idnits 2.17.1 draft-ietf-sasl-scram-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- 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 (July 30, 2009) is 5377 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) == Missing Reference: 'I-D.ietf-sasl-gs2' is mentioned on line 1065, but not defined == Missing Reference: 'RFC2743' is mentioned on line 1070, but not defined == Missing Reference: 'RFC4121' is mentioned on line 1079, but not defined == Missing Reference: 'RFC3962' is mentioned on line 1073, but not defined == Missing Reference: 'RFC4401' is mentioned on line 1084, but not defined == Missing Reference: 'RFC4402' is mentioned on line 1088, but not defined ** Obsolete undefined reference: RFC 4402 (Obsoleted by RFC 7802) == Missing Reference: 'RFC4086' is mentioned on line 1076, but not defined == Missing Reference: 'RFCXXXX' is mentioned on line 876, but not defined ** Downref: Normative reference to an Informational RFC: RFC 2104 ** Downref: Normative reference to an Informational RFC: RFC 3174 ** Obsolete normative reference: RFC 3454 (Obsoleted by RFC 7564) ** Obsolete normative reference: RFC 4013 (Obsoleted by RFC 7613) -- Obsolete informational reference (is this intentional?): RFC 2898 (Obsoleted by RFC 8018) -- Obsolete informational reference (is this intentional?): RFC 5246 (Obsoleted by RFC 8446) Summary: 6 errors (**), 0 flaws (~~), 9 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 NETWORK WORKING GROUP A. Menon-Sen 3 Internet-Draft Oryx Mail Systems GmbH 4 Intended status: Standards Track A. Melnikov 5 Expires: January 31, 2010 Isode Ltd 6 C. Newman 7 N. Williams 8 Sun Microsystems 9 July 30, 2009 11 Salted Challenge Response (SCRAM) SASL Mechanism 12 draft-ietf-sasl-scram-03.txt 14 Status of this Memo 16 This Internet-Draft is submitted to IETF in full conformance with the 17 provisions of BCP 78 and BCP 79. 19 Internet-Drafts are working documents of the Internet Engineering 20 Task Force (IETF), its areas, and its working groups. Note that 21 other groups may also distribute working documents as Internet- 22 Drafts. 24 Internet-Drafts are draft documents valid for a maximum of six months 25 and may be updated, replaced, or obsoleted by other documents at any 26 time. It is inappropriate to use Internet-Drafts as reference 27 material or to cite them other than as "work in progress." 29 The list of current Internet-Drafts can be accessed at 30 http://www.ietf.org/ietf/1id-abstracts.txt. 32 The list of Internet-Draft Shadow Directories can be accessed at 33 http://www.ietf.org/shadow.html. 35 This Internet-Draft will expire on January 31, 2010. 37 Copyright Notice 39 Copyright (c) 2009 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 in effect on the date of 44 publication of this document (http://trustee.ietf.org/license-info). 45 Please review these documents carefully, as they describe your rights 46 and restrictions with respect to this document. 48 Abstract 50 The secure authentication mechanism most widely deployed and used by 51 Internet application protocols is the transmission of clear-text 52 passwords over a channel protected by Transport Layer Security (TLS). 53 There are some significant security concerns with that mechanism, 54 which could be addressed by the use of a challenge response 55 authentication mechanism protected by TLS. Unfortunately, the 56 challenge response mechanisms presently on the standards track all 57 fail to meet requirements necessary for widespread deployment, and 58 have had success only in limited use. 60 This specification describes a family of Simple Authentication and 61 Security Layer (SASL, RFC 4422) authentication mechanisms called the 62 Salted Challenge Response Authentication Mechanism (SCRAM), which 63 addresses the security concerns and meets the deployability 64 requirements. When used in combination with TLS or an equivalent 65 security layer, a mechanism from this family could improve the 66 status-quo for application protocol authentication and provide a 67 suitable choice for a mandatory-to-implement mechanism for future 68 application protocol standards. 70 Table of Contents 72 1. Conventions Used in This Document . . . . . . . . . . 4 73 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . 4 74 1.2. Notation . . . . . . . . . . . . . . . . . . . . . . . 5 75 2. Introduction . . . . . . . . . . . . . . . . . . . . . 7 76 3. SCRAM Algorithm Overview . . . . . . . . . . . . . . . 9 77 4. SCRAM Mechanism Names . . . . . . . . . . . . . . . . 10 78 5. SCRAM Authentication Exchange . . . . . . . . . . . . 11 79 5.1. SCRAM Attributes . . . . . . . . . . . . . . . . . . . 12 80 6. Channel Binding . . . . . . . . . . . . . . . . . . . 15 81 6.1. Default Channel Binding . . . . . . . . . . . . . . . 16 82 7. Formal Syntax . . . . . . . . . . . . . . . . . . . . 17 83 8. SCRAM as a GSS-API Mechanism . . . . . . . . . . . . . 20 84 8.1. GSS-API Principal Name Types for SCRAM . . . . . . . . 20 85 8.2. GSS-API Per-Message Tokens for SCRAM . . . . . . . . . 20 86 8.3. GSS_Pseudo_random() for SCRAM . . . . . . . . . . . . 21 87 9. Security Considerations . . . . . . . . . . . . . . . 22 88 10. IANA Considerations . . . . . . . . . . . . . . . . . 24 89 11. Acknowledgements . . . . . . . . . . . . . . . . . . . 26 90 Appendix A. Other Authentication Mechanisms . . . . . . . . . . . 27 91 Appendix B. Design Motivations . . . . . . . . . . . . . . . . . . 28 92 Appendix C. Internet-Draft Change History . . . . . . . . . . . . 29 93 12. References . . . . . . . . . . . . . . . . . . . . . . 31 94 12.1. Normative References . . . . . . . . . . . . . . . . . 31 95 12.2. Normative References for GSS-API implementors . . . . 31 96 12.3. Informative References . . . . . . . . . . . . . . . . 32 97 Authors' Addresses . . . . . . . . . . . . . . . . . . 34 99 1. Conventions Used in This Document 101 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 102 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 103 document are to be interpreted as described in [RFC2119]. 