idnits 2.17.1 draft-ietf-radext-nai-07.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (23 September 2014) is 3496 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Obsolete informational reference (is this intentional?): RFC 4282 (Obsoleted by RFC 7542) -- Obsolete informational reference (is this intentional?): RFC 5335 (Obsoleted by RFC 6532) Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 RADEXT Working Group DeKok, Alan 3 INTERNET-DRAFT FreeRADIUS 4 Obsoletes: 4282 5 Category: Standards Track 6 7 23 September 2014 9 The Network Access Identifier 10 draft-ietf-radext-nai-07 12 Abstract 14 In order to provide inter-domain authentication services, it is 15 necessary to have a standardized method that domains can use to 16 identify each other's users. This document defines the syntax for 17 the Network Access Identifier (NAI), the user identity submitted by 18 the client prior to accessing network resources. This document is a 19 revised version of RFC 4282, which addresses issues with 20 international character sets, as well as a number of other 21 corrections to the previous document. 23 Status of this Memo 25 This Internet-Draft is submitted to IETF in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF), its areas, and its working groups. Note that 30 other groups may also distribute working documents as Internet- 31 Drafts. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 The list of current Internet-Drafts can be accessed at 39 http://www.ietf.org/ietf/1id-abstracts.txt. 41 The list of Internet-Draft Shadow Directories can be accessed at 42 http://www.ietf.org/shadow.html. 44 This Internet-Draft will expire on June 23, 2015. 46 Copyright Notice 48 Copyright (c) 2014 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 This document may contain material from IETF Documents or IETF 62 Contributions published or made publicly available before November 63 10, 2008. The person(s) controlling the copyright in some of this 64 material may not have granted the IETF Trust the right to allow 65 modifications of such material outside the IETF Standards Process. 66 Without obtaining an adequate license from the person(s) controlling 67 the copyright in such materials, this document may not be modified 68 outside the IETF Standards Process, and derivative works of it may 69 not be created outside the IETF Standards Process, except to format 70 it for publication as an RFC or to translate it into languages other 71 than English. 73 Table of Contents 75 Appendix A - Changes from RFC4282 ............................ 3 76 1. Introduction ............................................. 4 77 1.1. Terminology ......................................... 5 78 1.2. Requirements Language ............................... 6 79 1.3. Purpose ............................................. 7 80 1.4. Motivation .......................................... 7 81 2. NAI Definition ........................................... 8 82 2.1. UTF-8 Syntax and Normalization ...................... 8 83 2.2. Formal Syntax ....................................... 9 84 2.3. NAI Length Considerations ........................... 10 85 2.4. Support for Username Privacy ........................ 11 86 2.5. International Character Sets ........................ 11 87 2.6. The Normalization Process ........................... 12 88 2.6.1. Issues with the Normalization Process .......... 13 89 2.7. Use in Other Protocols .............................. 14 90 2.8. Using the NAI format for other identifiers .......... 15 91 3. Routing inside of AAA Systems ............................ 16 92 3.1. Compatibility with Email Usernames .................. 17 93 3.2. Compatibility with DNS .............................. 17 94 3.3. Realm Construction .................................. 18 95 3.3.1. Historical Practices ........................... 18 96 3.4. Examples ............................................ 19 97 4. Security Considerations .................................. 20 98 5. Administration of Names .................................. 21 99 6. IANA Considerations ...................................... 21 100 7. References ............................................... 21 101 7.1. Normative References ................................ 21 102 7.2. Informative References .............................. 22 103 Appendix A - Changes from RFC4282 ............................ 24 104 1. Introduction 106 Considerable interest exists for a set of features that fit within 107 the general category of inter-domain authentiction, or "roaming 108 capability" for network access, including dialup Internet users, 109 Virtual Private Network (VPN) usage, wireless LAN authentication, and 110 other applications. Interested parties have included the following: 112 * Regional Internet Service Providers (ISPs) operating within a 113 particular state or province, looking to combine their efforts 114 with those of other regional providers to offer dialup service 115 over a wider area. 117 * National ISPs wishing to combine their operations with those of 118 one or more ISPs in another nation to offer more comprehensive 119 dialup service in a group of countries or on a continent. 121 * Wireless LAN hotspots providing service to one or more ISPs. 123 * Businesses desiring to offer their employees a comprehensive 124 package of dialup services on a global basis. Those services may 125 include Internet access as well as secure access to corporate 126 intranets via a VPN, enabled by tunneling protocols such as the 127 Point-to-Point Tunneling Protocol (PPTP) [RFC2637], the Layer 2 128 Forwarding (L2F) protocol [RFC2341], the Layer 2 Tunneling 129 Protocol (L2TP) [RFC2661], and the IPsec tunnel mode [RFC4301]. 131 * Other protocols which are interested in leveraging the users 132 credentials in order to take advantage of an existing 133 authentication framework. 135 In order to enhance the interoperability of these services, it is 136 necessary to have a standardized method for identifying users. This 137 document defines syntax for the Network Access Identifier (NAI). 138 Examples of implementations that use the NAI, and descriptions of its 139 semantics, can be found in [RFC2194]. 