-port approach to TLS. 956 The first bullet was a correct critique. There are proposals in the 957 http community to address that, and use of SRV records as described 958 in RFC 6186 resolves that critique for email. The second bullet is 959 correct as well, but not very important because useful deployment of 960 security layers other than TLS in email is small enough to be 961 effectively irrelevant. The third bullet is incorrect because it 962 misses the desirable option of "use and latch-on TLS if available". 963 The fourth bullet may be correct, but is not a problem yet with 964 current port consumption rates. The fundamental error was 965 prioritizing a perceived better design based on a mostly valid 966 critique over real-world deployability. But getting security and 967 confidentiality facilities actually deployed is so important it 968 should trump design purity considerations. 970 Port 465 is presently used for two purposes: for submissions by a 971 large number of clients and service providers and for the "urd" 972 protocol by one vendor. Actually documenting this current state is 973 controversial as discussed in the IANA considerations section. 974 However, there is no good alternative. Registering a new port for 975 submissions when port 465 is widely used for that purpose already 976 will just create interoperability problems. Registering a port 977 that's only used if advertised by an SRV record (RFC 6186) would not 978 create interoperability problems but would require all client and 979 server deployments and software to change significantly which is 980 contrary to the goal of promoting more TLS use. Encouraging use of 981 STARTTLS on port 587 would not create interoperability problems, but 982 is unlikely to have impact on current undocumented use of port 465 983 and makes the guidance in this document less consistent. The 984 remaining option is to document the current state of the world and 985 support future use of port 465 for submission as this increases 986 consistency and ease-of-deployment for TLS email submission. 988 Appendix B. Change Log 990 Changes since draft-ietf-uta-email-deep-07: 992 o After discussion with the WG in Prague, removed BCP language and 993 once again made unambiguous that this is intended as a standards- 994 track document. 996 o Server implementations now MUST implement TLS 1.2, consistent with 997 RFC 7525. MUAs may still consider a TLS 1.1 session as meeting 998 minimum confidentiality requirements. 1000 o MSPs now MUST support TLS for POP, IMAP, Submission, and any other 1001 services that use username/password authentication. 1003 o Added text to clarify the purpose of recommending that MSPs use 1004 DNS SRV records to advertise services. 1006 o Changed text about MUAs not blindly trusting unsigned SRV records, 1007 to instead restate RFC 6186 requirements. 1009 Changes since draft-ietf-uta-email-deep-06: 1011 o On the recommendation of one of the co-chairs and some working 1012 group members, rewrote document with the intended status of BCP. 1013 This involved removing a great deal of text that consisted 1014 essentially of new protocol specification, especially the STS 1015 features, on the theory that a BCP should base its recommendations 1016 on current practice, and that new protocol features should be 1017 subject to the interoperability test requirements associated with 1018 normal standards-track documents. 1020 Changes since draft-ietf-uta-email-deep-05: 1022 o Clarify throughout that the confidentiality assurance level 1023 associated with a mail account is a minimum level; attempt to 1024 distinguish this from the current confidentiality level provided 1025 by a connection between client and server. 1027 o Change naming for confidentiality assurance levels: instead of 1028 "high" or "no" confidence, assign numbers 1 and 0 to them 1029 respectively. This because it seems likely that in the not-too- 1030 distant future, what was defined in -05 as "high" confidence will 1031 be considered insufficient, and calling that "high" confidence 1032 will become misleading. For example, relying entirely on a list 1033 of trusted CAs to validate server certificates from arbitrary 1034 parties, appears to be less and less reliable in practice at 1035 thwarting MITM attacks. 