105 Formal syntax is defined by [RFC5234] including the core rules 106 defined in Appendix B of [RFC5234]. 108 Example lines prefaced by "C:" are sent by the client and ones 109 prefaced by "S:" by the server. If a single "C:" or "S:" label 110 applies to multiple lines, then the line breaks between those lines 111 are for editorial clarity only, and are not part of the actual 112 protocol exchange. 114 1.1. Terminology 116 This document uses several terms defined in [RFC4949] ("Internet 117 Security Glossary") including the following: authentication, 118 authentication exchange, authentication information, brute force, 119 challenge-response, cryptographic hash function, dictionary attack, 120 eavesdropping, hash result, keyed hash, man-in-the-middle, nonce, 121 one-way encryption function, password, replay attack and salt. 122 Readers not familiar with these terms should use that glossary as a 123 reference. 125 Some clarifications and additional definitions follow: 127 o Authentication information: Information used to verify an identity 128 claimed by a SCRAM client. The authentication information for a 129 SCRAM identity consists of salt, iteration count, the "StoredKey" 130 and "ServerKey" (as defined in the algorithm overview) for each 131 supported cryptographic hash function. 133 o Authentication database: The database used to look up the 134 authentication information associated with a particular identity. 135 For application protocols, LDAPv3 (see [RFC4510]) is frequently 136 used as the authentication database. For network-level protocols 137 such as PPP or 802.11x, the use of RADIUS is more common. 139 o Base64: An encoding mechanism defined in [RFC4648] which converts 140 an octet string input to a textual output string which can be 141 easily displayed to a human. The use of base64 in SCRAM is 142 restricted to the canonical form with no whitespace. 144 o Octet: An 8-bit byte. 146 o Octet string: A sequence of 8-bit bytes. 148 o Salt: A random octet string that is combined with a password 149 before applying a one-way encryption function. This value is used 150 to protect passwords that are stored in an authentication 151 database. 153 1.2. Notation 155 The pseudocode description of the algorithm uses the following 156 notations: 158 o ":=": The variable on the left hand side represents the octet 159 string resulting from the expression on the right hand side. 161 o "+": Octet string concatenation. 163 o "[ ]": A portion of an expression enclosed in "[" and "]" may not 164 be included in the result under some circumstances. See the 165 associated text for a description of those circumstances. 167 o HMAC(key, str): Apply the HMAC keyed hash algorithm (defined in 168 [RFC2104]) using the octet string represented by "key" as the key 169 and the octet string "str" as the input string. The size of the 170 result is the hash result size for the hash function in use. For 171 example, it is 20 octets for SHA-1 (see [RFC3174]). 173 o H(str): Apply the cryptographic hash function to the octet string 174 "str", producing an octet string as a result. The size of the 175 result depends on the hash result size for the hash function in 176 use. 178 o XOR: Apply the exclusive-or operation to combine the octet string 179 on the left of this operator with the octet string on the right of 180 this operator. The length of the output and each of the two 181 inputs will be the same for this use. 183 o Hi(str, salt): 185 U0 := HMAC(str, salt + INT(1)) 186 U1 := HMAC(str, U0) 187 U2 := HMAC(str, U1) 188 ... 189 Ui-1 := HMAC(str, Ui-2) 190 Ui := HMAC(str, Ui-1) 192 Hi := U0 XOR U1 XOR U2 XOR ... XOR Ui 193 where "i" is the iteration count, "+" is the string concatenation 194 operator and INT(g) is a four-octet encoding of the integer g, 195 most significant octet first. 197 o This is, essentially, PBKDF2 [RFC2898] with HMAC() as the PRF and 198 with dkLen == output length of HMAC() == output length of H(). 200 2. Introduction 202 This specification describes a family of authentication mechanisms 203 called the Salted Challenge Response Authentication Mechanism (SCRAM) 204 which addresses the requirements necessary to deploy a challenge- 205 response mechanism more widely than past attempts. When used in 206 combination with Transport Layer Security (TLS, see [RFC5246]) or an 207 equivalent security layer, a mechanism from this family could improve 208 the status-quo for application protocol authentication and provide a 209 suitable choice for a mandatory-to-implement mechanism for future 210 application protocol standards. 212 For simplicity, this family of mechanisms does not presently include 213 negotiation of a security layer [RFC4422]. It is intended to be used 214 with an external security layer such as that provided by TLS or SSH, 215 with optional channel binding [RFC5056] to the external security 216 layer. 218 SCRAM is specified herein as a pure Simple Authentication and 219 Security Layer (SASL) [RFC4422] mechanism, but it conforms to the new 220 bridge between SASL and the Generic Security Services Application 221 Programming Interface (GSS-API) called "GS2" [I-D.ietf-sasl-gs2]. 222 This means that this document defines both, a SASL mechanism and a 223 GSS-API mechanism. 225 SCRAM provides the following protocol features: 227 o The authentication information stored in the authentication 228 database is not sufficient by itself to impersonate the client. 229 The information is salted to prevent a pre-stored dictionary 230 attack if the database is stolen. 232 o The server does not gain the ability to impersonate the client to 233 other servers (with an exception for server-authorized proxies). 235 o The mechanism permits the use of a server-authorized proxy without 236 requiring that proxy to have super-user rights with the back-end 237 server. 239 o Mutual authentication is supported, but only the client is named 240 (i.e., the server has no name). 