idnits 2.17.1 draft-ietf-httpauth-digest-06.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 obsoletes RFC2617, 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 == Line 560 has weird spacing: '...rameter that ...' -- The document date (April 9, 2014) is 3670 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) -- Possible downref: Normative reference to a draft: ref. 'BASIC' -- Obsolete informational reference (is this intentional?): RFC 2818 (Obsoleted by RFC 9110) Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 HTTPAuth Working Group R. Shekh-Yusef, Ed. 3 Internet-Draft D. Ahrens 4 Obsoletes: 2617 (if approved) Avaya 5 Intended Status: Standards Track S. Bremer 6 Expires: October 11, 2014 Netzkonform 7 April 9, 2014 9 HTTP Digest Access Authentication 10 draft-ietf-httpauth-digest-06 12 Abstract 14 HTTP provides a simple challenge-response authentication mechanism 15 that may be used by a server to challenge a client request and by a 16 client to provide authentication information. This document defines 17 the HTTP Digest Authentication scheme that may be used with the 18 authentication mechanism. 20 Status of this Memo 22 This Internet-Draft is submitted to IETF 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), its areas, and its working groups. Note that 27 other groups may also distribute working documents as 28 Internet-Drafts. 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 The list of current Internet-Drafts can be accessed at 36 http://www.ietf.org/1id-abstracts.html 38 The list of Internet-Draft Shadow Directories can be accessed at 39 http://www.ietf.org/shadow.html 41 Copyright and License 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 . . . . . . . . . . . . . . . . . . . . . . . . . . 4 59 1.1 Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 60 2 Syntax Convention . . . . . . . . . . . . . . . . . . . . . . . 4 61 2.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 4 62 2.2 Algorithm Variants . . . . . . . . . . . . . . . . . . . . . 4 63 2.3 ABNF . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 3 Digest Access Authentication Scheme . . . . . . . . . . . . . . 5 65 3.1 Overall Operation . . . . . . . . . . . . . . . . . . . . . 5 66 3.2 Representation of Digest Values . . . . . . . . . . . . . . 5 67 3.3 The WWW-Authenticate Response Header . . . . . . . . . . . . 5 68 3.4 The Authorization Request Header . . . . . . . . . . . . . . 8 69 3.4.1 Response . . . . . . . . . . . . . . . . . . . . . . . . 10 70 3.4.2 A1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 71 3.4.3 A2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 72 3.4.4 Username Hashing . . . . . . . . . . . . . . . . . . . . 11 73 3.4.5 Parameter Values and Quoted-String . . . . . . . . . . . 11 74 3.4.6 Various Considerations . . . . . . . . . . . . . . . . . 12 75 3.5 The Authentication-Info Header . . . . . . . . . . . . . . . 13 76 3.6 Digest Operation . . . . . . . . . . . . . . . . . . . . . . 15 77 3.7 Security Protocol Negotiation . . . . . . . . . . . . . . . 16 78 3.8 Proxy-Authenticate and Proxy-Authorization . . . . . . . . . 17 79 3.9 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 17 80 3.9.1 Example with SHA-256 and MD5 . . . . . . . . . . . . . . 17 81 3.9.2 Example with SHA-512-256, Charset, and Userhash . . . . 18 82 4 Internationalization . . . . . . . . . . . . . . . . . . . . . . 20 83 5 Security Considerations . . . . . . . . . . . . . . . . . . . . 20 84 5.1 Limitations . . . . . . . . . . . . . . . . . . . . . . . . 20 85 5.2 Authentication of Clients using Digest Authentication . . . 21 86 5.3 Limited Use Nonce Values . . . . . . . . . . . . . . . . . . 21 87 5.4 Replay Attacks . . . . . . . . . . . . . . . . . . . . . . . 22 88 5.5 Weakness Created by Multiple Authentication Schemes . . . . 23 89 5.6 Online dictionary attacks . . . . . . . . . . . . . . . . . 23 90 5.7 Man in the Middle . . . . . . . . . . . . . . . . . . . . . 23 91 5.8 Chosen plaintext attacks . . . . . . . . . . . . . . . . . . 24 92 5.9 Precomputed dictionary attacks . . . . . . . . . . . . . . . 24 93 5.10 Batch brute force attacks . . . . . . . . . . . . . . . . . 25 94 5.11 Spoofing by Counterfeit Servers . . . . . . . . . . . . . . 25 95 5.12 Storing passwords . . . . . . . . . . . . . . . . . . . . . 25 96 5.13 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 26 97 6 IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 27 98 6.1 HTTP Digest Hash Algorithms Registry . . . . . . . . . . . 27 99 6.2 Digest Scheme Registration . . . . . . . . . . . . . . . . 27 100 6.3 Authentication-Info Header Registration . . . . . . . . . . 27 101 7 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . 28 102 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 103 8.1 Normative References . . . . . . . . . . . . . . . . . . . . 29 104 8.2 Informative References . . . . . . . . . . . . . . . . . . . 30 105 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 30 107 1 Introduction 109 HTTP provides a simple challenge-response authentication mechanism 110 that may be used by a server to challenge a client request and by a 111 client to provide authentication information. This document defines 112 the HTTP Digest Authentication scheme that may be used with the 113 authentication mechanism. 115 The details of the challenge-response authentication mechanism are 116 specified in the [HTTP-P7] document. 118 The combination of this document with Basic [BASIC] and [HTTP-P7] 119 obsolete RFC2617. 121 1.1 Terminology 123 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 124 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 125 document are to be interpreted as described in RFC 2119 [RFC2119]. 127 2 Syntax Convention 129 2.1 Examples 131 In the interest of clarity and readability, the extended parameters 132 or the headers and parameters in the examples in this document might 133 be broken into multiple lines. Any line that is indented in this 134 document is a continuation of the preceding line. 136 2.2 Algorithm Variants 138 When used with the Digest mechanism, each one of the algorithms has 139 two variants: Session variant and non-Session variant. 141 The non-Session variant is denoted by "", e.g. "SHA-256", 142 and the Session variant is denoted by "-sess", e.g. "SHA- 143 256-sess". 145 2.3 ABNF 147 This specification uses the Augmented Backus-Naur Form (ABNF) 148 notation of [RFC5234]. 150 3 Digest Access Authentication Scheme 152 3.1 Overall Operation 154 The Digest scheme is based on a simple challenge-response paradigm. 155 The Digest scheme challenges using a nonce value. A valid response 156 contains a checksum of the username, the password, the given nonce 157 value, the HTTP method, and the requested URI. In this way, the 158 password is never sent in the clear. The username and password must 159 be prearranged in some fashion not addressed by this document. 161 3.2 Representation of Digest Values 163 An optional header allows the server to specify the algorithm used to 164 create the checksum or digest. This documents adds SHA-256 and SHA- 165 512/256 algorithms. To maintain backwards compatibility, the MD5 166 algorithm is still supported but not recommended. 