141 When the NAI was defined for network access, it had the side effect 142 of defining an identifier which could be used in non-AAA systems. 143 Some systems defined identifiers which were compatible with the NAI, 144 and deployments used the NAI. This process simplified the management 145 of credentials, by re-using the same credential in multiple 146 situations. We suggest that this re-use is good practice. The 147 alternative is to have protocol-specific identifiers, which increases 148 cost to both user and administrator. 150 The goal of this document is to define the format of an identifier 151 which can be used in many protocols. A protocol may transport an 152 encoded version of the NAI (e.g. '.' as %2E). However, the 153 definition of the NAI is protocol independent. We hope to encourage 154 the wide-spread adoption of the NAI as an identifier. This adoption 155 will decrease work required to leverage identification and 156 authentication in other protocols. It will also decrease the 157 complexity of systems for end users and administrators. 159 We note that this document only suggest that the NAI be used, but 160 does not require such use. Many protocols already define their own 161 identifier formats. Some of these are incompatible with the NAI, 162 while others allow the NAI in addition to non-NAI identifiers. The 163 definition of the NAI in this document has no requirements on 164 protocol specifications, implementations, or deployments. 166 However, we suggest that using one standard identifier format is 167 preferable to using multiple incompatible identifier formats. Where 168 identifiers need to be used in new protocols and/or specifications, 169 it is RECOMMENDED that the format of the NAI be used. That is, the 170 interpretation of the identifier is context-specific, while the 171 format of the identifier remains the same. These issues are 172 discussed in more detail in Section 2.8, below. 174 This document is a revised version of [RFC4282], which originally 175 defined internationalized NAIs. Differences and enhancements 176 compared to that document are listed in Appendix A. 178 1.1. Terminology 180 This document frequently uses the following terms: 182 "Local" or "localized" text 184 Text which is either in non-UTF-8, or in non-normalized form. The 185 character set, encoding, and locale are (in general) unknown to 186 Authentication, Authorization, and Accounting (AAA) network 187 protocols. The client which "knows" the locale may have a 188 different concept of this text than other AAA entities, which do 189 not know the same locale. 191 Network Access Identifier 193 The Network Access Identifier (NAI) is the user identity submitted 194 by the client during network access authentication. The purpose 195 of the NAI is to identify the user as well as to assist in the 196 routing of the authentication request. Please note that the NAI 197 may not necessarily be the same as the user's email address or the 198 user identity submitted in an application layer authentication. 200 Network Access Server 202 The Network Access Server (NAS) is the device that clients connect 203 to in order to get access to the network. In PPTP terminology, 204 this is referred to as the PPTP Access Concentrator (PAC), and in 205 L2TP terminology, it is referred to as the L2TP Access 206 Concentrator (LAC). In IEEE 802.11, it is referred to as an 207 Access Point. 209 Roaming Capability 211 Roaming capability can be loosely defined as the ability to use 212 any one of multiple Internet Service Providers (ISPs), while 213 maintaining a formal, customer-vendor relationship with only one. 214 Examples of cases where roaming capability might be required 215 include ISP "confederations" and ISP-provided corporate network 216 access support. 218 Normalization Canonicalization 220 These terms are defined in [RFC6365] Section 4. We incorporate 221 the definitions here by reference. 223 Locale 225 This term is defined in [RFC6365] Section 8. We incorporate the 226 definition here by reference. 228 Tunneling Service 230 A tunneling service is any network service enabled by tunneling 231 protocols such as PPTP, L2F, L2TP, and IPsec tunnel mode. One 232 example of a tunneling service is secure access to corporate 233 intranets via a Virtual Private Network (VPN). 235 1.2. Requirements Language 237 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 238 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 239 "OPTIONAL" in this document are to be interpreted as described in 240 [RFC2119]. 242 1.3. Purpose 244 As described in [RFC2194], there are a number of providers offering 245 network access services, and the number of Internet Service Providers 246 involved in roaming consortia is increasing rapidly. 248 In order to be able to offer roaming capability, one of the 249 requirements is to be able to identify the user's home authentication 250 server. For use in roaming, this function is accomplished via the 251 Network Access Identifier (NAI) submitted by the user to the NAS in 252 the initial network authentication. It is also expected that NASes 253 will use the NAI as part of the process of opening a new tunnel, in 254 order to determine the tunnel endpoint. 256 We also hope that other protocols can take advantage of the NAI. 257 Many protocols include authentication capabilities, including 258 defining their own identifier formats. These identifiers can then 259 end up being transported in AAA protocols, when those systems want to 260 leverage AAA for user authentication. There is therefore a need for 261 a definition of a user identifier which can be used in multiple 262 protocols. 264 While we define the NAI here, we recognize that existing protocols 265 and deployments do not always use it. AAA systems MUST therefore be 266 able to handle user identifiers which are not in the NAI format. The 267 process by which that is done is outside of the scope of this 268 document. 270 We note that this document does not make any protocol-specific 271 definitions for an identifier format, and it does not make changes to 272 any existing protocol. Instead, it defines a protocol-independent 273 form for the NAI. It is hoped that the NAI is a user identifier 274 which can be used in multiple protocols. 