242 For an in-depth discussion of why other challenge response mechanisms 243 are not considered sufficient, see appendix A. For more information 244 about the motivations behind the design of this mechanism, see 245 appendix B. 247 Comments regarding this draft may be sent either to the 248 ietf-sasl@imc.org mailing list or to the authors. 250 3. SCRAM Algorithm Overview 252 Note that this section omits some details, such as client and server 253 nonces. See Section 5 for more details. 255 To begin with, the SCRAM client is in possession of a username and 256 password. It sends the username to the server, which retrieves the 257 corresponding authentication information, i.e. a salt, StoredKey, 258 ServerKey and the iteration count i. (Note that a server 259 implementation may chose to use the same iteration count for all 260 accounts.) The server sends the salt and the iteration count to the 261 client, which then computes the following values and sends a 262 ClientProof to the server: 264 SaltedPassword := Hi(password, salt) 265 ClientKey := HMAC(SaltedPassword, "Client Key") 266 StoredKey := H(ClientKey) 267 AuthMessage := client-first-message-bare + "," + 268 server-first-message + "," + 269 client-final-message-without-proof 270 ClientSignature := HMAC(StoredKey, AuthMessage) 271 ClientProof := ClientKey XOR ClientSignature 272 ServerKey := HMAC(SaltedPassword, "Server Key") 273 ServerSignature := HMAC(ServerKey, AuthMessage) 275 The server authenticates the client by computing the ClientSignature, 276 exclusive-ORing that with the ClientProof to recover the ClientKey 277 and verifying the correctness of the ClientKey by applying the hash 278 function and comparing the result to the StoredKey. If the ClientKey 279 is correct, this proves that the client has access to the user's 280 password. 282 Similarly, the client authenticates the server by computing the 283 ServerSignature and comparing it to the value sent by the server. If 284 the two are equal, it proves that the server had access to the user's 285 ServerKey. 287 The AuthMessage is computed by concatenating messages from the 288 authentication exchange. The format of these messages is defined in 289 Section 7. 291 4. SCRAM Mechanism Names 293 A SCRAM mechanism name is a string "SCRAM-" followed by the 294 uppercased name of the underlying hash function taken from the IANA 295 "Hash Function Textual Names" registry (see http://www.iana.org), 296 optionally followed by the suffix "-PLUS" (see below). Note that 297 SASL mechanism names are limited to 20 characters, which means that 298 only hash function names with lengths shorter or equal to 9 299 characters (20-length("SCRAM-")-length("-PLUS") can be used. For 300 cases when the underlying hash function name is longer than 9 301 characters, an alternative 9 character (or shorter) name can be used 302 to construct the corresponding SCRAM mechanism name, as long as this 303 alternative name doesn't conflict with any other hash function name 304 from the IANA "Hash Function Textual Names" registry. 306 For interoperability, all SCRAM clients and servers MUST implement 307 the SCRAM-SHA-1 authentication mechanism, i.e. an authentication 308 mechanism from the SCRAM family that uses the SHA-1 hash function as 309 defined in [RFC3174]. 311 The "-PLUS" suffix is used only when the server supports channel 312 binding to the external channel. In this case the server will 313 advertise both, SCRAM-SHA-1 and SCRAM-SHA-1-PLUS, otherwise the 314 server will advertise only SCRAM-SHA-1. The "-PLUS" exists to allow 315 negotiation of the use of channel binding. See Section 6. 317 5. SCRAM Authentication Exchange 319 SCRAM is a SASL mechanism whose client response and server challenge 320 messages are text-based messages containing one or more attribute- 321 value pairs separated by commas. Each attribute has a one-letter 322 name. The messages and their attributes are described in 323 Section 5.1, and defined in Section 7. 325 This is a simple example of a SCRAM-SHA-1 authentication exchange 326 when the client doesn't support channel bindings: 328 C: n,,n=Chris Newman,r=ClientNonce 329 S: r=ClientNonceServerNonce,s=PxR/wv+epq,i=128 330 C: c=biwsCg==,r=ClientNonceServerNonce,p=WxPv/siO5l+qxN4 331 S: v=WxPv/siO5l+qxN4 333 [[anchor5: Note that the all hashes above are fake and will be fixed 334 during AUTH48.]] 336 With channel-binding data sent by the client this might look like 337 this (see [tls-server-end-point] for the definition of tls-server- 338 end-point TLS channel binding): 340 C: p=tls-server-end-point,,n=Chris Newman,r=ClientNonce 341 S: r=ClientNonceServerNonce,s=PxR/wv+epq,i=128 342 C: c=cD10bHMtc2VydmVyLWVuZC1wb2ludCwsy1hFtXOnZ+ySrQM6srFp 343 l/77uqvtxrg7nBY1BetEr/g=,r=ClientNonceServerNonce,p=Wx 344 Pv/siO5l+qxN4 345 S: v=WxPv/siO5l+qxN4 347 [[anchor6: Note that all hashes above are fake and will be fixed 348 during AUTH48.]] 350 First, the client sends a message containing: 352 o a GS2 header consisting of a flag indicating whether channel 353 binding is supported-but-not-used, not supported, or used, and an 354 optional SASL authorization identity; 356 o SCRAM username and a random, unique nonce attributes. 358 Note that the client's first message will always start with "n", "y" 359 or "p", otherwise the message is invalid and authentication MUST 360 fail. This is important, as it allows for GS2 extensibility (e.g., 361 to add support for security layers). 363 In response, the server sends the user's iteration count i, the 364 user's salt, and appends its own nonce to the client-specified one. 365 The client then responds with the same nonce and a ClientProof 366 computed using the selected hash function as explained earlier. The 367 server verifies the nonce and the proof, verifies that the 368 authorization identity (if supplied by the client in the first 369 message) is authorized to act as the authentication identity, and, 370 finally, it responds with a ServerSignature, concluding the 371 authentication exchange. The client then authenticates the server by 372 computing the ServerSignature and comparing it to the value sent by 373 the server. If the two are different, the client MUST consider the 374 authentication exchange to be unsuccessful and it might have to drop 375 the connection. 