idnits 2.17.00 (12 Aug 2021) /tmp/idnits31499/draft-ietf-uta-tls-bcp-01.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 date (June 24, 2014) is 2887 days in the past. Is this intentional? Checking references for intended status: Best Current Practice ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) ** Obsolete normative reference: RFC 4492 (Obsoleted by RFC 8422) ** Obsolete normative reference: RFC 5246 (Obsoleted by RFC 8446) ** Downref: Normative reference to an Informational RFC: RFC 7027 -- Obsolete informational reference (is this intentional?): RFC 2246 (Obsoleted by RFC 4346) -- Obsolete informational reference (is this intentional?): RFC 4346 (Obsoleted by RFC 5246) -- Obsolete informational reference (is this intentional?): RFC 5077 (Obsoleted by RFC 8446) Summary: 3 errors (**), 0 flaws (~~), 1 warning (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 UTA Y. Sheffer 3 Internet-Draft Porticor 4 Intended status: Best Current Practice R. Holz 5 Expires: December 26, 2014 TUM 6 P. Saint-Andre 7 &yet 8 June 24, 2014 10 Recommendations for Secure Use of TLS and DTLS 11 draft-ietf-uta-tls-bcp-01 13 Abstract 15 Transport Layer Security (TLS) and Datagram Transport Security Layer 16 (DTLS) are widely used to protect data exchanged over application 17 protocols such as HTTP, SMTP, IMAP, POP, SIP, and XMPP. Over the 18 last few years, several serious attacks on TLS have emerged, 19 including attacks on its most commonly used cipher suites and modes 20 of operation. This document provides recommendations for improving 21 the security of both software implementations and deployed services 22 that use TLS and DTLS. 24 Status of This Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at http://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on December 26, 2014. 41 Copyright Notice 43 Copyright (c) 2014 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (http://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 59 2. Conventions used in this document . . . . . . . . . . . . . . 3 60 3. Recommendations . . . . . . . . . . . . . . . . . . . . . . . 3 61 3.1. Protocol Versions . . . . . . . . . . . . . . . . . . . . 3 62 3.2. Fallback to SSL . . . . . . . . . . . . . . . . . . . . . 4 63 3.3. Always Use TLS . . . . . . . . . . . . . . . . . . . . . 4 64 3.4. Cipher Suites . . . . . . . . . . . . . . . . . . . . . . 5 65 3.5. Public Key Length . . . . . . . . . . . . . . . . . . . . 6 66 3.6. Compression . . . . . . . . . . . . . . . . . . . . . . . 7 67 3.7. Session Resumption . . . . . . . . . . . . . . . . . . . 7 68 3.8. Renegotiation . . . . . . . . . . . . . . . . . . . . . . 7 69 4. Detailed Guidelines . . . . . . . . . . . . . . . . . . . . . 7 70 4.1. Cipher Suite Negotiation Details . . . . . . . . . . . . 7 71 4.2. Alternative Cipher Suites . . . . . . . . . . . . . . . . 8 72 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 73 6. Security Considerations . . . . . . . . . . . . . . . . . . . 9 74 6.1. AES-GCM . . . . . . . . . . . . . . . . . . . . . . . . . 9 75 6.2. Forward Secrecy . . . . . . . . . . . . . . . . . . . . . 9 76 6.3. Certificate Revocation . . . . . . . . . . . . . . . . . 10 77 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 78 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 79 8.1. Normative References . . . . . . . . . . . . . . . . . . 11 80 8.2. Informative References . . . . . . . . . . . . . . . . . 11 81 Appendix A. Appendix: Change Log . . . . . . . . . . . . . . . . 13 82 A.1. draft-ietf-tls-bcp-01 . . . . . . . . . . . . . . . . . . 13 83 A.2. draft-ietf-tls-bcp-00 . . . . . . . . . . . . . . . . . . 13 84 A.3. draft-sheffer-tls-bcp-02 . . . . . . . . . . . . . . . . 13 85 A.4. draft-sheffer-tls-bcp-01 . . . . . . . . . . . . . . . . 13 86 A.5. draft-sheffer-tls-bcp-00 . . . . . . . . . . . . . . . . 14 87 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 89 1. Introduction 91 Transport Layer Security (TLS) and Datagram Transport Security Layer 92 (DTLS) are widely used to protect data exchanged over application 93 protocols such as HTTP, SMTP, IMAP, POP, SIP, and XMPP. Over the 94 last few years, several serious attacks on TLS have emerged, 95 including attacks on its most commonly used cipher suites and modes 96 of operation. For instance, both AES-CBC and RC4, which together 97 comprise most current usage, have been attacked in the context of 98 TLS. A companion document [I-D.sheffer-uta-tls-attacks] provides 99 detailed information about these attacks. 