276 Using a common identifier simplifies deployments, as there is no need 277 to maintain multiple identifiers for one user. It simplifies 278 protocols requiring authentication, as they no longer need to specify 279 protocol-specific format for user identifiers. It increases 280 security, as it multiple identifiers allow attackers to make 281 contradictory claims without being detected. 283 In short, having a standard is better than having no standard at all. 285 1.4. Motivation 287 The changes from [RFC4282] are listed in detail in Appendix A. 288 However, some additional discussion is appropriate to motivate those 289 changes. 291 The motivation to revise [RFC4282] began with internationalization 292 concerns raised in the context of [EDUROAM]. Section 2.1 of 293 [RFC4282] defines ABNF for realms which limits the realm grammar to 294 English letters, digits, and the hyphen "-" character. The intent 295 appears to have been to encode, compare, and transport realms with 296 the ToASCII operation defined in [RFC5890]. There are a number of 297 problems with this approach: 299 * The [RFC4282] ABNF is not aligned with internationalization of DNS. 301 * The requirement in Section 2.1 that realms are ASCII conflicts 302 with the Extensible Authentication Protocol (EAP) and RADIUS, 303 which are both 8-bit clean, and which both recommend the use of 304 UTF-8 for identitifiers. 306 * Section 2.4 required mappings that are language-specific, 307 and which are nearly impossible for intermediate nodes to perform 308 correctly without information about that language. 310 * Section 2.4 requires normalization of user names, which 311 may conflict with local system or administrative requirements. 313 * The recommendations in Section 2.4 for treatment of 314 bidirectional characters have proven to be unworkable. 316 * The prohibition against use of unassigned code points in 317 Section 2.4 effectively prohibits support for new scripts. 319 * No Authentication, Authorization, and Accounting (AAA) 320 client, proxy, or server has implemented any of the requirements 321 in [RFC4282] Section 2.4, among other sections. 323 With international roaming growing in popularity, it is important for 324 these issues to be corrected in order to provide robust and inter- 325 operable network services. 327 2. NAI Definition 329 2.1. UTF-8 Syntax and Normalization 331 UTF-8 characters can be defined in terms of octets using the 332 following ABNF [RFC5234], taken from [RFC3629]: 334 UTF8-xtra-char = UTF8-2 / UTF8-3 / UTF8-4 336 UTF8-2 = %xC2-DF UTF8-tail 337 UTF8-3 = %xE0 %xA0-BF UTF8-tail / 338 %xE1-EC 2(UTF8-tail) / 339 %xED %x80-9F UTF8-tail / 340 %xEE-EF 2(UTF8-tail) 342 UTF8-4 = %xF0 %x90-BF 2( UTF8-tail ) / 343 %xF1-F3 3( UTF8-tail ) / 344 %xF4 %x80-8F 2( UTF8-tail ) 346 UTF8-tail = %x80-BF 348 These are normatively defined in [RFC3629], but are repeated in this 349 document for reasons of convenience. 351 See [RFC5198] and section 2.6 of this specification for a discussion 352 of normalization. Strings which are not in Normal Form Composed (NFC) 353 are not valid NAIs and SHOULD NOT be treated as such. 354 Implementations which expect to receive a NAI, but which instead 355 receive non-normalised (but otherwise valid) UTF-8 strings instead 356 SHOULD attempt to create a local version of the NAI, which is 357 normalized from the input identifier. This local version can then be 358 used for local processing. 360 Systems MAY accept user identifiers in forms other than the NAI. 361 This specification does not forbid that practice. It only codifies 362 the format and interpretation of the NAI. We cannot expect to change 363 existing protocols or practices. We can, however, suggest that using 364 a consistent form for a user identifier is of a benefit to the 365 community. 367 Where protocols carry identifiers which are expected to be 368 transported over an AAA protocol, it is RECOMMENDED that the 369 identifiers be in NAI format. Where the identifiers are not in the 370 NAI format, it is up to the AAA systems to discover this, and to 371 process them. This document does not suggest how that is done. 372 However, existing practice indicates that it is possible. 374 We expect that with wider use of internationalized domain names, 375 existing practices will be inadequate. We therefore define the NAI, 376 which is a user identifier that can correctly deal with 377 internationalized identifiers. 379 2.2. Formal Syntax 381 The grammar for the NAI is given below, described in Augmented 382 Backus-Naur Form (ABNF) as documented in [RFC5234]. 384 nai = utf8-username 385 nai =/ "@" utf8-realm 386 nai =/ utf8-username "@" utf8-realm 388 utf8-username = dot-string 389 dot-string = string 390 dot-string =/ dot-string "." string 391 string = utf8-atext 392 string =/ string utf8-atext 394 utf8-atext = ALPHA / DIGIT / 395 "!" / "#" / 396 "$" / "%" / 397 "&" / "'" / 398 "*" / "+" / 399 "-" / "/" / 400 "=" / "?" / 401 "^" / "_" / 402 "`" / "{" / 403 "|" / "}" / 404 "~" / 405 UTF8-xtra-char 407 utf8-realm = 1*( label "." ) label 409 label = utf8-rtext *(ldh-str) 410 ldh-str = *( utf8-rtext / "-" ) utf8-rtext 411 utf8-rtext = ALPHA / DIGIT / UTF8-xtra-char 413 2.3. NAI Length Considerations 415 Devices handling NAIs MUST support an NAI length of at least 72 416 octets. Devices SHOULD support an NAI length of 253 octets. 417 However, the following implementation issues should be considered: 419 * NAI octet length constraints may impose a more severe constraint 420 on the number of UTF-8 characters. 422 * NAIs are often transported in the User-Name attribute of the 423 Remote Authentication Dial-In User Service (RADIUS) protocol. 424 Unfortunately, RFC 2865 [RFC2865], Section 5.1, states that "the 425 ability to handle at least 63 octets is recommended." As a 426 result, it may not be possible to transfer NAIs beyond 63 octets 427 through all devices. In addition, since only a single User-Name 428 attribute may be included in a RADIUS message and the maximum 429 attribute length is 253 octets; RADIUS is unable to support NAI 430 lengths beyond 253 octets. 