1037 o Clarify that if some services associated with a mail account don't 1038 meet the minimum confidentiality assurance level assigned to that 1039 account, other services that do meet that minimum confidentiality 1040 assurance level may continue to be used. 1042 o Clarify that successful negotiation of at least TLS version 1.1 is 1043 required as a condition of meeting confidentiality assurance level 1044 1. 1046 o Clarify that validation of a server certificate using either DANE 1047 or PKIX is sufficient to meet the certificate validation 1048 requirement of confidentiality assurance level 1. 1050 o Clarify that minimum confidentiality assurance levels are separate 1051 from security directives, and that the requirements of both 1052 mechanisms must be met. 1054 o Explicitly cite an example that a security directive of tls- 1055 version=1.2 won't be saved if the currently negotiated tls-version 1056 is 1.1. (This example already appeared a bit later in the text, 1057 but for author KM it seemed to make the mechanism clearer to use 1058 this example earlier.) 1060 o Clarify some protocol examples as to whether PKIX or DANE was used 1061 to verify a server's certificate. 1063 o Remove most references to DEEP as the conversion from DEEP to MUA- 1064 STS seemed incomplete, but kept the DEEP command for use in POP3 1065 on the assumption that author CN wanted it that way. 1067 o Removed most references to "latch" and derivative words. 1069 o Added pkix+dane as a value for the tls-cert directive, to indicate 1070 (from a server) that both PKIX and DANE validation will be 1071 supported, or (from a client) that both PKIX and DANE were used to 1072 validate a certificate. Also clarified what each of any, pkix, 1073 dane, and pkix+dane mean when advertised by a server and in 1074 particular that tls-cert=any provides no assurance of future PKIX 1075 verifiability in contrast to tls-cert=pkix or tls-cert=pkix+dane. 1076 It seemed important to support the ability to evolve to using 1077 multiple trust anchors for certificate validation, but also to 1078 allow servers to have the option to migrate from PKIX to DANE if 1079 that made sense for them. This change seemed less disruptive than 1080 either defining additional directives, or allowing multiple 1081 instances of the same directive with different values to appear in 1082 the same advertisement. 1084 o Clarify interaction of this specification with anti-virus / anti- 1085 spam mechanisms. 1087 Changes since draft-ietf-uta-email-deep-04: 1089 o Swap sections 5.1 and 5.3 ("Email Security Tags" and "Server DEEP 1090 Status") as that order may aid understanding of the model. Also 1091 rewrote parts of these two sections to try to make the model 1092 clearer. 1094 o Add text about versioning of security tags to make the model 1095 clearer. 1097 o Add example of security tag upgrade. 1099 o Convert remaining mention of TLS 1.0 to TLS 1.1. 1101 o Change document title from DEEP to MUA STS to align with SMTP 1102 relay STS. 1104 * Slight updates to abstract and introductions. 1106 * Rename security latches/tags to security directives. 1108 * Rename server DEEP status to STS policy. 1110 * Change syntax to use directive-style HSTS syntax. 1112 o Make HSTS reference normative. 1114 o Remove SMTP DSN header as that belongs in SMTP relay STS document. 1116 Changes since draft-ietf-uta-email-deep-03: 1118 o Add more references to ietf-uta-email-tls-certs in implementation 1119 requirements section. 1121 o Replace primary reference to RFC 6125 with ietf-uta-email-tls- 1122 certs, so move RFC 6125 to informative list for this 1123 specification. 1125 Changes since draft-ietf-uta-email-deep-02: 1127 o Make reference to design considerations explicit rather than 1128 "elsewhere in this document". 1130 o Change provider requirement so SMTP submission services are 1131 separate from SMTP MTA services as opposed to the previous 1132 phrasing that required the servers be separate (which is too 1133 restrictive). 1135 o Update DANE SMTP reference 1136 Changes since draft-ietf-uta-email-deep-01: 1138 o Change text in tls11 and tls12 registrations to clarify 1139 certificate rules, including additional PKIX and DANE references. 1141 o Change from tls10 to tls11 (including reference) as the minimum. 1143 o Fix typo in example 5. 