101 Because of these attacks, those who implement and deploy TLS and DTLS 102 need updated guidance on how TLS can be used securely. Note that 103 this document provides guidance for deployed services, as well as 104 software implementations. In fact, this document calls for the 105 deployment of algorithms that are widely implemented but not yet 106 widely deployed. 108 The recommendations herein take into consideration the security of 109 various mechanisms, their technical maturity and interoperability, 110 and their prevalence in implementatios at the time of writing. These 111 recommendations apply to both TLS and DTLS. TLS 1.3, when it is 112 standardized and deployed in the field, should resolve the current 113 vulnerabilities while providing significantly better functionality, 114 and will very likely obsolete this document. 116 These are minimum recommendations for the general use of TLS. 117 Individual specifications may have stricter requirements related to 118 one or more aspects of the protocol, and based on their particular 119 circumstances. When that is the case, implementers MUST adhere to 120 those stricter requirements. 122 Community knowledge about the strength of various algorithms and 123 feasible attacks can change quickly, and experience shows that a 124 security BCP is a point-in-time statement. Readers are advised to 125 seek out any errata or updates that apply to this document. 127 2. Conventions used in this document 129 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 130 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 131 document are to be interpreted as described in [RFC2119]. 133 3. Recommendations 135 3.1. Protocol Versions 137 It is important both to stop using old, less secure versions of SSL/ 138 TLS and to start using modern, more secure versions. Therefore: 140 o Implementations MUST NOT negotiate SSL version 2. 142 Rationale: SSLv2 has serious security vulnerabilities [RFC6176]. 144 o Implementations SHOULD NOT negotiate SSL version 3. 146 Rationale: SSLv3 [RFC6101] was an improvement over SSLv2 and 147 plugged some significant security holes, but did not support 148 strong cipher suites. 150 o Implementations SHOULD NOT negotiate TLS version 1.0 [RFC2246]. 152 Rationale: TLS 1.0 (published in 1999) includes a way to downgrade 153 the connection to SSLv3 and does not support more modern, strong 154 cipher suites. 156 o Implementations MAY negotiate TLS version 1.1 [RFC4346]. 158 Rationale: TLS 1.1 (published in 2006) prevents downgrade attacks 159 to SSL, but does not support certain stronger cipher suites. 161 o Implementations MUST support, and prefer to negotiate, TLS version 162 1.2 [RFC5246]. 164 Rationale: Several stronger cipher suites are available only with 165 TLS 1.2 (published in 2008). 167 As of the date of this writing, the latest version of TLS is 1.2. 168 When TLS is updated to a newer version, this document will be updated 169 to recommend support for the latest version. If this document is not 170 updated in a timely manner, it can be assumed that support for the 171 latest version of TLS is recommended. 173 3.2. Fallback to SSL 175 Some client implementations revert to SSLv3 if the server rejected 176 higher versions of SSL/TLS. This fallback can be forced by a MITM 177 attacker. Moreover, IP scans [[reference?]] show that SSLv3-only 178 servers amount to only about 3% of the current web server population. 179 Therefore, by default clients SHOULD NOT fall back from TLS to SSLv3. 181 3.3. Always Use TLS 183 Combining unprotected and TLS-protected communication opens the way 184 to SSL Stripping and similar attacks. In cases where an application 185 protocol allows implementations or deployments a choice between 186 strict TLS configuration and dynamic upgrade from unencrypted to TLS- 187 protected traffic (such as STARTTLS), clients and servers SHOULD 188 prefer strict TLS configuration. 190 When applicable, Web servers SHOULD advertise that they are willing 191 to accept TLS-only clients, using the HTTP Strict Transport Security 192 (HSTS) header [RFC6797]. 