168 The size of the digest depends on the algorithm used. The bits in 169 the digest are converted from the most significant to the least 170 significant bit, four bits at a time to the ASCII representation as 171 follows. Each four bits is represented by its familiar hexadecimal 172 notation from the characters 0123456789abcdef, that is binary 0000 is 173 represented by the character '0', 0001 by '1' and so on up to the 174 representation of 1111 as 'f'. If the MD5 algorithm is used to 175 calculate the digest, then the digest will be represented as 32 176 hexadecimal characters, SHA-256 and SHA-512/256 by 64 hexadecimal 177 characters. 179 3.3 The WWW-Authenticate Response Header 181 If a server receives a request for an access-protected object, and an 182 acceptable Authorization header is not sent, the server responds with 183 a "401 Unauthorized" status code, and a WWW-Authenticate header with 184 Digest scheme as per the framework defined above, and include some or 185 all of the following parameters: 187 realm 188 A string to be displayed to users so they know which username and 189 password to use. This string should contain at least the name of 190 the host performing the authentication and might additionally 191 indicate the collection of users who might have access. An example 192 might be "registered_users@gotham.news.com". (See section 2.2 of 193 [HTTP-P7] for more details). 195 domain 196 A quoted, space-separated list of URIs, as specified in RFC 3986 197 [RFC3986], that define the protection space. If a URI is an 198 abs_path, it is relative to the canonical root URL of the server 199 being accessed. An absolute-URI in this list may refer to a 200 different server than the one being accessed. The client can use 201 this list to determine the set of URIs for which the same 202 authentication information may be sent: any URI that has a URI in 203 this list as a prefix (after both have been made absolute) may be 204 assumed to be in the same protection space. If this parameter is 205 omitted or its value is empty, the client should assume that the 206 protection space consists of all URIs on the responding server. 208 This parameter is not meaningful in Proxy-Authenticate headers, 209 for which the protection space is always the entire proxy; if 210 present it should be ignored. 212 nonce 213 A server-specified data string which should be uniquely generated 214 each time a 401 response is made. It is recommended that this 215 string be base64 or hexadecimal data. Specifically, since the 216 string is passed in the header lines as a quoted string, the 217 double-quote character is not allowed. 219 The contents of the nonce are implementation dependent. The 220 quality of the implementation depends on a good choice. A nonce 221 might, for example, be constructed as the base 64 encoding of 223 time-stamp H(time-stamp ":" ETag ":" private-key) 225 where time-stamp is a server-generated time or other non-repeating 226 value, ETag is the value of the HTTP ETag header associated with 227 the requested entity, and private-key is data known only to the 228 server. With a nonce of this form a server would recalculate the 229 hash portion after receiving the client authentication header and 230 reject the request if it did not match the nonce from that header 231 or if the time-stamp value is not recent enough. In this way the 232 server can limit the time of the nonce's validity. The inclusion 233 of the ETag prevents a replay request for an updated version of 234 the resource. (Note: including the IP address of the client in the 235 nonce would appear to offer the server the ability to limit the 236 reuse of the nonce to the same client that originally got it. 237 However, that would break proxy farms, where requests from a 238 single user often go through different proxies in the farm. Also, 239 IP address spoofing is not that hard.) 241 An implementation might choose not to accept a previously used 242 nonce or a previously used digest, in order to protect against a 243 replay attack. Or, an implementation might choose to use one-time 244 nonces or digests for POST or PUT requests and a time-stamp for 245 GET requests. For more details on the issues involved see section 246 5 of this document. 248 The nonce is opaque to the client. 250 opaque 251 A string of data, specified by the server, which should be 252 returned by the client unchanged in the Authorization header of 253 subsequent requests with URIs in the same protection space. It is 254 recommended that this string be base64 or hexadecimal data. 256 stale 257 A case-insensitive flag, indicating that the previous request from 258 the client was rejected because the nonce value was stale. If 259 stale is TRUE, the client may wish to simply retry the request 260 with a new encrypted response, without reprompting the user for a 261 new username and password. The server should only set stale to 262 TRUE if it receives a request for which the nonce is invalid but 263 with a valid digest for that nonce (indicating that the client 264 knows the correct username/password). If stale is FALSE, or 265 anything other than TRUE, or the stale parameter is not present, 266 the username and/or password are invalid, and new values must be 267 obtained. 269 algorithm 270 A string indicating a pair of algorithms used to produce the 271 digest and a checksum. If this is not present it is assumed to be 272 "MD5". If the algorithm is not understood, the challenge should be 273 ignored (and a different one used, if there is more than one). 275 In this document the string obtained by applying the digest 276 algorithm to the data "data" with secret "secret" will be denoted 277 by KD(secret, data), and the string obtained by applying the 278 checksum algorithm to the data "data" will be denoted H(data). The 279 notation unq(X) means the value of the quoted-string X without the 280 surrounding quotes. 282 For "" and "-sess" 284 H(data) = (data) 286 and 288 KD(secret, data) = H(concat(secret, ":", data)) 290 For example: 292 For the "SHA-256" and "SHA-256-sess" algorithms 294 H(data) = SHA-256(data) 296 i.e., the digest is the SHA-256 of the secret concatenated with 297 a colon concatenated with the data. The "SHA-256-sess" 298 algorithm is intended to allow efficient 3rd party 299 authentication servers; for the difference in usage, see the 300 description in section 3.4.2. 302 qop 303 This parameter MUST be used by all implementations compliant with 304 this version of the Digest scheme. It is a quoted string of one or 305 more tokens indicating the "quality of protection" values 306 supported by the server. The value "auth" indicates 307 authentication; the value "auth-int" indicates authentication with 308 integrity protection; see the descriptions below for calculating 309 the response parameter value for the application of this choice. 310 Unrecognized options MUST be ignored. 312 charset 313 This is an optional parameter that is used by the server to 314 indicate the encoding scheme it supports. 316 userhash 317 This is an optional parameter that is used by the server to 318 indicate that it supports username hashing. Valid value are: 319 "true" or "false". 