432 * NAIs can also be transported in the User-Name attribute of 433 Diameter [RFC6733], which supports content lengths up to 2^24 - 9 434 octets. As a result, NAIs processed only by Diameter nodes can be 435 very long. However, an NAI transported over Diameter may 436 eventually be translated to RADIUS, in which case the above 437 limitations will apply. 439 * NAIs may be transported in other protocols. Each protocol 440 can have its own limitations on maximum NAI length. 441 The above criteria should permit the widest use, and widest possible 442 inter-operability of the NAI. 444 2.4. Support for Username Privacy 446 Interpretation of the username part of the NAI depends on the realm 447 in question. Therefore, the utf8-username portion SHOULD be treated 448 as opaque data when processed by nodes that are not a part of the 449 authoritative domain (in the sense of Section 4) for that realm. 451 In some situations, NAIs are used together with a separate 452 authentication method that can transfer the username part in a more 453 secure manner to increase privacy. In this case, NAIs MAY be 454 provided in an abbreviated form by omitting the username part. 455 Omitting the username part is RECOMMENDED over using a fixed username 456 part, such as "anonymous", since it provides an unambiguous way to 457 determine whether the username is intended to uniquely identify a 458 single user. However, current practice is to use the username 459 "anonymous" instead of omitting the username part. This behavior is 460 also permitted. 462 For roaming purposes, it is typically necessary to locate the 463 appropriate backend authentication server for the given NAI before 464 the authentication conversation can proceed. As a result, the realm 465 portion is typically required in order for the authentication 466 exchange to be routed to the appropriate server. 468 2.5. International Character Sets 470 This specification allows both international usernames and realms. 471 International usernames are based on the use of Unicode characters, 472 encoded as UTF-8. Internationalization of the realm portion of the 473 NAI is based on "Internationalized Email Headers" [RFC5335]. 475 In order to ensure a canonical representation, characters of the 476 username portion in an NAI MUST match the ABNF in this specification 477 as well as the requirements specified in [RFC5891]. In practice, 478 these requirements consist of the following item: 480 * Realms MUST be of the form that can be registered as a 481 Fully Qualified Domain Name (FQDN) within the DNS. 483 This list is significantly shorter and simpler than the list in 484 Section 2.4 of [RFC4282]. The form suggested in [RFC4282] depended 485 on intermediate nodes performing canonicalizations based on 486 insufficient information, which meant that the form was not 487 canonical. 489 Specifying the realm requirement as above means that the requirements 490 depend on specifications that are referenced here, rather than copied 491 here. This allows the realm definition to be updated when the 492 referenced documents change, without requiring a revision of this 493 specification. 495 One caveat on the above recommendation is the issues noted in 496 [RFC6912]. That document notes that there are additional 497 restrictions around DNS registration which forbid some code points 498 from being valid in a DNS U-label. These restrictions cannot be 499 expressed algorithmically. 501 For this specification, that caveat means the following. Realms not 502 matching the above ABNF are not valid NAIs. However, some realms 503 which do match the ABNF are still invalid NAIs. That is, matching 504 the ABNF is a necessary, but not sufficient, requirement for an NAI. 506 In general, the above requirement means following the requirements 507 specified in [RFC5891]. 509 2.6. The Normalization Process 511 Conversion to Unicode as well as normalization SHOULD be performed by 512 end systems that take "local" text as input. These systems are best 513 suited to determine the users intent, and can best convert from 514 "local" text to a normalized form. 516 Other AAA systems such as proxies do not have access to locale and 517 character set information that is available to end systems. 518 Therefore, they can not always convert local input to Unicode. 520 That is, all processing of NAIs from "local" character sets and 521 locales to UTF-8 SHOULD be performed by edge systems, prior to the 522 NAIs entering the AAA system. Inside of an AAA system, NAIs are sent 523 over the wire in their canonical form, and this canonical form is 524 used for all NAI and/or realm comparisons. 526 Copying of localized text into fields that can subsequently be placed 527 into the RADIUS User-Name attribute is problematic. This practice 528 can result in a AAA proxy encountering non-UTF8 characters within 529 what it expects to be an NAI. An example of this requirement is 530 [RFC3579] Section 2.1, which states: 532 the NAS MUST copy the contents of the Type-Data field of the 533 EAP-Response/Identity received from the peer into the User-Name 534 attribute 536 As a result, AAA proxies expect the contents of the EAP- 537 Response/Identity sent by an EAP supplicant to consist of UTF-8 538 characters, not localized text. Using localized text in AAA username 539 or identity fields means that realm routing becomes difficult or 540 impossible. 542 In contrast to [RFC4282] Section 2.4, we expect AAA systems to 543 perform NAI comparisons, matching, and AAA routing based on the NAI 544 as it is received. This specification provides a canonical 545 representation, ensures that intermediate systems such as AAA proxies 546 do not need to perform translations, and can be expected to work 547 through systems that are unaware of international character sets. 549 In short, 551 * End systems using "localized" text SHOULD normalize the NAI 552 prior to it being used as an identifier in an authentication 553 protocol. 555 * AAA systems SHOULD NOT normalize the NAI, as they may not have 556 sufficient information to perform the normalization. 