377 5.1. SCRAM Attributes 379 This section describes the permissible attributes, their use, and the 380 format of their values. All attribute names are single US-ASCII 381 letters and are case-sensitive. 383 Note that the order of attributes in client or server messages is 384 fixed, with the exception of extension attributes (described by the 385 "extensions" ABNF production), which can appear in any order in the 386 designated positions. See the ABNF section for authoritative 387 reference. 389 o a: This is an optional attribute, and is part of the GS2 390 [I-D.ietf-sasl-gs2] bridge between the GSS-API and SASL. This 391 attribute specifies an authorization identity. A client may 392 include it in its first message to the server if it wants to 393 authenticate as one user, but subsequently act as a different 394 user. This is typically used by an administrator to perform some 395 management task on behalf of another user, or by a proxy in some 396 situations. 398 Upon the receipt of this value the server verifies its 399 correctness according to the used SASL protocol profile. 400 Failed verification results in failed authentication exchange. 402 If this attribute is omitted (as it normally would be), the 403 authorization identity is assumed to be derived from the 404 username specified with the (required) "n" attribute. 406 The server always authenticates the user specified by the "n" 407 attribute. If the "a" attribute specifies a different user, 408 the server associates that identity with the connection after 409 successful authentication and authorization checks. 411 The syntax of this field is the same as that of the "n" field 412 with respect to quoting of '=' and ','. 414 o n: This attribute specifies the name of the user whose password is 415 used for authentication (a.k.a. "authentication identity" 416 [RFC4422]). A client MUST include it in its first message to the 417 server. If the "a" attribute is not specified (which would 418 normally be the case), this username is also the identity which 419 will be associated with the connection subsequent to 420 authentication and authorization. 422 Before sending the username to the server, the client MUST 423 prepare the username using the "SASLPrep" profile [RFC4013] of 424 the "stringprep" algorithm [RFC3454]. If the preparation of 425 the username fails or results in an empty string, the client 426 SHOULD abort the authentication exchange (*). 428 (*) An interactive client can request a repeated entry of the 429 username value. 431 Upon receipt of the username by the server, the server SHOULD 432 prepare it using the "SASLPrep" profile [RFC4013] of the 433 "stringprep" algorithm [RFC3454]. If the preparation of the 434 username fails or results in an empty string, the server SHOULD 435 abort the authentication exchange. 437 The characters ',' or '=' in usernames are sent as '=2C' and 438 '=3D' respectively. If the server receives a username which 439 contains '=' not followed by either '2C' or '3D', then the 440 server MUST fail the authentication. 442 o m: This attribute is reserved for future extensibility. In this 443 version of SCRAM, its presence in a client or a server message 444 MUST cause authentication failure when the attribute is parsed by 445 the other end. 447 o r: This attribute specifies a sequence of random printable 448 characters excluding ',' which forms the nonce used as input to 449 the hash function. No quoting is applied to this string. As 450 described earlier, the client supplies an initial value in its 451 first message, and the server augments that value with its own 452 nonce in its first response. It is important that this value be 453 different for each authentication. The client MUST verify that 454 the initial part of the nonce used in subsequent messages is the 455 same as the nonce it initially specified. The server MUST verify 456 that the nonce sent by the client in the second message is the 457 same as the one sent by the server in its first message. 459 o c: This REQUIRED attribute specifies base64-encoded of a header 460 and the channel-binding data. It is sent by the client in its 461 second authentication message. The header consist of: 463 * the GS2 header from the client's first message (recall: a 464 channel binding flag and an optional authzid). This header is 465 going to include channel binding type prefix (see [RFC5056]), 466 if and only if the client is using channel binding; 468 * followed by the external channel's channel binding data, if and 469 only if the client is using channel binding. 471 o s: This attribute specifies the base64-encoded salt used by the 472 server for this user. It is sent by the server in its first 473 message to the client. 475 o i: This attribute specifies an iteration count for the selected 476 hash function and user, and MUST be sent by the server along with 477 the user's salt. 479 For SCRAM-SHA-1/SCRAM-SHA-1-PLUS SASL mechanism servers SHOULD 480 announce a hash iteration-count of at least 4096. Note that a 481 client implementation MAY cache SaltedPassword/ClientKey for 482 later reauthentication to the same service, as it is likely 483 that the server is going to advertise the same salt value upon 484 reauthentication. This might be useful for mobile clients 485 where CPU usage is a concern. 487 o p: This attribute specifies a base64-encoded ClientProof. The 488 client computes this value as described in the overview and sends 489 it to the server. 491 o v: This attribute specifies a base64-encoded ServerSignature. It 492 is sent by the server in its final message, and is used by the 493 client to verify that the server has access to the user's 494 authentication information. This value is computed as explained 495 in the overview. 497 6. Channel Binding 499 SCRAM supports channel binding to external secure channels, such as 500 TLS. Clients and servers may or may not support channel binding, 501 therefore the use of channel binding is negotiable. SCRAM does not 502 provide security layers, however, therefore it is imperative that 503 SCRAM provide integrity protection for the negotiation of channel 504 binding. 