321 3.4 The Authorization Request Header 323 The client is expected to retry the request, passing an 324 Authorization header line with Digest scheme, which is defined 325 according to the framework above. The values of the opaque and 326 algorithm fields must be those supplied in the WWW-Authenticate 327 response header for the entity being requested. 329 The request includes some or all of the following parameters: 331 response 332 A string of the hex digits computed as defined below, which proves 333 that the user knows a password. 335 username 336 The user's name in the specified realm. 338 uri 339 The URI from request-target of the Request-Line; duplicated here 340 because proxies are allowed to change the Request-Line in transit. 342 qop 343 Indicates what "quality of protection" the client has applied to 344 the message. Its value MUST be one of the alternatives the server 345 indicated it supports in the WWW-Authenticate header. These values 346 affect the computation of the response. Note that this is a single 347 token, not a quoted list of alternatives as in WWW-Authenticate. 348 .in 3 350 cnonce 351 This MUST be specified if a qop parameter is sent (see above), and 352 MUST NOT be specified if the server did not send a qop parameter 353 in the WWW-Authenticate header field. The cnonce value is an 354 opaque quoted string value provided by the client and used by both 355 client and server to avoid chosen plaintext attacks, to provide 356 mutual authentication, and to provide some message integrity 357 protection. See the descriptions below of the calculation of the 358 rspauth and response values. 360 nc 361 The "nc" parameter stands for "nonce count". This MUST be 362 specified if a qop parameter is sent (see above), and MUST NOT be 363 specified if the server did not send a qop parameter in the WWW- 364 Authenticate header field. The nc value is the hexadecimal count 365 of the number of requests (including the current request) that the 366 client has sent with the nonce value in this request. For 367 example, in the first request sent in response to a given nonce 368 value, the client sends "nc=00000001". The purpose of this 369 parameter is to allow the server to detect request replays by 370 maintaining its own copy of this count - if the same nc value is 371 seen twice, then the request is a replay. See the description 372 below of the construction of the response value. 374 userhash 375 This optional parameter is used by the client to indicate that the 376 username has been hashed. Valid value are: "true" or "false". 378 If a parameter or its value is improper, or required parameters are 379 missing, the proper response is 400 Bad Request. If the request- 380 digest is invalid, then a login failure should be logged, since 381 repeated login failures from a single client may indicate an attacker 382 attempting to guess passwords. 384 The definition of response above indicates the encoding for its 385 value. The following definitions show how the value is computed. 387 3.4.1 Response 389 If the "qop" value is "auth" or "auth-int": 391 response = <"> < KD ( H(A1), unq(nonce) 392 ":" nc 393 ":" unq(cnonce) 394 ":" unq(qop) 395 ":" H(A2) 396 ) <"> 398 See below for the definitions for A1 and A2. 400 3.4.2 A1 402 If the "algorithm" parameter's value is "", e.g. "SHA- 403 256", then A1 is: 405 A1 = unq(username) ":" unq(realm) ":" passwd 407 where 409 passwd = < user's password > 411 If the "algorithm" parameter's value is "-sess", e.g. 412 "SHA-256-sess", then A1 is calculated only once - on the first 413 request by the client following receipt of a WWW-Authenticate 414 challenge from the server. It uses the server nonce from that 415 challenge, and the first client nonce value to construct A1 as 416 follows: 418 A1 = H( unq(username) ":" unq(realm) 419 ":" passwd ) 420 ":" unq(nonce) ":" unq(cnonce) 422 This creates a 'session key' for the authentication of subsequent 423 requests and responses which is different for each "authentication 424 session", thus limiting the amount of material hashed with any one 425 key. (Note: see further discussion of the authentication session in 426 section 3.6.) Because the server need only use the hash of the user 427 credentials in order to create the A1 value, this construction could 428 be used in conjunction with a third party authentication service so 429 that the web server would not need the actual password value. The 430 specification of such a protocol is beyond the scope of this 431 specification. 433 3.4.3 A2 435 If the "qop" parameter's value is "auth" or is unspecified, then A2 436 is: 438 A2 = Method ":" request-uri 440 If the "qop" value is "auth-int", then A2 is: 442 A2 = Method ":" request-uri ":" H(entity-body) 444 3.4.4 Username Hashing 446 To protect the transport of the username from the client to the 447 server, the server SHOULD set the "userhash" parameter with the value 448 of "true" in the WWW-Authentication header. 450 If the client supports the "userhash" parameter, and the "userhash" 451 parameter value in the WWW-Authentication header is set to "true", 452 then the client MUST calculate a hash of the username after any other 453 hash calculation and include the "userhash" parameter with the value 454 of "true" in the Authorization Request Header. If the client does not 455 provide the "username" as a hash value or the "userhash" parameter 456 with the value of "true", the server MAY reject the request. 458 The following is the operation that the client will take to hash the 459 username: 461 username = H( unq(username) ":" unq(realm) ) 463 3.4.5 Parameter Values and Quoted-String 465 Note that the value of many of the parameters, such as "username" 466 value, are defined as a "quoted-string". However, the "unq" notation 467 indicates that surrounding quotation marks are removed in forming the 468 string A1. Thus if the Authorization header includes the fields 470 username="Mufasa", realm=myhost@testrealm.com 472 and the user Mufasa has password "Circle Of Life" then H(A1) would be 473 H(Mufasa:myhost@testrealm.com:Circle Of Life) with no quotation marks 474 in the digested string. 476 No white space is allowed in any of the strings to which the digest 477 function H() is applied unless that white space exists in the quoted 478 strings or entity body whose contents make up the string to be 479 digested. For example, the string A1 illustrated above must be 481 Mufasa:myhost@testrealm.com:Circle Of Life 483 with no white space on either side of the colons, but with the white 484 space between the words used in the password value. Likewise, the 485 other strings digested by H() must not have white space on either 486 side of the colons which delimit their fields unless that white space 487 was in the quoted strings or entity body being digested. 