558 For example, much of the common realm routing can be done on the 559 "utf8-realm" portion of NAI, through simple checks for equality. 560 This routing can be done even if the AAA proxy is unaware of 561 internalized domain names. All that is required is for the AAA proxy 562 to be able to enter, store, and compare 8-bit data. 564 2.6.1. Issues with the Normalization Process 566 We recognize that the requirements in the preceding section are not 567 implemented today. For example, most EAP implementations use a user 568 identifier which is passed to them from some other local system. 569 This identifier is treated as an opaque blob, and is placed as-is 570 into the EAP Identity field. Any subsequent system which receives 571 that identifier is assumed to be able to understand and process it. 573 This opaque blob unfortunately can contain localized text, which 574 means that the AAA systems have to process that text. 576 These limitations have the following theoretical and practical 577 implications. 579 * "local" systems used today generally do not normalize the NAI 581 * Therefore AAA systems SHOUD attempt to normalize the NAI 583 The suggestion in the above sentence contradicts the suggestion in 584 the previous section. This is the reality of imperfect protocols. 586 Where the user identifier can be normalized, or determined to be in 587 normal form, the normal form MUST be used as the NAI. In all other 588 circumstances, the user identifier MUST NOT be treated as an NAI. 589 That data is still, however, a user identifier. AAA systems MUST NOT 590 fail authentication simply because the user identifier is not an NAI. 592 That is, when the realm portion of the NAI is not recognized by an 593 AAA server, it SHOULD try to normalize the NAI into NFC form. That 594 normalized form can then be used to see if the realm matches a known 595 realm. If no match is found, the original form of the NAI SHOULD be 596 used in all subsequent processing. 598 The AAA server may also convert realms to punycode, and perform all 599 realm comparisons on the resulting punycode strings. This conversion 600 follows the recommendations above, but may have different operational 601 effects and failure modes. 603 2.7. Use in Other Protocols 605 As noted earlier, the NAI MAY be used in other, non-AAA protocols. 606 It is RECOMMENDED that the definition given here be used unchanged. 607 Using other definitions for user identifiers may hinder 608 interoperability, along with the users ability to authenticate 609 successfully. It is RECOMMENDED that protocols requiring the use of 610 a user identifier reference this specification, and suggest that the 611 use of an NAI is RECOMMENDED. 613 We cannot require other protocols to use the NAI for user 614 identifiers. Their needs are unknown, and unknowable. We simply 615 suggest that interoperability and inter-domain authentication is 616 useful, and should be encouraged. 618 Where a protocol is 8-bit clean, it can likely transport the NAI as- 619 is, without further modification. 621 Where a protocol is not 8-bit clean, it cannot transport the NAI as- 622 is. Instead, we presume that a protocol-specific transport layer 623 takes care of encoding the NAI on input to the protocol, and decoding 624 it when the NAI exits the protocol. The encoded or escaped version 625 of the NAI is not a valid NAI, and MUST NOT be presented to the AAA 626 system. 628 For example, HTTP carries user identifiers, but escapes the '.' 629 character as "%2E" (among others). When we desire HTTP to transport 630 the NAI "fred@example.com", the data as transported will be in the 631 form "fred@example%2Ecom". That data exists only within HTTP, and 632 has no relevance to any AAA system. 634 Any comparison, validation, or use of the NAI MUST be done on its un- 635 escaped (i.e. utf8-clean) form. 637 2.8. Using the NAI format for other identifiers 639 As discussed in Section 1, above, is RECOMMENDED that the NAI format 640 be used as the standard format for user identifiers. This section 641 discusses that use in more detail. 643 It is often useful to create new identifiers for use in specific 644 contexts. These identifiers may have a number of different 645 properties, most of which are unimportant to this document. For our 646 purposes, we are interested in identifiers which need to be in a 647 well-known format, and to have namespaces. The NAI format fits these 648 requirements. 650 One example of such use is the "private user identity", which is 651 defined by the 3rd-Generation Partnership Project (3GPP). That 652 identity is used to uniquely identify the user to the network. The 653 identity is used for authorization, authentication, accounting, 654 administation, etc. The private user identity is globally unique, 655 and is defined by the home network operator. The format of the 656 identity is explicitly the NAI, as stated by Section 13.3 of [3GPP]: 658 The private user identity shall take the form of an NAI, and shall 659 have the form username@realm as specified in clause 2.1 of IETF 660 RFC 4282 662 For 3GPP, the "username" portion is a unique identifier which is 663 derived from device-specific information. The "realm" portion is 664 composed of information about the home network, followed by the base 665 string "3gppnetwork.org". e.g. 666 234150999999999@ims.mnc015.mcc234.3gppnetwork.org. 668 This format ensures that the identifier is globally unique, as it is 669 based off of the "3gppnetwork.org" domain. It ensures that the 670 "realm" portion is specific to a particular home network (or 671 organization), via the "ims.mnc015.mcc234" prefix to the realm. 672 Finally, it ensures that the "username" portion follows a well-known 673 format. 675 We suggest that the NAI format be used for all new specifications 676 and/or protocols where a user identifier is required. It is 677 RECOMMENDED that methods similar to that described above for 3GPP be 678 used to derive the identifier. 680 3. Routing inside of AAA Systems 682 Many AAA systems use the "utf8-realm" portion of the NAI to route 683 requests within a AAA proxy network. The semantics of this operation 684 involves a logical AAA routing table, where the "utf8-realm" portion 685 acts as a key, and the values stored in the table are one or more 686 "next hop" AAA servers. 