194 3.4. Cipher Suites 196 It is important both to stop using old, insecure cipher suites and to 197 start using modern, more secure cipher suites. Therefore: 199 o Implementations MUST NOT negotiate the NULL cipher suites. 201 Rationale: The NULL cipher suites offer no encryption whatsoever 202 and thus are completely insecure. 204 o Implementations MUST NOT negotiate RC4 cipher suites 206 Rationale: The RC4 stream cipher has a variety of cryptographic 207 weaknesses, as documented in [I-D.popov-tls-prohibiting-rc4]. 209 o Implementations MUST NOT negotiate cipher suites offering only so- 210 called "export-level" encryption (including algorithms with 40 211 bits or 56 bits of security). 213 Rationale: These cipher suites are deliberately "dumbed down" and 214 are very easy to break. 216 o Implementations SHOULD NOT negotiate cipher suites that use 217 algorithms offering less than 128 bits of security. Note that 218 some legacy cipher suites (e.g. 168-bit 3DES) have an effective 219 key length which is smaller than their nominal key length. Such 220 cipher suites should be evaluated accoring to their effective key 221 length. 223 Rationale: Although these cipher suites are not actively subject 224 to breakage, their useful life is short enough that stronger 225 cipher suites are desirable. 227 o Implementations SHOULD prefer cipher suites that use algorithms 228 with at least 128 (and, if possible, 256) bits of security. 230 Rationale: Although the useful life of such cipher suites is 231 unknown, it is probably at least several years for the 128-bit 232 ciphers and "until the next fundamental technology breakthrough" 233 for 256-bit ciphers. 235 o Implementations MUST support, and SHOULD prefer to negotiate, 236 cipher suites offering forward secrecy, such as those in the 237 "EDH", "DHE", and "ECDHE" families. 239 Rationale: Forward secrecy (sometimes called "perfect forward 240 secrecy") prevents the recovery of information that was encrypted 241 with older session keys, thus limiting the amount of time during 242 which attacks can be successful. 244 Given the foregoing considerations, implementation of the following 245 cipher suites is RECOMMENDED (see [RFC5289] for details): 247 o TLS_DHE_RSA_WITH_AES_128_GCM_SHA256 249 o TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 251 o TLS_DHE_RSA_WITH_AES_256_GCM_SHA384 253 o TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 255 We suggest that TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 be preferred in 256 general. 258 Unfortunately, those cipher suites are supported only in TLS 1.2 259 since they are authenticated encryption (AEAD) algorithms [RFC5116]. 260 A future version of this document might recommend cipher suites for 261 earlier versions of TLS. 263 [RFC4492] allows clients and servers to negotiate ECDH parameters 264 (curves). Clients and servers SHOULD prefer verifiably random curves 265 (specifically Brainpool P-256, brainpoolp256r1 [RFC7027]), and fall 266 back to the commonly used NIST P-256 (secp256r1) curve [RFC4492]. In 267 addition, clients SHOULD send an ec_point_formats extension with a 268 single element, "uncompressed". 270 3.5. Public Key Length 272 Because Diffie-Hellman keys of 1024 bits are estimated to be roughly 273 equivalent to 80-bit symmetric keys, it is better to use longer keys 274 for the "DH" family of cipher suites. Unfortunately, some existing 275 software cannot handle (or cannot easily handle) key lengths greater 276 than 1024 bits. The most common workaround for these systems is to 277 prefer the "ECDHE" family of cipher suites instead of the "DH" 278 family, then use longer keys. Key lengths of at least 2048 bits are 279 RECOMMENDED, since they are estimated to be roughly equivalent to 280 112-bit symmetric keys and might be sufficient for at least the next 281 10 years. 283 In addition to 2048-bit server certificates, the use of SHA-256 284 fingerprints is RECOMMENDED (see [CAB-Baseline] for more details). 285 Clients SHOULD indicate to servers that they request SHA-256, by 286 using the "Signature Algorithms" extension defined in TLS 1.2. 288 Note: The foregoing recommendations are preliminary and will likely 289 be corrected and enhanced in a future version of this document. 