1145 o Remove open issues section; enough time has passed so not worth 1146 waiting for more input. 1148 Changes since draft-ietf-uta-email-deep-00: 1150 o Update and clarify abstract 1152 o use term confidentiality instead of privacy in most cases. 1154 o update open issues to request input for missing text. 1156 o move certificate pinning sub-section to account setup section and 1157 attempt to define it more precisely. 1159 o Add note about end-to-end encryption in AVAS section. 1161 o swap order of DNSSEC and TLSA sub-sections. 1163 o change meaning of 'tls10' and 'tls12' latches to require 1164 certificate validation. 1166 o Replace cipher suite advice with reference to RFC 7525. Change 1167 examples to use TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 as cipher 1168 suite. 1170 o Add text to update IMAP, POP3 and Message Submission standards 1171 with newer TLS advice. 1173 o Add clearer text in introduction that this does not cover SMTP 1174 relay. 1176 o Update references to uta-tls-certs. 1178 o Add paragraph to Implicit TLS for SMTP Submission section 1179 recommending that STARTTLS also be implemented. 1181 Changes since draft-newman-email-deep-02: 1183 o Changed "privacy assurance" to "confidentiality assurance" 1184 o Changed "low privacy assurance" to "no confidentiality assurance" 1186 o Attempt to improve definition of confidentiality assurance level. 1188 o Add SHOULD indicate when MUA is showing list of mail accounts. 1190 o Add SHOULD NOT latch tls10, tls12 tags until TLS negotiated. 1192 o Removed sentence about deleting and re-creating the account in 1193 latch failure section. 1195 o Remove use of word "fallback" with respect to TLS version 1196 negotiation. 1198 o Added bullet about changes to Internet facing servers to MSP 1199 section. 1201 o minor wording improvements based on feedback 1203 Changes since -01: 1205 o Updated abstract, introduction and document structure to focus 1206 more on mail user agent privacy assurance. 1208 o Added email account privacy section, also moving section on 1209 account setup using SRV records to that section. 1211 o Finished writing IANA considerations section 1213 o Remove provisional concept and instead have server explicitly list 1214 security tags clients should latch. 1216 o Added note that rules for the submissions port follow the same 1217 rules as those for the submit port. 1219 o Reference and update advice in [RFC5068]. 1221 o Fixed typo in Client Certificate Authentication section. 1223 o Removed tls-pfs security latch and all mention of perfect forward 1224 secrecy as it was controversial. 1226 o Added reference to HSTS. 1228 Changes since -00: 1230 o Rewrote introduction to merge ideas from draft-moore-email-tls-00. 1232 o Added Implicit TLS section, Account configuration section and IANA 1233 port registration updates based on draft-moore-email-tls-00. 1235 o Add protocol details necessary to standardize implicit TLS for 1236 POP/IMAP/submission, using ideas from draft-melnikov-pop3-over- 1237 tls. 1239 o Reduce initial set of security tags based on feedback. 1241 o Add deep status concept to allow a window for software updates to 1242 be backed out before latches make that problematic, as well as to 1243 provide service providers with a mechanism they can use to assist 1244 customers in the event of a privacy failure. 1246 o Add DNS SRV section from draft-moore-email-tls-00. 1248 o Write most of the missing IANA considerations section. 1250 o Rewrite most of implementation requirements section based more on 1251 draft-moore-email-tls-00. Remove new cipher requirements for now 1252 because those may be dealt with elsewhere. 1254 Appendix C. Acknowledgements 1256 Thanks to Ned Freed for discussion of the initial latch concepts in 1257 this document. Thanks to Alexey Melnikov for draft-melnikov-pop3- 1258 over-tls-02, which was the basis of the POP3 Implicit TLS text. 1259 Thanks to Russ Housley, Alexey Melnikov and Dan Newman for review 1260 feedback. Thanks to Paul Hoffman for interesting feedback in initial 1261 conversations about this idea. 1263 Authors' Addresses 1265 Keith Moore 1266 Windrock, Inc. 1267 PO Box 1934 1268 Knoxville, TN 37901 1269 US 1271 Email: moore@network-heretics.com 1272 Chris Newman 1273 Oracle 1274 440 E. Huntington Dr., Suite 400 1275 Arcadia, CA 91006 1276 US 1278 Email: chris.newman@oracle.com