688 Intermediate nodes MUST use the "utf8-realm" portion of the NAI 689 without modification to perform this lookup. As noted earlier, 690 intermediate nodes may not have access to the same locale information 691 as the system which injected the NAI into the AAA routing systems. 692 Therefore, almost all "case insensitive" comparisons will be wrong. 693 Where the "utf8-realm" is entirely ASCII, current systems sometimes 694 perform case-insensitive matching on realms. This practice MAY be 695 continued, as it has been shown to work in practice. 697 We also note that many existing systems have user identifiers which 698 are similar in format to the NAI, but which are not compliant with 699 this specification. For example, they may use non-NFC form, or they 700 may have multiple "@" characters in the user identifier. 701 Intermediate nodes SHOULD normalize non-NFC identifiers to NFC, prior 702 to looking up the "utf8-realm" in the logical routing table. 703 Intermediate nodes MUST NOT modify the identifiers that they forward. 704 The data as entered by the user is inviolate. 706 The "utf8-realm" provisioned in the logical AAA routing table SHOULD 707 be provisioned to the proxy prior to it receiving any AAA traffic. 708 The "utf8-realm" SHOULD be supplied by the "next hop" or "home" 709 system that also supplies the routing information necessary for 710 packets to reach the next hop. 712 This "next hop" information may be any of, or all of, the following 713 information: IP address; port; RADIUS shared secret; TLS certificate; 714 DNS host name; or instruction to use dyanmic DNS discovery (i.e. look 715 up a record in the "utf8-realm" domain). This list is not 716 exhaustive, and my be extended by future specifications. 718 It is RECOMMENDED to use the entirety of the "utf8-realm" for the 719 routing decisions. However, systems MAY use a portion of the 720 "utf8-realm" portion, so long as that portion is a valid 721 "utf8-realm", and that portion is handled as above. For example, 722 routing "fred@example.com" to a "com" destination is forbidden, 723 because "com" is not a valid "utf8-realm". However, routing 724 "fred@sales.example.com" to the "example.com" destination is 725 permissible. 727 Another reason to forbid the use of a single label (e.g. 728 "fred@sales") is that many systems treat a single label as being a 729 local identifier within their realm. That is, a user logging in as 730 "fred@sales" to a domain "example.com", would be treated as if the 731 NAI was instead "fred@sales.example.com". Permitting the use of a 732 single label would mean changing the interpretation and meaning of a 733 single label, which cannot be done. 735 3.1. Compatibility with Email Usernames 737 As proposed in this document, the Network Access Identifier is of the 738 form "user@realm". Please note that while the user portion of the 739 NAI is based on the BNF described in [RFC5198], it has been modified 740 for the purposes of Section 2.2. It does not permit quoted text 741 along with "folding" or "non-folding" whitespace that is commonly 742 used in email addresses. As such, the NAI is not necessarily 743 equivalent to usernames used in e-mail. 745 However, it is a common practice to use email addresses as user 746 identifiers in AAA systems. The ABNF in Section 2.2 is defined to be 747 close to the "utf8-addr-spec" portion of [RFC5335], while still being 748 compatible with [RFC4282]. 750 In contrast to [RFC4282] Section 2.5, we state that the 751 internationalization requirements for NAIs and email addresses are 752 substantially similar. The NAI and email identifiers may be the 753 same, and both need to be entered by the user and/or the operator 754 supplying network access to that user. There is therefore good 755 reason for the internationalization requirements to be similar. 757 3.2. Compatibility with DNS 759 The "utf8-realm" portion of the NAI is intended to be compatible with 760 Internationalized Domain Names (IDNs) [RFC5890]. As defined above, 761 the "utf8-realm" portion as transported within an 8-bit clean 762 protocol such as RADIUS and EAP can contain any valid UTF8 character. 763 There is therefore no reason for a NAS to apply the ToAscii function 764 to the "utf8-realm" portion of an NAI, prior to placing the NAI into 765 a RADIUS User-Name attribute. 767 The NAI does not make a distinction between A-labels and U-labels, as 768 those are terms specific to DNS. It is instead an IDNA-valid label, 769 as per the first item in Section 2.3.2.1 of [RFC5890]. As noted in 770 that section, the term "IDNA-valid label" encompases both of the 771 terms A-label and U-label. 773 When the realm portion of the NAI is used as the basis for name 774 resolution, it may be necessary to convert internationalized realm 775 names to ASCII using the ToASCII operation defined in [RFC5890]. As 776 noted in [RFC6055] Section 2, resolver Application Programming 777 Interfaces (APIs) are not necessarily DNS-specific, so that the 778 ToASCII operation needs to be applied carefully: 780 Applications which convert an IDN to A-label form before calling (for 781 example) getaddrinfo() will result in name resolution failures if the 782 Punycode name is directly used in such protocols. Having libraries 783 or protocols to convert from A-labels to the encoding scheme defined 784 by the protocol (e.g., UTF-8) would require changes to APIs and/or 785 servers, which IDNA was intended to avoid. 787 As a result, applications SHOULD NOT assume that non-ASCII names are 788 resolvable using the public DNS and blindly convert them to A-labels 789 without knowledge of what protocol will be selected by the name 790 resolution library. 792 3.3. Realm Construction 794 The home realm usually appears in the "utf8-realm" portion of the 795 NAI, but in some cases a different realm can be used. This may be 796 useful, for instance, when the home realm is reachable only via 797 intermediate proxies. 