489 Also note that if integrity protection is applied (qop=auth-int), the 490 H(entity-body) is the hash of the entity body, not the message body - 491 it is computed before any transfer encoding is applied by the sender 492 and after it has been removed by the recipient. Note that this 493 includes multipart boundaries and embedded headers in each part of 494 any multipart content-type. 496 3.4.6 Various Considerations 498 The "Method" value is the HTTP request method as specified in section 499 3.1.1 of [HTTP-P1]. The "request-target" value is the request-target 500 from the request line as specified in section 3.1.1 of [HTTP-P1]. 501 This may be "*", an "absolute-URI" or an "absolute-path" as specified 502 in section 2.7 of [HTTP-P1], but it MUST agree with the request- 503 target. In particular, it MUST be an "absolute-URI" if the request- 504 target is an "absolute-URI". The "cnonce" value is an optional 505 client-chosen value whose purpose is to foil chosen plaintext 506 attacks. 508 The authenticating server must assure that the resource designated by 509 the "uri" parameter is the same as the resource specified in the 510 Request-Line; if they are not, the server SHOULD return a 400 Bad 511 Request error. (Since this may be a symptom of an attack, server 512 implementers may want to consider logging such errors.) The purpose 513 of duplicating information from the request URL in this field is to 514 deal with the possibility that an intermediate proxy may alter the 515 client's Request-Line. This altered (but presumably semantically 516 equivalent) request would not result in the same digest as that 517 calculated by the client. 519 Implementers should be aware of how authenticated transactions 520 interact with shared caches. The HTTP/1.1 protocol specifies that 521 when a shared cache (see [HTTP-P6]) has received a request containing 522 an Authorization header and a response from relaying that request, it 523 MUST NOT return that response as a reply to any other request, unless 524 one of two Cache-Control (see section 3.2 of [HTTP-P6]) directive was 525 present in the response. If the original response included the "must- 526 revalidate" Cache-Control directive, the cache MAY use the entity of 527 that response in replying to a subsequent request, but MUST first 528 revalidate it with the origin server, using the request headers from 529 the new request to allow the origin server to authenticate the new 530 request. Alternatively, if the original response included the 531 "public" Cache-Control directive, the response entity MAY be returned 532 in reply to any subsequent request. 534 3.5 The Authentication-Info Header 536 The Authentication-Info header is used by the server to communicate 537 some information regarding the successful authentication in the 538 response. 540 Authentication-Info = auth-info 542 auth-info = *auth-param 544 The request includes some or all of the following parameters: 546 nextnonce 548 The value of the nextnonce parameter is the nonce the server 549 wishes the client to use for a future authentication response. 550 The server may send the Authentication-Info header with a 551 nextnonce field as a means of implementing one-time or otherwise 552 changing nonces. If the nextnonce field is present the client 553 SHOULD use it when constructing the Authorization header for its 554 next request. Failure of the client to do so may result in a 555 request to re-authenticate from the server with the "stale=TRUE". 557 Server implementations should carefully consider the 558 performance implications of the use of this mechanism; 559 pipelined requests will not be possible if every response 560 includes a nextnonce parameter that must be used on the next 561 request received by the server. Consideration should be given 562 to the performance vs. security tradeoffs of allowing an old 563 nonce value to be used for a limited time to permit request 564 pipelining. Use of the "nc" parameter can retain most of the 565 security advantages of a new server nonce without the 566 deleterious affects on pipelining. 568 qop 569 Indicates the "quality of protection" options applied to the 570 response by the server. The value "auth" indicates 571 authentication; the value "auth-int" indicates authentication with 572 integrity protection. The server SHOULD use the same value for the 573 qop parameter in the response as was sent by the client in the 574 corresponding request. 576 rspauth 578 The optional response digest in the "rspauth" parameter supports 579 mutual authentication -- the server proves that it knows the 580 user's secret, and with qop=auth-int also provides limited 581 integrity protection of the response. The "rspauth" value is 582 calculated as for the response in the Authorization header, except 583 that if "qop=auth" or is not specified in the Authorization header 584 for the request, A2 is 586 A2 = ":" request-uri 588 and if "qop=auth-int", then A2 is 590 A2 = ":" request-uri ":" H(entity-body) 592 cnonce and nc 594 The "cnonce" value and "nc" value MUST be the ones for the client 595 request to which this message is the response. The "rspauth", 596 "cnonce", and "nc" parameters MUST be present if "qop=auth" or 597 "qop=auth-int" is specified. 599 The Authentication-Info header is allowed in the trailer of an HTTP 600 message transferred via chunked transfer-coding. 602 3.6 Digest Operation 604 Upon receiving the Authorization header, the server may check its 605 validity by looking up the password that corresponds to the submitted 606 username. Then, the server must perform the same digest operation 607 (e.g., MD5) performed by the client, and compare the result to the 608 given response value. 610 Note that the HTTP server does not actually need to know the user's 611 cleartext password. As long as H(A1) is available to the server, the 612 validity of an Authorization header may be verified. 614 The client response to a WWW-Authenticate challenge for a protection 615 space starts an authentication session with that protection space. 616 The authentication session lasts until the client receives another 617 WWW-Authenticate challenge from any server in the protection space. A 618 client should remember the username, password, nonce, nonce count and 619 opaque values associated with an authentication session to use to 620 construct the Authorization header in future requests within that 621 protection space. The Authorization header may be included 622 preemptively; doing so improves server efficiency and avoids extra 623 round trips for authentication challenges. The server may choose to 624 accept the old Authorization header information, even though the 625 nonce value included might not be fresh. Alternatively, the server 626 may return a 401 response with a new nonce value, causing the client 627 to retry the request; by specifying stale=TRUE with this response, 628 the server tells the client to retry with the new nonce, but without 629 prompting for a new username and password. 