idnits 2.17.00 (12 Aug 2021) /tmp/idnits50596/draft-ietf-abfab-arch-08.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 document seems to lack a both a reference to RFC 2119 and the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. RFC 2119 keyword, line 505: '...method to be used. The IdP SHOULD NOT...' RFC 2119 keyword, line 507: '... this case the IdP MUST accept a realm...' RFC 2119 keyword, line 1341: '...A infrastructure MAY hide the initiato...' RFC 2119 keyword, line 1829: '... by an IdP, care MUST be taken in the ...' Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (November 3, 2013) is 3120 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- -- Looks like a reference, but probably isn't: '1' on line 142 -- Looks like a reference, but probably isn't: '2' on line 754 == Unused Reference: 'RFC2903' is defined on line 1897, but no explicit reference was found in the text == Unused Reference: 'I-D.nir-tls-eap' is defined on line 1901, but no explicit reference was found in the text == Unused Reference: 'RFC4017' is defined on line 1924, but no explicit reference was found in the text == Unused Reference: 'RFC5106' is defined on line 1928, but no explicit reference was found in the text == Unused Reference: 'RFC2138' is defined on line 1944, but no explicit reference was found in the text == Unused Reference: 'RFC2904' is defined on line 1984, but no explicit reference was found in the text ** Obsolete normative reference: RFC 3588 (Obsoleted by RFC 6733) == Outdated reference: draft-ietf-abfab-gss-eap has been published as RFC 7055 == Outdated reference: draft-ietf-abfab-aaa-saml has been published as RFC 7833 == Outdated reference: draft-ietf-radext-nai has been published as RFC 7542 == Outdated reference: draft-ietf-oauth-v2 has been published as RFC 6749 == Outdated reference: draft-iab-privacy-considerations has been published as RFC 6973 == Outdated reference: A later version (-06) exists of draft-perez-radext-radius-fragmentation-05 -- Obsolete informational reference (is this intentional?): RFC 2138 (Obsoleted by RFC 2865) -- Obsolete informational reference (is this intentional?): RFC 5849 (Obsoleted by RFC 6749) == Outdated reference: draft-ietf-emu-crypto-bind has been published as RFC 7029 -- No information found for draft-ietf-emu-eap-tunnel-method - is the name correct? == Outdated reference: draft-ietf-radext-dtls has been published as RFC 7360 == Outdated reference: draft-ietf-radext-dynamic-discovery has been published as RFC 7585 Summary: 2 errors (**), 0 flaws (~~), 16 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 ABFAB J. Howlett 3 Internet-Draft JANET(UK) 4 Intended status: Informational S. Hartman 5 Expires: May 7, 2014 Painless Security 6 H. Tschofenig 7 Nokia Siemens Networks 8 E. Lear 9 Cisco Systems GmbH 10 J. Schaad 11 Soaring Hawk Consulting 12 November 3, 2013 14 Application Bridging for Federated Access Beyond Web (ABFAB) 15 Architecture 16 draft-ietf-abfab-arch-08.txt 18 Abstract 20 Over the last decade a substantial amount of work has occurred in the 21 space of federated access management. Most of this effort has 22 focused on two use cases: network access and web-based access. 23 However, the solutions to these use cases that have been proposed and 24 deployed tend to have few common building blocks in common. 26 This memo describes an architecture that makes use of extensions to 27 the commonly used security mechanisms for both federated and non- 28 federated access management, including the Remote Authentication Dial 29 In User Service (RADIUS) and the Diameter protocol, the Generic 30 Security Service (GSS), the Extensible Authentication Protocol (EAP) 31 and the Security Assertion Markup Language (SAML). The architecture 32 addresses the problem of federated access management to primarily 33 non-web-based services, in a manner that will scale to large numbers 34 of identity providers, relying parties, and federations. 36 Status of This Memo 38 This Internet-Draft is submitted in full conformance with the 39 provisions of BCP 78 and BCP 79. 41 Internet-Drafts are working documents of the Internet Engineering 42 Task Force (IETF). Note that other groups may also distribute 43 working documents as Internet-Drafts. The list of current Internet- 44 Drafts is at http://datatracker.ietf.org/drafts/current/. 46 Internet-Drafts are draft documents valid for a maximum of six months 47 and may be updated, replaced, or obsoleted by other documents at any 48 time. It is inappropriate to use Internet-Drafts as reference 49 material or to cite them other than as "work in progress." 51 This Internet-Draft will expire on May 7, 2014. 