291 3.6. Compression 293 Implementations and deployments SHOULD disable TLS-level compression 294 ([RFC5246], Sec. 6.2.2), because it has been subject to security 295 attacks. 297 3.7. Session Resumption 299 If TLS session resumption is used, care ought to be taken to do so 300 safely. In particular, the resumption information (either session 301 IDs [RFC5246] or session tickets [RFC5077]) needs to be authenticated 302 and encrypted to prevent modification or eavesdropping by an 303 attacker. For session tickets, a strong cipher suite MUST be used 304 when encrypting the ticket (as least as strong as the main TLS cipher 305 suite); ticket keys MUST be changed regularly, e.g. once every week, 306 so as not to negate the effect of forward secrecy. Session ticket 307 validity SHOULD be limited to a reasonable duration (e.g. 1 day), so 308 as not to negate the benefits of forward secrecy. 310 3.8. Renegotiation 312 Where handshake renegotiation is implemented, both clients and 313 servers MUST implement the renegotiation_info extension, as defined 314 in [RFC5746]. 316 4. Detailed Guidelines 318 The following sections provide more detailed information about the 319 recommendations listed above. 321 4.1. Cipher Suite Negotiation Details 323 Clients SHOULD include TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 as the 324 first proposal to any server, unless they have prior knowledge that 325 the server cannot respond to a TLS 1.2 client_hello message. 327 Servers SHOULD prefer this cipher suite (or a similar but stronger 328 one) whenever it is proposed, even if it is not the first proposal. 330 Both clients and servers SHOULD include the "Supported Elliptic 331 Curves" extension [RFC4492]. 333 Clients are of course free to offer stronger cipher suites, e.g. 334 using AES-256; when they do, the server SHOULD prefer the stronger 335 cipher suite unless there are compelling reasons (e.g., seriously 336 degraded performance) to choose otherwise. 338 Note that other profiles of TLS 1.2 exist that use different cipher 339 suites. For example, [RFC6460] defines a profile that uses the 340 TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 and 341 TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384 cipher suites. 343 This document is not an application profile standard, in the sense of 344 Sec. 9 of [RFC5246]. As a result, clients and servers are still 345 required to support the TLS mandatory cipher suite, 346 TLS_RSA_WITH_AES_128_CBC_SHA. 348 4.2. Alternative Cipher Suites 350 Elliptic Curves Cryptography is not universally deployed for several 351 reasons, including its complexity compared to modular arithmetic and 352 longstanding IPR concerns. On the other hand, there are two related 353 issues hindering effective use of modular Diffie-Hellman cipher 354 suites in TLS: 356 o There are no protocol mechanisms to negotiate the DH groups or 357 parameter lengths supported by client and server. 359 o There are widely deployed client implementations that reject 360 received DH parameters, if they are longer than 1024 bits. 362 We note that with DHE and ECDHE cipher suites, the TLS master key 363 only depends on the Diffie Hellman parameters and not on the strength 364 the the RSA certificate; moreover, 1024 bits DH parameters are 365 generally considered insufficient at this time. 367 Because of the above, we recommend using (in priority order): 369 1. Elliptic Curve DHE with negotiated parameters [RFC5289] 371 2. TLS_DHE_RSA_WITH_AES_128_GCM_SHA256 [RFC5288], with 2048-bit 372 Diffie-Hellman parameters 374 3. The same cipher suite, with 1024-bit parameters. 376 With modular ephemeral DH, deployers SHOULD carefully evaluate 377 interoperability vs. security considerations when configuring their 378 TLS endpoints. 380 5. IANA Considerations 382 This document requests no actions of IANA. 384 6. Security Considerations 386 6.1. AES-GCM 388 Please refer to [RFC5246], Sec. 11 for general security 389 considerations when using TLS 1.2, and to [RFC5288], Sec. 6 for 390 security considerations that apply specifically to AES-GCM when used 391 with TLS. 393 6.2. Forward Secrecy 395 Forward secrecy (also often called Perfect Forward Secrecy or "PFS") 396 is a defense against an attacker who records encrypted conversations 397 where the session keys are only encrypted with the communicating 398 parties' long-term keys. Should the attacker be able to obtain these 399 long-term keys at some point later in the future, he will be able to 400 decrypt the session keys and thus the entire conversation. In the 401 context of TLS and DTLS, such compromise of long-term keys is not 402 entirely implausible. It can happen, for example, due to: 404 o A client or server being attacked by some other attack vector, and 405 the private key retrieved. 