506 Use of channel binding is negotiated as follows: 508 o Servers SHOULD advertise both non-PLUS (SCRAM-) and 509 the PLUS-variant (SCRAM--PLUS) SASL mechanism 510 names. If the server cannot support channel binding, it MAY 511 advertise only the non-PLUS variant. If the server would never 512 succeed authentication of the non-PLUS variant due to policy 513 reasons, it MAY advertise only the PLUS-variant. 515 o If the client negotiates mechanisms then the client MUST select 516 SCRAM--PLUS if offered by the server and the client 517 wants to select SCRAM with the given hash function. Otherwise 518 (the client does not negotiate mechanisms), if the client has no 519 prior knowledge about mechanisms supported by the server and 520 wasn't explicitly configured to use a particular variant of the 521 SCRAM mechanism, then it MUST select only SCRAM- 522 (not suffixed with "-PLUS"). 524 o If the client supports channel binding and the server appears to 525 support it (i.e., the client sees SCRAM--PLUS), or 526 if the client wishes to use channel binding but the client does 527 not negotiate mechanisms, then the client MUST set the GS2 channel 528 binding flag to "p" in order to indicate the channel binding type 529 it is using and it MUST include the channel binding data for the 530 external channel in the computation of the "c=" attribute (see 531 Section 5.1). 533 o If the client supports channel binding but the server does not 534 appear to (i.e., the client did not see SCRAM-- 535 PLUS) then the client MUST either fail authentication or it MUST 536 choose the non-PLUS mechanism and set the GS2 channel binding flag 537 to "y" and MUST NOT include channel binding data for the external 538 channel in the computation of the "c=" attribute (see 539 Section 5.1). 541 o If the client does not support channel binding then the client 542 MUST set the GS2 channel binding flag to "n" and MUST NOT include 543 channel binding data for the external channel in the computation 544 of the "c=" attribute (see Section 5.1). 546 o Upon receipt of the client first message the server checks the GS2 547 channel binding flag (gs2-cb-flag). 549 * If the flag is set to "y" and the server supports channel 550 binding the server MUST fail authentication. This is because 551 if the client sets the GS2 channel binding flag set to "y" then 552 the client must have believed that the server did not support 553 channel binding -- if the server did in fact support channel 554 binding then this is an indication that there has been a 555 downgrade attack (e.g., an attacker changed the server's 556 mechanism list to exclude the -PLUS suffixed SCRAM mechanism 557 name(s)). 559 * If the channel binding flag was "p" and the server does not 560 support the indicated channel binding type then the server MUST 561 fail authentication. 563 The server MUST always validate the client's "c=" field. The server 564 does this by constructing the value of the "c=" attribute and then 565 checking that it matches the client's c= attribute value. 567 For more discussions of channel bindings, and the syntax of the 568 channel binding data for various security protocols, see [RFC5056]. 570 6.1. Default Channel Binding 572 A default channel binding type agreement process for all SASL 573 application protocols that do not provide their own channel binding 574 type agreement is provided as follows. 576 Clients and servers MUST implement the "tls-unique" [tls-unique] 577 channel binding type. Clients and servers SHOULD choose the highest- 578 layer/innermost end-to-end TLS channel as the channel to bind to. 580 Clients SHOULD choose the tls-unique channel binding type. 581 Conversely, clients MAY choose a different channel binding type based 582 on user input, configuration, or a future, as-yet undefined channel 583 binding type negotiation protocol. Servers MUST choose the channel 584 binding type indicated by the client, if they support it. 586 7. Formal Syntax 588 The following syntax specification uses the Augmented Backus-Naur 589 Form (ABNF) notation as specified in [RFC5234]. "UTF8-2", "UTF8-3" 590 and "UTF8-4" non-terminal are defined in [RFC3629]. 592 ALPHA = 593 DIGIT = 594 UTF8-2 = 595 UTF8-3 = 596 UTF8-4 = 598 attr-val = ALPHA "=" value 599 ;; Generic syntax of any attribute sent 600 ;; by server or client 602 value = 1*value-char 604 value-safe-char = %x01-2B / %x2D-3C / %x3E-7F / 605 UTF8-2 / UTF8-3 / UTF8-4 606 ;; UTF8-char except NUL, "=", and ",". 608 value-char = value-safe-char / "=" 610 base64-char = ALPHA / DIGIT / "/" / "+" 612 base64-4 = 4base64-char 614 base64-3 = 3base64-char "=" 616 base64-2 = 2base64-char "==" 618 base64 = *base64-4 [base64-3 / base64-2] 620 posit-number = %x31-39 *DIGIT 621 ;; A positive number 623 saslname = 1*(value-safe-char / "=2C" / "=3D") 624 ;; Conforms to 626 authzid = "a=" saslname 627 ;; Protocol specific. 629 cb-name = 1*(ALPHA / DIGIT / "." / "-") 630 ;; See RFC 5056 section 7. 631 ;; E.g. "tls-server-end-point" or 632 ;; "tls-unique" 634 gs2-cbind-flag = "p=" cb-name / "n" / "y" 635 ;; "n" -> client doesn't support channel binding 636 ;; "y" -> client does support channel binding 637 ;; but thinks the server does not. 638 ;; "p" -> client requires channel binding. 639 ;; The selected channel binding follows "p=". 641 gs2-header = gs2-cbind-flag "," [ authzid ] "," 642 ;; GS2 header for SCRAM 643 ;; (the actual GS2 header includes an optional 644 ;; flag to indicate that the GSS mechanism is not 645 ;; "standard" but since SCRAM is "standard" we 646 ;; don't include that flag). 648 username = "n=" saslname 649 ;; Usernames are prepared using SASLPrep. 651 reserved-mext = "m=" 1*(value-char) 652 ;; Reserved for signalling mandatory extensions. 653 ;; The exact syntax will be defined in 654 ;; the future. 656 channel-binding = "c=" base64 657 ;; base64 encoding of cbind-input 659 proof = "p=" base64 661 nonce = "r=" c-nonce [s-nonce] 662 ;; Second part provided by server. 