53 Copyright Notice 55 Copyright (c) 2013 IETF Trust and the persons identified as the 56 document authors. All rights reserved. 58 This document is subject to BCP 78 and the IETF Trust's Legal 59 Provisions Relating to IETF Documents 60 (http://trustee.ietf.org/license-info) in effect on the date of 61 publication of this document. Please review these documents 62 carefully, as they describe your rights and restrictions with respect 63 to this document. Code Components extracted from this document must 64 include Simplified BSD License text as described in Section 4.e of 65 the Trust Legal Provisions and are provided without warranty as 66 described in the Simplified BSD License. 68 Table of Contents 70 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 71 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 5 72 1.1.1. Channel Binding . . . . . . . . . . . . . . . . . . . 6 73 1.2. An Overview of Federation . . . . . . . . . . . . . . . . 7 74 1.3. Challenges for Contemporary Federation . . . . . . . . . 11 75 1.4. An Overview of ABFAB-based Federation . . . . . . . . . . 11 76 1.5. Design Goals . . . . . . . . . . . . . . . . . . . . . . 14 77 2. Architecture . . . . . . . . . . . . . . . . . . . . . . . . 15 78 2.1. Relying Party to Identity Provider . . . . . . . . . . . 16 79 2.1.1. AAA, RADIUS and Diameter . . . . . . . . . . . . . . 17 80 2.1.2. Discovery and Rules Determination . . . . . . . . . . 18 81 2.1.3. Routing and Technical Trust . . . . . . . . . . . . . 19 82 2.1.4. AAA Security . . . . . . . . . . . . . . . . . . . . 21 83 2.1.5. SAML Assertions . . . . . . . . . . . . . . . . . . . 21 84 2.2. Client To Identity Provider . . . . . . . . . . . . . . . 23 85 2.2.1. Extensible Authentication Protocol (EAP) . . . . . . 24 86 2.2.2. EAP Channel Binding . . . . . . . . . . . . . . . . . 25 87 2.3. Client to Relying Party . . . . . . . . . . . . . . . . . 26 88 2.3.1. GSS-API . . . . . . . . . . . . . . . . . . . . . . . 26 89 2.3.2. Protocol Transport . . . . . . . . . . . . . . . . . 28 90 2.3.3. Reauthentication . . . . . . . . . . . . . . . . . . 28 91 3. Application Security Services . . . . . . . . . . . . . . . . 28 92 3.1. Authentication . . . . . . . . . . . . . . . . . . . . . 29 93 3.2. GSS-API Channel Binding . . . . . . . . . . . . . . . . . 30 94 3.3. Host-Based Service Names . . . . . . . . . . . . . . . . 31 95 3.4. Additional GSS-API Services . . . . . . . . . . . . . . . 33 96 4. Privacy Considerations . . . . . . . . . . . . . . . . . . . 33 97 4.1. Entities and their roles . . . . . . . . . . . . . . . . 34 98 4.2. Privacy Aspects of ABFAB Communication Flows . . . . . . 35 99 4.2.1. Client to RP . . . . . . . . . . . . . . . . . . . . 35 100 4.2.2. Client to IdP (via Federation Substrate) . . . . . . 36 101 4.2.3. IdP to RP (via Federation Substrate) . . . . . . . . 37 102 4.3. Relationship between User and Entities . . . . . . . . . 38 103 4.4. Accounting Information . . . . . . . . . . . . . . . . . 38 104 4.5. Collection and retention of data and identifiers . . . . 38 105 4.6. User Participation . . . . . . . . . . . . . . . . . . . 39 106 5. Security Considerations . . . . . . . . . . . . . . . . . . . 39 107 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 40 108 7. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 40 109 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 41 110 8.1. Normative References . . . . . . . . . . . . . . . . . . 41 111 8.2. Informative References . . . . . . . . . . . . . . . . . 42 112 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 44 114 1. Introduction 116 The Internet uses numerous security mechanisms to manage access to 117 various resources. These mechanisms have been generalized and scaled 118 over the last decade through mechanisms such as Simple Authentication 119 and Security Layer (SASL) with the Generic Security Server 120 Application Program Interface (GSS-API) (known as the GS2 family) 121 [RFC5801], Security Assertion Markup Language (SAML) 122 [OASIS.saml-core-2.0-os], and the Authentication, Authorization, and 123 Accounting (AAA) architecture as embodied in RADIUS [RFC2865] and 124 Diameter [RFC3588]. 126 A Relying Party (RP) is the entity that manages access to some 127 resource. The entity that is requesting access to that resource is 128 often described as the Client. Many security mechanisms are 129 manifested as an exchange of information between these entities. The 130 RP is therefore able to decide whether the Client is authorized, or 131 not. 133 Some security mechanisms allow the RP to delegate aspects of the 134 access management decision to an entity called the Identity Provider 135 (IdP). This delegation requires technical signaling, trust and a 136 common understanding of semantics between the RP and IdP. These 137 aspects are generally managed within a relationship known as a 138 'federation'. This style of access management is accordingly 139 described as 'federated access management'. 