799 Such usage may prevent interoperability unless the parties involved 800 have a mutual agreement that the usage is allowed. In particular, 801 NAIs MUST NOT use a different realm than the home realm unless the 802 sender has explicit knowledge that (a) the specified other realm is 803 available and (b) the other realm supports such usage. The sender 804 may determine the fulfillment of these conditions through a database, 805 dynamic discovery, or other means not specified here. Note that the 806 first condition is affected by roaming, as the availability of the 807 other realm may depend on the user's location or the desired 808 application. 810 The use of the home realm MUST be the default unless otherwise 811 configured. 813 3.3.1. Historical Practices 815 Some systems have historically used NAI modifications with multiple 816 "prefix" and "suffix" decorations to perform explicit routing through 817 multiple proxies inside of a AAA network. This practice is NOT 818 RECOMMENDED for the following reasons: 820 * Using explicit routing paths is fragile, and is unresponsive to 821 changes in the network due to servers going up or down, or to 822 changing business relationships. 824 * There is no RADIUS routing protocol, meaning that routing paths 825 have to be communicated "out of band" to all intermediate AAA 826 nodes, and also to all end-user systems (supplicants) expecting to 827 obtain network access. 829 * Using explicit routing paths requires thousands, if not 830 millions of end-user systems to be updated with new path 831 information when a AAA routing path changes. This adds huge 832 expense for updates that would be better done at only a few AAA 833 systems in the network. 835 * Manual updates to RADIUS paths are expensive, time-consuming, 836 and prone to error. 838 * Creating compatible formats for the NAI is difficult 839 when locally-defined "prefixes" and "suffixes" conflict with 840 similar practices elsewhere in the network. These conflicts mean 841 that connecting two networks may be impossible in some cases, as 842 there is no way for packets to be routed properly in a way that 843 meets all requirements at all intermediate proxies. 845 * Leveraging the DNS name system for realm names establishes 846 a globally unique name space for realms. 848 In summary, network practices and capabilities have changed 849 significantly since NAIs were first overloaded to define AAA routes 850 through a network. While explicit path routing was once useful, the 851 time has come for better methods to be used. 853 3.4. Examples 855 Examples of valid Network Access Identifiers include the following: 857 bob 858 joe@example.com 859 fred@foo-9.example.com 860 jack@3rd.depts.example.com 861 fred.smith@example.com 862 fred_smith@example.com 863 fred$@example.com 864 fred=?#$&*+-/^smith@example.com 865 nancy@eng.example.net 866 eng.example.net!nancy@example.net 867 eng%nancy@example.net 868 @privatecorp.example.net 869 \(user\)@example.net 871 An additional valid NAI is the following, given as a hex string, as 872 this document can only contain ASCII characters. 874 626f 6240 ceb4 cebf ceba ceb9 cebc ceae 2e63 6f6d 876 Examples of invalid Network Access Identifiers include the following: 878 fred@example 879 fred@example_9.com 880 fred@example.net@example.net 881 fred.@example.net 882 eng:nancy@example.net 883 eng;nancy@example.net 884 (user)@example.net 885 @example.net 887 One example given in [RFC4282] is still permitted by the ABNF, but it 888 is NOT RECOMMMENDED because of the use of the ToAscii function to 889 create an ASCII encoding from what is now a valid UTF-8 string. 891 alice@xn--tmonesimerkki-bfbb.example.net 893 4. Security Considerations 895 Since an NAI reveals the home affiliation of a user, it may assist an 896 attacker in further probing the username space. Typically, this 897 problem is of most concern in protocols that transmit the username in 898 clear-text across the Internet, such as in RADIUS, described in 899 [RFC2865] and [RFC2866]. In order to prevent snooping of the 900 username, protocols may use confidentiality services provided by 901 protocols transporting them, such as RADIUS protected by IPsec 902 [RFC3579] or Diameter protected by TLS [RFC6733]. 904 This specification adds the possibility of hiding the username part 905 in the NAI, by omitting it. As discussed in Section 2.4, this is 906 possible only when NAIs are used together with a separate 907 authentication method that can transfer the username in a secure 908 manner. In some cases, application-specific privacy mechanism have 909 also been used with NAIs. For instance, some EAP methods apply 910 method-specific pseudonyms in the username part of the NAI [RFC3748]. 911 While neither of these approaches can protect the realm part, their 912 advantage over transport protection is that privacy of the username 913 is protected, even through intermediate nodes such as NASes. 915 5. Administration of Names 916 In order to avoid creating any new administrative procedures, 917 administration of the NAI realm namespace piggybacks on the 918 administration of the DNS namespace. 920 NAI realm names are required to be unique, and the rights to use a 921 given NAI realm for roaming purposes are obtained coincident with 922 acquiring the rights to use a particular Fully Qualified Domain Name 923 (FQDN). Those wishing to use an NAI realm name should first acquire 924 the rights to use the corresponding FQDN. Administrators MUST NOT 925 publicly use an NAI realm without first owning the corresponding 926 FQDN. Private use of unowned NAI realms within an administative 927 domain is allowed, though it is RECOMMENDED that example names be 928 used, such as "example.com". 930 Note that the use of an FQDN as the realm name does not require use 931 of the DNS for location of the authentication server. While Diameter 932 [RFC6733] supports the use of DNS for location of authentication 933 servers, existing RADIUS implementations typically use proxy 934 configuration files in order to locate authentication servers within 935 a domain and perform authentication routing. The implementations 936 described in [RFC2194] did not use DNS for location of the 937 authentication server within a domain. Similarly, existing 938 implementations have not found a need for dynamic routing protocols 939 or propagation of global routing information. Note also that there 940 is no requirement that the NAI represent a valid email address. 