631 Because the client is required to return the value of the opaque 632 parameter given to it by the server for the duration of a session, 633 the opaque data may be used to transport authentication session state 634 information. (Note that any such use can also be accomplished more 635 easily and safely by including the state in the nonce.) For example, 636 a server could be responsible for authenticating content that 637 actually sits on another server. It would achieve this by having the 638 first 401 response include a domain parameter whose value includes a 639 URI on the second server, and an opaque parameter whose value 640 contains the state information. The client will retry the request, at 641 which time the server might respond with a 301/302 redirection, 642 pointing to the URI on the second server. The client will follow the 643 redirection, and pass an Authorization header , including the 644 data. 646 As with the basic scheme, proxies must be completely transparent in 647 the Digest access authentication scheme. That is, they must forward 648 the WWW-Authenticate, Authentication-Info and Authorization headers 649 untouched. If a proxy wants to authenticate a client before a request 650 is forwarded to the server, it can be done using the Proxy- 651 Authenticate and Proxy-Authorization headers described in section 3.6 652 below. 654 3.7 Security Protocol Negotiation 656 It is useful for a server to be able to know which security schemes a 657 client is capable of handling. 659 It is possible that a server may want to require Digest as its 660 authentication method, even if the server does not know that the 661 client supports it. A client is encouraged to fail gracefully if the 662 server specifies only authentication schemes it cannot handle. 664 When a server receives a request to access a resource, the server 665 might challenge the client by responding with "401 Unauthorized" 666 status code, and include one or more WWW-Authenticate headers. If the 667 server challenges with multiple Digest headers, then each one of 668 these headers MUST use a different digest algorithm. The server MUST 669 add these Digest headers to the response in order of preference, 670 starting with the most preferred header, followed by the less 671 preferred headers. 673 This specification defines the following preference list, starting 674 with the most preferred algorithm: 676 * SHA2-256 677 * SHA2-512/256 678 * MD5 (for backward compatibility). 680 When the client receives the response it SHOULD use the topmost 681 header that it supports, unless a local policy dictates otherwise. 682 The client should ignore any challenge it does not understand. 684 3.8 Proxy-Authenticate and Proxy-Authorization 686 The digest authentication scheme may also be used for authenticating 687 users to proxies, proxies to proxies, or proxies to origin servers by 688 use of the Proxy-Authenticate and Proxy-Authorization headers. These 689 headers are instances of the Proxy-Authenticate and Proxy- 690 Authorization headers specified in sections 4.2 and 4.3 of the 691 HTTP/1.1 specification [HTTP-P7] and their behavior is subject to 692 restrictions described there. The transactions for proxy 693 authentication are very similar to those already described. Upon 694 receiving a request which requires authentication, the proxy/server 695 must issue the "407 Proxy Authentication Required" response with a 696 "Proxy-Authenticate" header. The digest-challenge used in the Proxy- 697 Authenticate header is the same as that for the WWW- Authenticate 698 header as defined above in section 3.2.1. 700 The client/proxy must then re-issue the request with a Proxy- 701 Authorization header, with parameters as specified for the 702 Authorization header in section 3.4 above. 704 On subsequent responses, the server sends Proxy-Authenticate-Info 705 with parameters the same as those for the Authentication-Info header 706 field. 708 Note that in principle a client could be asked to authenticate itself 709 to both a proxy and an end-server, but never in the same response. 711 3.9 Examples 713 3.9.1 Example with SHA-256 and MD5 715 The following example assumes that an access protected document is 716 being requested from the server via a GET request. The URI of the 717 document is http://www.nowhere.org/dir/index.html". Both client and 718 server know that the username for this document is "Mufasa" and the 719 password is "Circle of Life" ( with one space between each of the 720 three words). 722 The first time the client requests the document, no Authorization 723 header is sent, so the server responds with: 725 HTTP/1.1 401 Unauthorized 726 WWW-Authenticate: Digest 727 realm = "testrealm@host.com", 728 qop="auth, auth-int", 729 algorithm="SHA-256", 730 nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", 731 opaque="5ccc069c403ebaf9f0171e9517f40e41" 732 WWW-Authenticate: Digest 733 realm="testrealm@host.com", 734 qop="auth, auth-int", 735 algorithm="MD5", 736 nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", 737 opaque="5ccc069c403ebaf9f0171e9517f40ef41" 739 The client may prompt the user for their username and password, after 740 which it will respond with a new request, including the following 741 Authorization header if the client chooses MD5 digest: 743 Authorization:Digest username="Mufasa", 744 realm="testrealm@host.com", 745 nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", 746 uri="/dir/index.html", 747 qop="auth", 748 algorithm="MD5", 749 nc=00000001, 750 cnonce="0a4f113b", 751 response="6629fae49393a05397450978507c4ef1", 752 opaque="5ccc069c403ebaf9f0171e9517f40e41" 754 If the client chooses to use the SHA-256 algorithm for calculating 755 the response, the client responds with a new request including the 756 following Authorization header: 758 Authorization:Digest username="Mufasa", 759 realm="testrealm@host.com", 760 nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", 761 uri="/dir/index.html", 762 qop="auth", 763 algorithm="SHA-256", 764 nc=00000001, 765 cnonce="0a4f113b", 766 response="5abdd07184ba512a22c53f41470e5eea7dcaa3a93 767 a59b630c13dfe0a5dc6e38b", 768 opaque="5ccc069c403ebaf9f0171e9517f40e41" 770 3.9.2 Example with SHA-512-256, Charset, and Userhash 771 The following example assumes that an access protected document is 772 being requested from the server via a GET request. The URI for the 773 request is "http://api.example.org/doe.json". Both client and server 774 know the userhash of the username, support the UTF-8 charset, and use 775 the SHA-512-256 algorithm. The username for the request is "Jason 776 Doe" and the password is "Secret, or not?". 