664 c-nonce = value 666 s-nonce = value 668 salt = "s=" base64 670 verifier = "v=" base64 671 ;; base-64 encoded ServerSignature. 673 iteration-count = "i=" posit-number 674 ;; A positive number 676 client-first-message-bare = 677 [reserved-mext ","] 678 username "," nonce ["," extensions] 680 client-first-message = 681 gs2-header client-first-message-bare 683 server-first-message = 684 [reserved-mext ","] nonce "," salt "," 685 iteration-count ["," extensions] 687 client-final-message-without-proof = 688 channel-binding "," nonce ["," 689 extensions] 691 client-final-message = 692 client-final-message-without-proof "," proof 694 gss-server-error = "e=" value 695 server-final-message = gss-server-error / 696 verifier ["," extensions] 697 ;; The error message is only for the GSS-API 698 ;; form of SCRAM, and it is OPTIONAL to 699 ;; implement it. 701 extensions = attr-val *("," attr-val) 702 ;; All extensions are optional, 703 ;; i.e. unrecognized attributes 704 ;; not defined in this document 705 ;; MUST be ignored. 707 cbind-data = 1*OCTET 709 cbind-input = gs2-header [ cbind-data ] 710 ;; cbind-data MUST be present for 711 ;; gs2-cbind-flag of "p" and MUST be absent 712 ;; for "y" or "n". 714 8. SCRAM as a GSS-API Mechanism 716 This section and its sub-sections and all normative references of it 717 not referenced elsewhere in this document are INFORMATIONAL for SASL 718 implementors, but they are NORMATIVE for GSS-API implementors. 720 SCRAM is actually also GSS-API mechanism. The messages are the same, 721 but a) the GS2 header on the client's first message and channel 722 binding data is excluded when SCRAM is used as a GSS-API mechanism, 723 and b) the RFC2743 section 3.1 initial context token header is 724 prefixed to the client's first authentication message (context 725 token). 727 The GSS-API mechanism OID for SCRAM is (see Section 10). 729 8.1. GSS-API Principal Name Types for SCRAM 731 SCRAM does not name acceptors. Therefore only GSS_C_NO_NAME and 732 names of type GSS_C_NT_ANONYMOUS shall be allowed as the target name 733 input of GSS_Init_sec_context() when using a SCRAM mechanism. 735 SCRAM supports only a single name type for initiators: 736 GSS_C_NT_USER_NAME. GSS_C_NT_USER_NAME is the default name type for 737 SCRAM. 739 There is no name canonicalization procedure for SCRAM beyond applying 740 SASLprep as described in Section 5.1. 742 The query, display and exported name syntax for SCRAM principal names 743 is the same: there is no syntax -- SCRAM principal names are free- 744 form. (The exported name token does, of course, conform to [RFC2743] 745 section 3.2, but the "NAME" part of the token is just a SCRAM user 746 name.) 748 8.2. GSS-API Per-Message Tokens for SCRAM 750 The per-message tokens for SCRAM as a GSS-API mechanism SHALL be the 751 same as those for the Kerberos V GSS-API mechanism [RFC4121], using 752 the Kerberos V "aes128-cts-hmac-sha1-96" enctype [RFC3962]. 754 The 128-bit session key SHALL be derived by using the least 755 significant (right-most) 128 bits of HMAC(StoredKey, "GSS-API session 756 key" || ClientKey || AuthMessage). 758 SCRAM does support PROT_READY, and is PROT_READY on the initiator 759 side first upon receipt of the server's reply to the initial security 760 context token. 762 8.3. GSS_Pseudo_random() for SCRAM 764 The GSS_Pseudo_random() [RFC4401] for SCRAM SHALL be the same as for 765 the Kerberos V GSS-API mechanism [RFC4402]. There is no acceptor- 766 asserted sub-session key for SCRAM, thus GSS_C_PRF_KEY_FULL and 767 GSS_C_PRF_KEY_PARTIAL are equivalent for SCRAM's GSS_Pseudo_random(). 769 9. Security Considerations 771 If the authentication exchange is performed without a strong security 772 layer, then a passive eavesdropper can gain sufficient information to 773 mount an offline dictionary or brute-force attack which can be used 774 to recover the user's password. The amount of time necessary for 775 this attack depends on the cryptographic hash function selected, the 776 strength of the password and the iteration count supplied by the 777 server. An external security layer with strong encryption will 778 prevent this attack. 780 If the external security layer used to protect the SCRAM exchange 781 uses an anonymous key exchange, then the SCRAM channel binding 782 mechanism can be used to detect a man-in-the-middle attack on the 783 security layer and cause the authentication to fail as a result. 784 However, the man-in-the-middle attacker will have gained sufficient 785 information to mount an offline dictionary or brute-force attack. 786 For this reason, SCRAM includes the ability to increase the iteration 787 count over time. 789 If the authentication information is stolen from the authentication 790 database, then an offline dictionary or brute-force attack can be 791 used to recover the user's password. The use of salt mitigates this 792 attack somewhat by requiring a separate attack on each password. 793 Authentication mechanisms which protect against this attack are 794 available (e.g., the EKE class of mechanisms). 796 If an attacker obtains the authentication information from the 797 authentication repository and either eavesdrops on one authentication 798 exchange or impersonates a server, the attacker gains the ability to 799 impersonate that user to all servers providing SCRAM access using the 800 same hash function, password, iteration count and salt. For this 801 reason, it is important to use randomly-generated salt values. 803 SCRAM does not negotiate a hash function to use. Hash function 804 negotiation is left to the SASL mechanism negotiation. It is 805 important that clients be able to sort a locally available list of 806 mechanisms by preference so that the client may pick the most 807 preferred of a server's advertised mechanism list. This preference 808 order is not specified here as it is a local matter. The preference 809 order should include objective and subjective notions of mechanism 810 cryptographic strength (e.g., SCRAM with a successor to SHA-1 may be 811 preferred over SCRAM with SHA-1). 