407 o A long-term key retrieved from a device that has been sold or 408 otherwise decommissioned without prior wiping. 410 o A long-term key used on a device as a default key [Heninger2012]. 412 o A key generated by a Trusted Third Party like a CA, and later 413 retrieved from it either by extortion or compromise 414 [Soghoian2011]. 416 o A cryptographic break-through, or the use of asymmetric keys with 417 insufficient length [Kleinjung2010]. 419 PFS ensures in such cases that the session keys cannot be determined 420 even by an attacker who obtains the long-term keys some time after 421 the conversation. It also protects against an attacker who is in 422 possession of the long-term keys, but remains passive during the 423 conversation. 425 PFS is generally achieved by using the Diffie-Hellman scheme to 426 derive session keys. The Diffie-Hellman scheme has both parties 427 maintain private secrets and send parameters over the network as 428 modular powers over certain cyclic groups. The properties of the so- 429 called Discrete Logarithm Problem (DLP) allow to derive the session 430 keys without an eavesdropper being able to do so. There is currently 431 no known attack against DLP if sufficiently large parameters are 432 chosen. 434 Unfortunately, many TLS/DTLS cipher suites were defined that do not 435 enable PFS, e.g. TLS_RSA_WITH_AES_256_CBC_SHA256. We thus advocate 436 strict use of PFS-only ciphers. 438 6.3. Certificate Revocation 440 Unfortunately there is currently no effective, Internet-scale 441 mechanism to affect certificate revocation: 443 o Certificate Revocation Lists (CRLs) are non-scalable and therefore 444 often unused. 446 o The On-Line Certification Status Prorocol (OCSP) presents both 447 scaling and privacy issues when used for heavy traffic Web 448 servers. In addition, clients typically "soft-fail", meaning they 449 do not abort the TLS connection if the OCSP server does not 450 respond. 452 o OCSP stapling (Sec. 8 of [RFC6066]) resolves the operational 453 issues with OCSP, but is still ineffective in the presence of a 454 MITM atacker, because they can simply ignore the client's request. 456 o Proprietary mechanisms that embed revocation lists in the Web 457 browser's configuration database cannot scale beyond a small 458 number of the most heavily used Web servers. 460 The current consensus appears to be that OCSP stapling, combined with 461 a "must staple" mechanism similar to HSTS, would finally resolve this 462 problem. But such a mechanism has not been standardized yet. 464 7. Acknowledgements 466 We would like to thank Stephen Farrell, Simon Josefsson, Johannes 467 Merkle, Yoav Nir, Kenny Paterson, Patrick Pelletier, Tom Ritter and 468 Rich Salz for their review. Thanks to Brian Smith whose "browser 469 cipher suites" page is a great resource. Finally, thanks to all 470 others who commented on the TLS and other lists and are not mentioned 471 here by name. 473 8. References 475 8.1. Normative References 477 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 478 Requirement Levels", BCP 14, RFC 2119, March 1997. 480 [RFC4492] Blake-Wilson, S., Bolyard, N., Gupta, V., Hawk, C., and B. 481 Moeller, "Elliptic Curve Cryptography (ECC) Cipher Suites 482 for Transport Layer Security (TLS)", RFC 4492, May 2006. 484 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 485 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 487 [RFC5288] Salowey, J., Choudhury, A., and D. McGrew, "AES Galois 488 Counter Mode (GCM) Cipher Suites for TLS", RFC 5288, 489 August 2008. 491 [RFC5289] Rescorla, E., "TLS Elliptic Curve Cipher Suites with 492 SHA-256/384 and AES Galois Counter Mode (GCM)", RFC 5289, 493 August 2008. 495 [RFC5746] Rescorla, E., Ray, M., Dispensa, S., and N. Oskov, 496 "Transport Layer Security (TLS) Renegotiation Indication 497 Extension", RFC 5746, February 2010. 