778 The first time the client requests the document, no Authorization 779 header is sent, so the server responds with: 781 HTTP/2.0 401 Unauthorized 782 WWW-Authenticate: Digest 783 realm="api@example.org", 784 qop=auth, 785 algorithm=SHA-512-256, 786 nonce="e145a96d70d40739596e60c6340f13be03290bd73c676d 787 3f25c01271af522eb2", 788 opaque="192cbcf2a2576846522c1a367c1dfdf0359a87719c5cc1 789 839e4f3d2ffeb82517", 790 charset=UTF-8, 791 userhash=true 793 The client may prompt the user for the required credentials and send 794 a new request with following Authorization header: 796 Authorization: Digest 797 username="298bc3decec198ec5e7ecc1d69f059ca33044dd15baf45 798 a1f87bbd7adb3784fd", 799 realm="api@example.org", 800 uri="/doe.json", 801 algorithm=SHA-512-256, 802 nonce="e145a96d70d40739596e60c6340f13be03290bd73c676d 803 3f25c01271af522eb2", 804 nc=00000001, 805 cnonce="cde966df34a49d5d842a263604159141c81db8d468e1bf 806 657230429424fc337a", 807 qop=auth, 808 response="ec180fc03b7a0dcd43c414f66f2335399bbe5f4d4ad469 809 f8233106ba453213c8", 810 opaque="192cbcf2a2576846522c1a367c1dfdf0359a87719c5cc1 811 839e4f3d2ffeb82517", 812 userhash=true 814 If the client can not provide a hashed username for any reason, the 815 client may try a request with this Authorization header: 817 Authorization: Digest 818 username="Jason Doe", 819 realm="api@example.org", 820 uri="/doe.json", 821 algorithm=SHA-512-256, 822 nonce="e145a96d70d40739596e60c6340f13be03290bd73c676d 823 3f25c01271af522eb2", 824 nc=00000001, 825 cnonce="cde966df34a49d5d842a263604159141c81db8d468e1bf 826 657230429424fc337a", 827 qop=auth, 828 response="ec180fc03b7a0dcd43c414f66f2335399bbe5f4d4ad469 829 f8233106ba453213c8", 830 opaque="192cbcf2a2576846522c1a367c1dfdf0359a87719c5cc1 831 839e4f3d2ffeb82517", 832 userhash=false 834 4 Internationalization 836 In challenges, servers SHOULD use the "charset" authentication 837 parameter (case-insensitive) to express the character encoding they 838 expect the user agent to use when generating A1 (see section 3.4.2) 839 and username hashing (see section 3.4.4). 841 The only allowed value is "UTF-8", to be matched case-insensitively 842 (see [RFC2978], Section 2.3). It indicates that the server expects 843 user name and password to be converted to Unicode Normalization Form 844 C ("NFC", see Section 3 of [RFC5198]) and to be encoded into octets 845 using the UTF-8 character encoding scheme ([RFC3629]). 847 If the user agent does not support the encoding indicated by the 848 server, it MUST fail the request. 850 5 Security Considerations 852 5.1 Limitations 854 HTTP Digest authentication, when used with human-memorable passwords, 855 is vulnerable to dictionary attacks. Such attacks are much easier 856 than cryptographic attacks on any widely used algorithm, including 857 those that are no longer considered secure. In other words, algorithm 858 agility does not make this usage any more secure. 860 As a result, Digest authentication SHOULD be used only with passwords 861 that have a reasonable amount of entropy, e.g. 128-bit or more. Such 862 passwords typically cannot be memorized by humans but can be used for 863 automated web services. 865 Digest authentication SHOULD be used over a secure channel like HTTPS 866 [RFC2818]. 868 5.2 Authentication of Clients using Digest Authentication 870 Digest Authentication does not provide a strong authentication 871 mechanism, when compared to public key based mechanisms, for example. 873 However, it is significantly stronger than (e.g.) CRAM-MD5, which has 874 been proposed for use with LDAP [RFC4513], POP and IMAP (see 875 [RFC2195]). It is intended to replace the much weaker and even more 876 dangerous Basic mechanism. 878 Digest Authentication offers no confidentiality protection beyond 879 protecting the actual username and password. All of the rest of the 880 request and response are available to an eavesdropper. 882 Digest Authentication offers only limited integrity protection for 883 the messages in either direction. If qop=auth-int mechanism is used, 884 those parts of the message used in the calculation of the WWW- 885 Authenticate and Authorization header field response parameter values 886 (see section 3.2 above) are protected. Most header fields and their 887 values could be modified as a part of a man-in-the-middle attack. 889 Many needs for secure HTTP transactions cannot be met by Digest 890 Authentication. For those needs TLS or SHTTP are more appropriate 891 protocols. In particular Digest authentication cannot be used for any 892 transaction requiring confidentiality protection. Nevertheless many 893 functions remain for which Digest authentication is both useful and 894 appropriate. 896 5.3 Limited Use Nonce Values 898 The Digest scheme uses a server-specified nonce to seed the 899 generation of the response value (as specified in section 3.4.1 900 above). As shown in the example nonce in section 3.2.1, the server 901 is free to construct the nonce such that it may only be used from a 902 particular client, for a particular resource, for a limited period of 903 time or number of uses, or any other restrictions. Doing so 904 strengthens the protection provided against, for example, replay 905 attacks (see 4.5). However, it should be noted that the method 906 chosen for generating and checking the nonce also has performance and 907 resource implications. For example, a server may choose to allow 908 each nonce value to be used only once by maintaining a record of 909 whether or not each recently issued nonce has been returned and 910 sending a next-nonce parameter in the Authentication-Info header 911 field of every response. This protects against even an immediate 912 replay attack, but has a high cost checking nonce values, and perhaps 913 more important will cause authentication failures for any pipelined 914 requests (presumably returning a stale nonce indication). Similarly, 915 incorporating a request-specific element such as the Etag value for a 916 resource limits the use of the nonce to that version of the resource 917 and also defeats pipelining. Thus it may be useful to do so for 918 methods with side effects but have unacceptable performance for those 919 that do not. 921 5.4 Replay Attacks 923 A replay attack against Digest authentication would usually be 924 pointless for a simple GET request since an eavesdropper would 925 already have seen the only document he could obtain with a replay. 926 This is because the URI of the requested document is digested in the 927 client request and the server will only deliver that document. By 928 contrast under Basic Authentication once the eavesdropper has the 929 user's password, any document protected by that password is open to 930 him. 932 Thus, for some purposes, it is necessary to protect against replay 933 attacks. A good Digest implementation can do this in various ways. 934 The server created "nonce" value is implementation dependent, but if 935 it contains a digest of the client IP, a time-stamp, the resource 936 ETag, and a private server key (as recommended above) then a replay 937 attack is not simple. An attacker must convince the server that the 938 request is coming from a false IP address and must cause the server 939 to deliver the document to an IP address different from the address 940 to which it believes it is sending the document. An attack can only 941 succeed in the period before the time-stamp expires. Digesting the 942 client IP and time-stamp in the nonce permits an implementation which 943 does not maintain state between transactions. 945 For applications where no possibility of replay attack can be 946 tolerated the server can use one-time nonce values which will not be 947 honored for a second use. This requires the overhead of the server 949 remembering which nonce values have been used until the nonce time- 950 stamp (and hence the digest built with it) has expired, but it 951 effectively protects against replay attacks. 