813 Note that to protect the SASL mechanism negotiation applications 814 normally must list the server mechs twice: once before and once after 815 authentication, the latter using security layers. Since SCRAM does 816 not provide security layers the only ways to protect the mechanism 817 negotiation are: a) use channel binding to an external channel, or b) 818 use an external channel that authenticates a user-provided server 819 name. 821 A hostile server can perform a computational denial-of-service attack 822 on clients by sending a big iteration count value. 824 See [RFC4086] for more information about generating randomness. 826 10. IANA Considerations 828 IANA is requested to add the following family of SASL mechanisms to 829 the SASL Mechanism registry established by [RFC4422]: 831 To: iana@iana.org 832 Subject: Registration of a new SASL family SCRAM 834 SASL mechanism name (or prefix for the family): SCRAM-* 835 Security considerations: Section 7 of [RFCXXXX] 836 Published specification (optional, recommended): [RFCXXXX] 837 Person & email address to contact for further information: 838 IETF SASL WG 839 Intended usage: COMMON 840 Owner/Change controller: IESG 841 Note: Members of this family must be explicitly registered 842 using the "IETF Consensus" registration procedure. 843 Reviews must be requested on the SASL WG mailing list. 845 "IETF Consensus" registration procedure MUST be used for registering 846 new mechanisms in this family. The SASL mailing list 847 (or a successor designated by the responsible 848 Security AD) MUST be used for soliciting reviews on such 849 registrations. 851 Note to future SCRAM- mechanism designers: each new SCRAM- SASL 852 mechanism MUST be explicitly registered with IANA and MUST comply 853 with SCRAM- mechanism naming convention defined in Section 4 of this 854 document. 856 IANA is requested to add the following entries to the SASL Mechanism 857 registry established by [RFC4422]: 859 To: iana@iana.org 860 Subject: Registration of a new SASL mechanism SCRAM-SHA-1 862 SASL mechanism name (or prefix for the family): SCRAM-SHA-1 863 Security considerations: Section 7 of [RFCXXXX] 864 Published specification (optional, recommended): [RFCXXXX] 865 Person & email address to contact for further information: 866 IETF SASL WG 867 Intended usage: COMMON 868 Owner/Change controller: IESG 869 Note: 871 To: iana@iana.org 872 Subject: Registration of a new SASL mechanism SCRAM-SHA-1-PLUS 874 SASL mechanism name (or prefix for the family): SCRAM-SHA-1-PLUS 875 Security considerations: Section 7 of [RFCXXXX] 876 Published specification (optional, recommended): [RFCXXXX] 877 Person & email address to contact for further information: 878 IETF SASL WG 879 Intended usage: COMMON 880 Owner/Change controller: IESG 881 Note: 883 This document also requests IANA to assign a GSS-API mechanism OID 884 for SCRAM. 886 11. Acknowledgements 888 This document benefited from discussions on the SASL WG mailing list. 889 The authors would like to specially thank Dave Cridland, Simon 890 Josefsson and Jeffrey Hutzelman for their contributions to this 891 document. 893 Appendix A. Other Authentication Mechanisms 895 The DIGEST-MD5 [I-D.ietf-sasl-digest-to-historic] mechanism has 896 proved to be too complex to implement and test, and thus has poor 897 interoperability. The security layer is often not implemented, and 898 almost never used; everyone uses TLS instead. For a more complete 899 list of problems with DIGEST-MD5 which lead to the creation of SCRAM 900 see [I-D.ietf-sasl-digest-to-historic]. 902 The CRAM-MD5 SASL mechanism, while widely deployed has also some 903 problems, in particular it is missing some modern SASL features such 904 as support for internationalized usernames and passwords, support for 905 passing of authorization identity, support for channel bindings. It 906 also doesn't support server authentication. For a more complete list 907 of problems with CRAM-MD5 see [I-D.ietf-sasl-crammd5-to-historic]. 909 The PLAIN [RFC4616] SASL mechanism allows a malicious server or 910 eavesdropper to impersonate the authenticating user to any other 911 server for which the user has the same password. It also sends the 912 password in the clear over the network, unless TLS is used. Server 913 authentication is not supported. 915 Appendix B. Design Motivations 917 The following design goals shaped this document. Note that some of 918 the goals have changed since the initial version of the document. 920 o The SASL mechanism has all modern SASL features: support for 921 internationalized usernames and passwords, support for passing of 922 authorization identity, support for channel bindings. 924 o The protocol supports mutual authentication. 926 o The authentication information stored in the authentication 927 database is not sufficient by itself to impersonate the client. 929 o The server does not gain the ability to impersonate the client to 930 other servers (with an exception for server-authorized proxies), 931 unless such other servers allow SCRAM authentication and use the 932 same salt and iteration count for the user. 934 o The mechanism is extensible, but [hopefully] not overengineered in 935 this respect. 937 o Easier to implement than DIGEST-MD5 in both clients and servers. 939 Appendix C. Internet-Draft Change History 941 (RFC Editor: Please delete everything after this point) 943 Changes since -10 945 o Converted the source for this I-D to XML. 947 o Added text to make SCRAM compliant with the new GS2 design. 949 o Added text on channel binding negotiation. 951 o Added text on channel binding, including a reference to RFC5056. 953 o Added text on SCRAM as a GSS-API mechanism. This noted as not 954 relevant to SASL-only implementors -- the normative references for 955 SCRAM as a GSS-API mechanism are segregated as well. 957 Changes since -07 959 o Updated References. 961 o Clarified purpose of the m= attribute. 963 o Fixed a problem with authentication/authorization identity's ABNF 964 not allowing for some characters. 