499 [RFC6176] Turner, S. and T. Polk, "Prohibiting Secure Sockets Layer 500 (SSL) Version 2.0", RFC 6176, March 2011. 502 [RFC7027] Merkle, J. and M. Lochter, "Elliptic Curve Cryptography 503 (ECC) Brainpool Curves for Transport Layer Security 504 (TLS)", RFC 7027, October 2013. 506 8.2. Informative References 508 [CAB-Baseline] 509 "Baseline Requirements for the Issuance and Management of 510 Publicly-Trusted Certificates Version 1.1.6", 2013, 511 . 513 [Heninger2012] 514 Heninger, N., Durumeric, Z., Wustrow, E., and J. 515 Halderman, "Mining Your Ps and Qs: Detection of Widespread 516 Weak Keys in Network Devices", Usenix Security Symposium 517 2012, 2012. 519 [I-D.popov-tls-prohibiting-rc4] 520 Popov, A., "Prohibiting RC4 Cipher Suites", draft-popov- 521 tls-prohibiting-rc4-02 (work in progress), April 2014. 523 [I-D.sheffer-uta-tls-attacks] 524 Sheffer, Y., Holz, R., and P. Saint-Andre, "Summarizing 525 Current Attacks on TLS and DTLS", draft-sheffer-uta-tls- 526 attacks-00 (work in progress), February 2014. 528 [Kleinjung2010] 529 Kleinjung, T., "Factorization of a 768-Bit RSA Modulus", 530 CRYPTO 10, 2010. 532 [RFC2246] Dierks, T. and C. Allen, "The TLS Protocol Version 1.0", 533 RFC 2246, January 1999. 535 [RFC4346] Dierks, T. and E. Rescorla, "The Transport Layer Security 536 (TLS) Protocol Version 1.1", RFC 4346, April 2006. 538 [RFC5077] Salowey, J., Zhou, H., Eronen, P., and H. Tschofenig, 539 "Transport Layer Security (TLS) Session Resumption without 540 Server-Side State", RFC 5077, January 2008. 542 [RFC5116] McGrew, D., "An Interface and Algorithms for Authenticated 543 Encryption", RFC 5116, January 2008. 545 [RFC6066] Eastlake, D., "Transport Layer Security (TLS) Extensions: 546 Extension Definitions", RFC 6066, January 2011. 548 [RFC6101] Freier, A., Karlton, P., and P. Kocher, "The Secure 549 Sockets Layer (SSL) Protocol Version 3.0", RFC 6101, 550 August 2011. 552 [RFC6460] Salter, M. and R. Housley, "Suite B Profile for Transport 553 Layer Security (TLS)", RFC 6460, January 2012. 555 [RFC6797] Hodges, J., Jackson, C., and A. Barth, "HTTP Strict 556 Transport Security (HSTS)", RFC 6797, November 2012. 558 [Soghoian2011] 559 Soghoian, C. and S. Stamm, "Certified lies: Detecting and 560 defeating government interception attacks against SSL.", 561 Proc. 15th Int. Conf. Financial Cryptography and Data 562 Security , 2011. 564 Appendix A. Appendix: Change Log 566 Note to RFC Editor: please remove this section before publication. 568 A.1. draft-ietf-tls-bcp-01 570 o Clarified that specific TLS-using protocols may have stricter 571 requirements. 573 o Changed TLS 1.0 from MAY to SHUOLD NOT. 575 o Added discussion of "optional TLS" and HSTS. 577 o Recommended use of the Signature Algorithm and Renegotiation Info 578 extensions. 580 o Use of a strong cipher for a resumption ticket: changed SHOULD to 581 MUST. 583 o Added an informational discussion of certificate revocation, but 584 no recommendations. 586 A.2. draft-ietf-tls-bcp-00 588 o Initial WG version, with only updated references. 590 A.3. draft-sheffer-tls-bcp-02 592 o Reorganized the content to focus on recommendations. 594 o Moved description of attacks to a separate document (draft- 595 sheffer-uta-tls-attacks). 597 o Strengthened recommendations regarding session resumption. 599 A.4. draft-sheffer-tls-bcp-01 601 o Clarified our motivation in the introduction. 603 o Added a section justifying the need for PFS. 605 o Added recommendations for RSA and DH parameter lengths. Moved 606 from DHE to ECDHE, with a discussion on whether/when DHE is 607 appropriate. 609 o Recommendation to avoid fallback to SSLv3. 611 o Initial information about browser support - more still needed! 612 o More clarity on compression. 614 o Client can offer stronger cipher suites. 616 o Discussion of the regular TLS mandatory cipher suite. 618 A.5. draft-sheffer-tls-bcp-00 620 o Initial version. 622 Authors' Addresses 624 Yaron Sheffer 625 Porticor 626 29 HaHarash St. 627 Hod HaSharon 4501303 628 Israel 630 Email: yaronf.ietf@gmail.com 632 Ralph Holz 633 Technische Universitaet Muenchen 634 Boltzmannstr. 3 635 Garching 85748 636 Germany 638 Email: holz@net.in.tum.de 640 Peter Saint-Andre 641 &yet 643 Email: ietf@stpeter.im