953 An implementation must give special attention to the possibility of 954 replay attacks with POST and PUT requests. Unless the server employs 955 one-time or otherwise limited-use nonces and/or insists on the use of 956 the integrity protection of qop=auth-int, an attacker could replay 957 valid credentials from a successful request with counterfeit form 958 data or other message body. Even with the use of integrity protection 959 most metadata in header fields is not protected. Proper nonce 960 generation and checking provides some protection against replay of 961 previously used valid credentials, but see 4.8. 963 5.5 Weakness Created by Multiple Authentication Schemes 965 An HTTP/1.1 server may return multiple challenges with a 401 966 (Authenticate) response, and each challenge may use a different auth- 967 scheme. A user agent MUST choose to use the strongest auth- scheme it 968 understands and request credentials from the user based upon that 969 challenge. 971 Note that many browsers will only recognize Basic and will require 972 that it be the first auth-scheme presented. Servers should only 973 include Basic if it is minimally acceptable. 975 When the server offers choices of authentication schemes using the 976 WWW-Authenticate header, the strength of the resulting authentication 977 is only as good as that of the of the weakest of the authentication 978 schemes. See section 5.7 below for discussion of particular attack 979 scenarios that exploit multiple authentication schemes. 981 5.6 Online dictionary attacks 983 If the attacker can eavesdrop, then it can test any overheard 984 nonce/response pairs against a list of common words. Such a list is 985 usually much smaller than the total number of possible passwords. The 986 cost of computing the response for each password on the list is paid 987 once for each challenge. 989 The server can mitigate this attack by not allowing users to select 990 passwords that are in a dictionary. 992 5.7 Man in the Middle 994 Both Basic and Digest authentication are vulnerable to "man in the 995 middle" (MITM) attacks, for example, from a hostile or compromised 996 proxy. Clearly, this would present all the problems of eavesdropping. 997 But it also offers some additional opportunities to the attacker. 999 A possible man-in-the-middle attack would be to add a weak 1000 authentication scheme to the set of choices, hoping that the client 1001 will use one that exposes the user's credentials (e.g. password). For 1002 this reason, the client should always use the strongest scheme that 1003 it understands from the choices offered. 1005 An even better MITM attack would be to remove all offered choices, 1006 replacing them with a challenge that requests only Basic 1007 authentication, then uses the cleartext credentials from the Basic 1008 authentication to authenticate to the origin server using the 1009 stronger scheme it requested. A particularly insidious way to mount 1010 such a MITM attack would be to offer a "free" proxy caching service 1011 to gullible users. 1013 User agents should consider measures such as presenting a visual 1014 indication at the time of the credentials request of what 1015 authentication scheme is to be used, or remembering the strongest 1016 authentication scheme ever requested by a server and produce a 1017 warning message before using a weaker one. It might also be a good 1018 idea for the user agent to be configured to demand Digest 1019 authentication in general, or from specific sites. 1021 Or, a hostile proxy might spoof the client into making a request the 1022 attacker wanted rather than one the client wanted. Of course, this is 1023 still much harder than a comparable attack against Basic 1024 Authentication. 1026 5.8 Chosen plaintext attacks 1028 With Digest authentication, a MITM or a malicious server can 1029 arbitrarily choose the nonce that the client will use to compute the 1030 response. This is called a "chosen plaintext" attack. The ability to 1031 choose the nonce is known to make cryptanalysis much easier. 1033 However, no way to analyze the MD5 one-way function used by Digest 1034 using chosen plaintext is currently known. 1036 The countermeasure against this attack is for clients to be 1037 configured to require the use of the optional "cnonce" parameter; 1038 this allows the client to vary the input to the hash in a way not 1039 chosen by the attacker. 1041 5.9 Precomputed dictionary attacks 1043 With Digest authentication, if the attacker can execute a chosen 1044 plaintext attack, the attacker can precompute the response for many 1045 common words to a nonce of its choice, and store a dictionary of 1046 (response, password) pairs. Such precomputation can often be done in 1047 parallel on many machines. It can then use the chosen plaintext 1048 attack to acquire a response corresponding to that challenge, and 1049 just look up the password in the dictionary. Even if most passwords 1050 are not in the dictionary, some might be. Since the attacker gets to 1051 pick the challenge, the cost of computing the response for each 1052 password on the list can be amortized over finding many passwords. A 1053 dictionary with 100 million password/response pairs would take about 1054 3.2 gigabytes of disk storage. 1056 The countermeasure against this attack is to for clients to be 1057 configured to require the use of the optional "cnonce" parameter. 1059 5.10 Batch brute force attacks 1061 With Digest authentication, a MITM can execute a chosen plaintext 1062 attack, and can gather responses from many users to the same nonce. 1063 It can then find all the passwords within any subset of password 1064 space that would generate one of the nonce/response pairs in a single 1065 pass over that space. It also reduces the time to find the first 1066 password by a factor equal to the number of nonce/response pairs 1067 gathered. This search of the password space can often be done in 1068 parallel on many machines, and even a single machine can search large 1069 subsets of the password space very quickly -- reports exist of 1070 searching all passwords with six or fewer letters in a few hours. 1072 The countermeasure against this attack is to for clients to be 1073 configured to require the use of the optional "cnonce" parameter. 1075 5.11 Spoofing by Counterfeit Servers 1077 Basic Authentication is vulnerable to spoofing by counterfeit 1078 servers. If a user can be led to believe that she is connecting to a 1079 host containing information protected by a password she knows, when 1080 in fact she is connecting to a hostile server, then the hostile 1081 server can request a password, store it away for later use, and feign 1082 an error. This type of attack is more difficult with Digest 1083 Authentication -- but the client must know to demand that Digest 1084 authentication be used, perhaps using some of the techniques 1085 described above to counter "man-in-the-middle" attacks. Again, the 1086 user can be helped in detecting this attack by a visual indication of 1087 the authentication mechanism in use with appropriate guidance in 1088 interpreting the implications of each scheme. 1090 5.12 Storing passwords 1091 Digest authentication requires that the authenticating agent (usually 1092 the server) store some data derived from the user's name and password 1093 in a "password file" associated with a given realm. Normally this 1094 might contain pairs consisting of username and H(A1), where H(A1) is 1095 the digested value of the username, realm, and password as described 1096 above. 