966 o Updated ABNF for nonce to show client-generated and server- 967 generated parts. 969 o Only register SCRAM-SHA-1 with IANA and require explicit 970 registrations of all other SCRAM- mechanisms. 972 Changes since -06 974 o Removed hash negotiation from SCRAM and turned it into a family of 975 SASL mechanisms. 977 o Start using "Hash Function Textual Names" IANA registry for SCRAM 978 mechanism naming. 980 o Fixed definition of Hi(str, salt) to be consistent with [RFC2898]. 982 o Clarified extensibility of SCRAM: added m= attribute (for future 983 mandatory extensions) and specified that all unrecognized 984 attributes must be ignored. 986 Changes since -05 987 o Changed the mandatory to implement hash algorithm to SHA-1 (as per 988 WG consensus). 990 o Added text about use of SASLPrep for username canonicalization/ 991 validation. 993 o Clarified that authorization identity is canonicalized/verified 994 according to SASL protocol profile. 996 o Clarified that iteration count is per-user. 998 o Clarified how clients select the authentication function. 1000 o Added IANA registration for the new mechanism. 1002 o Added missing normative references (UTF-8, SASLPrep). 1004 o Various editorial changes based on comments from Hallvard B 1005 Furuseth, Nico William and Simon Josefsson. 1007 Changes since -04 1009 o Update Base64 and Security Glossary references. 1011 o Add Formal Syntax section. 1013 o Don't bother with "v=". 1015 o Make MD5 mandatory to implement. Suggest i=128. 1017 Changes since -03 1019 o Seven years have passed, in which it became clear that DIGEST-MD5 1020 suffered from unacceptably bad interoperability, so SCRAM-MD5 is 1021 now back from the dead. 1023 o Be hash agnostic, so MD5 can be replaced more easily. 1025 o General simplification. 1027 12. References 1029 12.1. Normative References 1031 [RFC2104] Krawczyk, H., Bellare, M., and R. Canetti, "HMAC: Keyed- 1032 Hashing for Message Authentication", RFC 2104, 1033 February 1997. 1035 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1036 Requirement Levels", BCP 14, RFC 2119, March 1997. 1038 [RFC3174] Eastlake, D. and P. Jones, "US Secure Hash Algorithm 1 1039 (SHA1)", RFC 3174, September 2001. 1041 [RFC3454] Hoffman, P. and M. Blanchet, "Preparation of 1042 Internationalized Strings ("stringprep")", RFC 3454, 1043 December 2002. 1045 [RFC3629] Yergeau, F., "UTF-8, a transformation format of ISO 1046 10646", STD 63, RFC 3629, November 2003. 1048 [RFC4013] Zeilenga, K., "SASLprep: Stringprep Profile for User Names 1049 and Passwords", RFC 4013, February 2005. 1051 [RFC4422] Melnikov, A. and K. Zeilenga, "Simple Authentication and 1052 Security Layer (SASL)", RFC 4422, June 2006. 1054 [RFC4648] Josefsson, S., "The Base16, Base32, and Base64 Data 1055 Encodings", RFC 4648, October 2006. 1057 [RFC5056] Williams, N., "On the Use of Channel Bindings to Secure 1058 Channels", RFC 5056, November 2007. 1060 [RFC5234] Crocker, D. and P. Overell, "Augmented BNF for Syntax 1061 Specifications: ABNF", STD 68, RFC 5234, January 2008. 1063 12.2. Normative References for GSS-API implementors 1065 [I-D.ietf-sasl-gs2] 1066 Josefsson, S. and N. Williams, "Using GSS-API Mechanisms 1067 in SASL: The GS2 Mechanism Family", draft-ietf-sasl-gs2-12 1068 (work in progress), April 2009. 1070 [RFC2743] Linn, J., "Generic Security Service Application Program 1071 Interface Version 2, Update 1", RFC 2743, January 2000. 1073 [RFC3962] Raeburn, K., "Advanced Encryption Standard (AES) 1074 Encryption for Kerberos 5", RFC 3962, February 2005. 1076 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 1077 Requirements for Security", BCP 106, RFC 4086, June 2005. 1079 [RFC4121] Zhu, L., Jaganathan, K., and S. Hartman, "The Kerberos 1080 Version 5 Generic Security Service Application Program 1081 Interface (GSS-API) Mechanism: Version 2", RFC 4121, 1082 July 2005. 1084 [RFC4401] Williams, N., "A Pseudo-Random Function (PRF) API 1085 Extension for the Generic Security Service Application 1086 Program Interface (GSS-API)", RFC 4401, February 2006. 1088 [RFC4402] Williams, N., "A Pseudo-Random Function (PRF) for the 1089 Kerberos V Generic Security Service Application Program 1090 Interface (GSS-API) Mechanism", RFC 4402, February 2006. 1092 [tls-unique] 1093 Zhu, L., "Registration of TLS unique channel binding 1094 (generic)", IANA http://www.iana.org/assignments/ 1095 channel-binding-types/tls-unique, July 2008. 1097 12.3. Informative References 1099 [I-D.ietf-sasl-crammd5-to-historic] 1100 Zeilenga, K., "CRAM-MD5 to Historic", 1101 draft-ietf-sasl-crammd5-to-historic-00 (work in progress), 1102 November 2008. 1104 [I-D.ietf-sasl-digest-to-historic] 1105 Melnikov, A., "Moving DIGEST-MD5 to Historic", 1106 draft-ietf-sasl-digest-to-historic-00 (work in progress), 1107 July 2008. 1109 [RFC2898] Kaliski, B., "PKCS #5: Password-Based Cryptography 1110 Specification Version 2.0", RFC 2898, September 2000. 1112 [RFC4510] Zeilenga, K., "Lightweight Directory Access Protocol 1113 (LDAP): Technical Specification Road Map", RFC 4510, 1114 June 2006. 1116 [RFC4616] Zeilenga, K., "The PLAIN Simple Authentication and 1117 Security Layer (SASL) Mechanism", RFC 4616, August 2006. 1119 [RFC4949] Shirey, R., "Internet Security Glossary, Version 2", 1120 RFC 4949, August 2007. 1122 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 1123 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 1125 [tls-server-end-point] 1126 Zhu, L., "Registration of TLS server end-point channel 1127 bindings", IANA http://www.iana.org/assignments/ 1128 channel-binding-types/tls-server-end-point, July 2008. 1130 Authors' Addresses 1132 Abhijit Menon-Sen 1133 Oryx Mail Systems GmbH 1135 Email: ams@oryx.com 1137 Alexey Melnikov 1138 Isode Ltd 1140 Email: Alexey.Melnikov@isode.com 1142 Chris Newman 1143 Sun Microsystems 1144 1050 Lakes Drive 1145 West Covina, CA 91790 1146 USA 1148 Email: chris.newman@sun.com 1150 Nicolas Williams 1151 Sun Microsystems 1152 5300 Riata Trace Ct 1153 Austin, TX 78727 1154 USA 1156 Email: Nicolas.Williams@sun.com