942 6. IANA Considerations 944 This document has no actions for IANA. 946 7. References 948 7.1. Normative References 950 [RFC2119] 951 Bradner, S., "Key words for use in RFCs to Indicate Requirement 952 Levels", RFC 2119, March, 1997. 954 [RFC3629] 955 Yergeau, F., "UTF-8, a transformation format of ISO 10646", STD 63, 956 RFC 3629, November 2003. 958 [RFC5198] 959 Klensin J., and Padlipsky M., "Unicode Format for Network 960 Interchange", RFC 5198, March 2008 962 [RFC5234] 963 Crocker, D. and P. Overell, "Augmented BNF for Syntax 964 Specifications: ABNF", RFC 5234, January 2008. 966 [RFC5890] 967 Faltstrom, P., Hoffman, P., and A. Costello, "Internationalizing 968 Domain Names in Applications (IDNA)", RFC 5890, August 2010 970 [RFC6365] 971 Hoffman, P., and Klensin, J., "Terminology Used in 972 Internationalization in the IETF", RFC 6365, September 2011 974 7.2. Informative References 976 [RFC2194] 977 Aboba, B., Lu, J., Alsop, J., Ding, J., and W. Wang, "Review of 978 Roaming Implementations", RFC 2194, September 1997. 980 [RFC2341] 981 Valencia, A., Littlewood, M., and T. Kolar, "Cisco Layer Two 982 Forwarding (Protocol) "L2F"", RFC 2341, May 1998. 984 [RFC2637] 985 Hamzeh, K., Pall, G., Verthein, W., Taarud, J., Little, W., and G. 986 Zorn, "Point-to-Point Tunneling Protocol", RFC 2637, July 1999. 988 [RFC2661] 989 Townsley, W., Valencia, A., Rubens, A., Pall, G., Zorn, G., and B. 990 Palter, "Layer Two Tunneling Protocol "L2TP"", RFC 2661, August 991 1999. 993 [RFC2865] 994 Rigney, C., Willens, S., Rubens, A. and W. Simpson, "Remote 995 Authentication Dial In User Service (RADIUS)", RFC 2865, June 2000. 997 [RFC2866] 998 Rigney, C., "RADIUS Accounting", RFC 2866, June 2000. 1000 [RFC3579] 1001 Aboba, B. and P. Calhoun, "RADIUS (Remote Authentication Dial In 1002 User Service) Support For Extensible Authentication Protocol 1003 (EAP)", RFC 3579, September 2003. 1005 [RFC3748] 1006 Aboba, B., Blunk, L., Vollbrecht, J., Carlson, J., and H. 1007 Levkowetz, "Extensible Authentication Protocol (EAP)", RFC 3748, 1008 June 2004. 1010 [RFC4282] 1011 Aboba, B. et al., "The Network Access Identifier", RFC 4282, 1012 December 2005. 1014 [RFC4301] 1015 Kent, S. and S. Keo, "Security Architecture for the Internet 1016 Protocol", RFC 4301, December 2005. 1018 [RFC5335] 1019 Y. Abel, Ed., "Internationalized Email Headers", RFC 5335, 1020 September 2008. 1022 [EDUROAM] 1023 http://eduroam.org, "eduroam (EDUcational ROAMing)" 1025 [RFC5891] 1026 Klensin, J., "Internationalized Domain Names in Applications 1027 (IDNA): Protocol", RFC 5891 1029 [RFC6055] 1030 Thaler, D., et al, "IAB Thoughts on Encodings for Internationalized 1031 Domain Names", RFC 6055, February 2011. 1033 [RFC6733] 1034 V. Fajardo, Ed., et al, "Diameter Base Protocol", RFC 6733, October 1035 2012. 1037 [RFC6912] 1038 Sullivan, A., et al, "Principles for Unicode Code Point Inclusion 1039 in Labels in the DNS", RFC 6912, April 2013. 1041 [3GPP] 1042 3GPP, "TS 23.003 Numbering, addressing, and Identification (Release 1043 12)", July 2014, 1044 ftp://ftp.3gpp.org/Specs/archive/23_series/23.003/. 1046 Acknowledgments 1048 The initial text for this document was [RFC4282], which was then 1049 heavily edited. The original authors of [RFC4282] were Bernard 1050 Aboba, Mark A. Beadles, Jari Arkko, and Pasi Eronen. 1052 The ABNF validator at http://www.apps.ietf.org/abnf.html was used to 1053 verify the syntactic correctness of the ABNF in Section 2. 1055 Appendix A - Changes from RFC4282 1057 This document contains the following updates with respect to the 1058 previous NAI definition in RFC 4282 [RFC4282]: 1060 * The formal syntax in Section 2.1 has been updated to forbid 1061 non-UTF8 characters. e.g. characters with the "high bit" set. 1063 * The formal syntax in Section 2.1 has been updated to allow 1064 UTF-8 in the "realm" portion of the NAI. 1066 * The formal syntax in [RFC4282] Section 2.1 applied to the 1067 NAI after it was "internationalized" via the ToAscii function. 1068 The contents of the NAI before it was "internationalized" were 1069 left indeterminate. This document updates the formal syntax to 1070 define an internationalized form of the NAI, and forbids the use 1071 of the ToAscii function for NAI "internationalization". 1073 * The grammar for the user and realm portion is based on a 1074 combination 1075 of the "nai" defined in [RFC4282] Section 2.1, and the "utf8-addr- 1076 spec" defined in [RFC5335] Section 4.4. 1078 * All use of the ToAscii function has been moved to normal 1079 requirements on DNS implementations when realms are used as the 1080 basis for DNS lookups. This involves no changes to the existing 1081 DNS infrastructure. 1083 * The discussions on internationalized character sets in Section 2.4 1084 have been updated. The suggestion to use the ToAscii function for 1085 realm comparisons has been removed. No AAA system has implemented 1086 these suggestions, so this change should have no operational 1087 impact. 1089 * The section "Routing inside of AAA Systems" section is new in this 1090 document. The concept of a "local AAA routing table" is also new, 1091 although it accurately describes the functionality of wide-spread 1092 implementations. 1094 * The "Compatibility with EMail Usernames" and "Compatibility 1095 with DNS" sections have been revised and updated. We now note 1096 that the ToAscii function is suggested to be used only when a 1097 realm name is used for DNS lookups, and even then the function is 1098 only used by a resolving API on the local system, and even then we 1099 recommend that only the home network perform this conversion. 1101 * The "Realm Construction" section has been updated to note 1102 that editing of the NAI is NOT RECOMMENDED. 1104 * The "Examples" section has been updated to remove the instance 1105 of the IDN being converted to ASCII. This behavior is now 1106 forbidden. 1108 Authors' Addresses 1110 Alan DeKok 1111 The FreeRADIUS Server Project 1113 Email: aland@freeradius.org