1098 The security implications of this are that if this password file is 1099 compromised, then an attacker gains immediate access to documents on 1100 the server using this realm. Unlike, say a standard UNIX password 1101 file, this information need not be decrypted in order to access 1102 documents in the server realm associated with this file. On the other 1103 hand, decryption, or more likely a brute force attack, would be 1104 necessary to obtain the user's password. This is the reason that the 1105 realm is part of the digested data stored in the password file. It 1106 means that if one Digest authentication password file is compromised, 1107 it does not automatically compromise others with the same username 1108 and password (though it does expose them to brute force attack). 1110 There are two important security consequences of this. First the 1111 password file must be protected as if it contained unencrypted 1112 passwords, because for the purpose of accessing documents in its 1113 realm, it effectively does. 1115 A second consequence of this is that the realm string should be 1116 unique among all realms which any single user is likely to use. In 1117 particular a realm string should include the name of the host doing 1118 the authentication. The inability of the client to authenticate the 1119 server is a weakness of Digest Authentication. 1121 5.13 Summary 1123 By modern cryptographic standards Digest Authentication is weak. But 1124 for a large range of purposes it is valuable as a replacement for 1125 Basic Authentication. It remedies some, but not all, weaknesses of 1126 Basic Authentication. Its strength may vary depending on the 1127 implementation. In particular the structure of the nonce (which is 1128 dependent on the server implementation) may affect the ease of 1129 mounting a replay attack. A range of server options is appropriate 1130 since, for example, some implementations may be willing to accept the 1131 server overhead of one-time nonces or digests to eliminate the 1132 possibility of replay. Others may satisfied with a nonce like the one 1133 recommended above restricted to a single IP address and a single ETag 1134 or with a limited lifetime. 1136 The bottom line is that *any* compliant implementation will be 1137 relatively weak by cryptographic standards, but *any* compliant 1138 implementation will be far superior to Basic Authentication. 1140 6 IANA Considerations 1142 6.1 HTTP Digest Hash Algorithms Registry 1144 This specification creates a new IANA registry named "HTTP Digest 1145 Hash Algorithms". When registering a new hash algorithm, the 1146 following information MUST be provided: 1148 o Hash Algorithm 1149 The textual name of the hash algorithm. 1151 o Digest Size 1152 The size of the algorithm's output in hexadecimal characters. 1154 o Reference 1155 A reference to the specification that describes the new algorithm. 1157 The update policy for this registry shall be Specification Required. 1159 The initial registry will contain the following entries: 1161 Hash Algorithm Digest Size Reference 1162 -------------- ----------- --------- 1163 "MD5" 32 RFC XXXX 1164 "SHA-512-256" 64 RFC XXXX 1165 "SHA-256" 64 RFC XXXX 1167 Each one of the algorithms defined in the registry might have a -sess 1168 variant, e.g. MD5-sess, SHA-256-sess, etc. 1170 6.2 Digest Scheme Registration 1172 This specification registers the Digest scheme with the 1173 Authentication Scheme Registry. 1175 Authentication Scheme Name: Digest 1177 Pointer to specification text: RFCXXX 1179 6.3 Authentication-Info Header Registration 1181 This specification registers the Authentication-Info Header with the 1182 Message Header Field Registry. 1184 Header Field Name: Authentication-Info 1186 Protocol: http 1188 Status: standard 1190 Reference: RFCXXXX, Section 3.5 1192 7 Acknowledgments 1194 The authors of this document would like to thank the authors of 1195 RFC2617, as this document heavily borrows text from their document to 1196 provide a complete description of the digest scheme and its 1197 operations. 1199 The authors would like to thank Stephen Farrell, Yoav Nir, Phillip 1200 Hallam-Baker, Manu Sporny, Paul Hoffman, Julian Reschke, Yaron 1201 Sheffer, Sean Turner, Geoff Baskwill, Eric Cooper, Bjoern Hoehrmann, 1202 Martin Durst, Peter Saint-Andre, Michael Sweet, Daniel Stenberg, 1203 Brett Tate, Paul Leach, and Ilari Liusvaara for their careful review 1204 and comments. 1206 The authors would like to thank Jonathan Stoke, Nico Williams, Harry 1207 Halpin, and Phil Hunt for their comments on the mailing list when 1208 discussing various aspects of this document. 1210 The authors would like to thank Paul Kyzivat and Dale Worley for 1211 their careful review and feedback on some aspects of this document. 1213 8 References 1215 8.1 Normative References 1217 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1218 Requirement Levels", BCP 14, RFC 2119, March 1997. 1220 [RFC2978] Freed, N. and J. Postel, "IANA Charset Registration 1221 Procedures", BCP 19, RFC 2978, October 2000. 1223 [RFC3629] Yergeau, F., "UTF-8, a transformation format of ISO 1224 10646", STD 63, RFC 3629, November 2003. 1226 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 1227 Resource Identifier (URI): Generic Syntax", STD 66, 1228 RFC 3986, January 2005. 1230 [RFC4513] Harrison, R., Ed., "Lightweight Directory Access Protocol 1231 (LDAP): Authentication Methods and Security Mechanisms", 1232 RFC 4513, June 2006. 1234 [RFC5198] Klensin, J. and M. Padlipsky, "Unicode Format for Network 1235 Interchange", RFC 5198, March 2008. 1237 [RFC5234] Crocker, D., Ed., and P. Overell, "Augmented BNF for 1238 Syntax Specifications: ABNF", STD 68, RFC 5234, January 1239 2008. 1241 [HTTP-P1] Fielding, R., Reschke, J., "Hypertext Transfer Protocol 1242 (HTTP/1.1): Message Syntax and Routing", draft-ietf- 1243 httpbis-p1-messaging (Work in Progress), November 2013. 1245 [HTTP-P6] Fielding, R., Nottingham, M., Reschke, J., "Hypertext 1246 Transfer Protocol (HTTP/1.1): Caching", draft-ietf- 1247 httpbis-p6-cache (Work in Progress), November 2013. 1249 [HTTP-P7] Fielding, R., Reschke, J., "Hypertext Transfer Protocol 1250 (HTTP/1.1): Authentication", draft-ietf-httpbis-p7-auth 1251 (Work in Progress), November 2013. 1253 [BASIC] Reschke, J., "The 'Basic' HTTP Authentication Scheme", 1254 draft-ietf-httpauth-basicauth-enc (Work in Progress), 1255 September 2013. 1257 8.2 Informative References 1259 [RFC2195] Klensin, J., Catoe, R., and P. Krumviede, "IMAP/POP 1260 AUTHorize Extension for Simple Challenge/Response", 1261 RFC 2195, September 1997. 1263 [RFC2818] Rescorla, E., "HTTP Over TLS", RFC 2818, May 2000. 1265 Authors' Addresses 1267 Rifaat Shekh-Yusef (Editor) 1268 Avaya 1269 250 Sydney Street 1270 Belleville, Ontario 1271 Canada 1273 Phone: +1-613-967-5267 1274 Email: rifaat.ietf@gmail.com 1276 David Ahrens 1277 Avaya 1278 California 1279 USA 1281 EMail: ahrensdc@gmail.com 1283 Sophie Bremer 1284 Netzkonform 1285 Germany 1287 Email: sophie.bremer@netzkonform.de