141 Federated access management has evolved over the last decade through 142 specifications like SAML [OASIS.saml-core-2.0-os], OpenID [1], OAuth 143 [RFC5849], [I-D.ietf-oauth-v2] and WS-Trust [WS-TRUST]. The benefits 144 of federated access management include: 146 Single or Simplified sign-on: 148 An Internet service can delegate access management, and the 149 associated responsibilities such as identity management and 150 credentialing, to an organization that already has a long-term 151 relationship with the Client. This is often attractive as Relying 152 Parties frequently do not want these responsibilities. The Client 153 also requires fewer credentials, which is also desirable. 155 Data Minimization and User Participation: 157 Often a Relying Party does not need to know the identity of a 158 Client to reach an access management decision. It is frequently 159 only necessary for the Relying Party to know specific attributes 160 about the client, for example, that the client is affiliated with 161 a particular organization or has a certain role or entitlement. 162 Sometimes the RP only needs to know a pseudonym of the client. 164 Prior to the release of attributes to the RP from the IdP, the IdP 165 will check configuration and policy to determine if the attributes 166 are to be released. There is currently no direct client 167 participation in this decision. 169 Provisioning: 171 Sometimes a Relying Party needs, or would like, to know more about 172 a client than an affiliation or a pseudonym. For example, a 173 Relying Party may want the Client's email address or name. Some 174 federated access management technologies provide the ability for 175 the IdP to supply this information, either on request by the RP or 176 unsolicited. 178 This memo describes the Application Bridging for Federated Access 179 Beyond the Web (ABFAB) architecture. This architecture makes use of 180 extensions to the commonly used security mechanisms for both 181 federated and non-federated access management, including the RADIUS 182 and the Diameter protocols, the Generic Security Service (GSS), the 183 Extensible Authentication Protocol (EAP) and SAML. The architecture 184 addresses the problem of federated access management primarily for 185 non-web-based services. It does so in a manner that will scale to 186 large numbers of identity providers, relying parties, and 187 federations. 189 1.1. Terminology 191 This document uses identity management and privacy terminology from 192 [I-D.iab-privacy-considerations]. In particular, this document uses 193 the terms identity provider, relying party, identifier, pseudonymity, 194 unlinkability, and anonymity. 196 In this architecture the IdP consists of the following components: an 197 EAP server, a RADIUS or a Diameter server, and optionally a SAML 198 Assertion service. 200 This document uses the term Network Access Identifier (NAI), as 201 defined in [I-D.ietf-radext-nai]. An NAI consists of a realm 202 identifier, which is associated with an IdP and a username which is 203 associated with a specific client of the IdP. 205 One of the problems people will find with reading this document is 206 that the terminology sometimes appears to be inconsistent. This is 207 due the fact that the terms used by the different standards we are 208 referencing are not consistent. In general the document uses either 209 the ABFAB term or the term associated with the standard under 210 discussion as appropriate. For reference we include this table which 211 maps the different terms into a single table. 213 +------------+-------------+---------------------+------------------+ 214 | Protocol | Client | Relying Party | Identity | 215 | | | | Provider | 216 +------------+-------------+---------------------+------------------+ 217 | ABFAB | Client | Relying Party (RP) | Identity | 218 | | | | Provider (IdP) | 219 | | | | | 220 | | Initiator | Acceptor | | 221 | | | | | 222 | | | Server | | 223 | | | | | 224 | SAML | Subject | Service Provider | Issuer | 225 | | | | | 226 | GSS-API | Initiator | Acceptor | | 227 | | | | | 228 | EAP | EAP peer | EAP Authenticator | EAP server | 229 | | | | | 230 | AAA | | AAA Client | AAA server | 231 | | | | | 232 | RADIUS | user | NAS | RADIUS server | 233 | | | | | 234 | | | RADIUS client | | 235 +------------+-------------+---------------------+------------------+ 237 Table 1. Terminology 239 Note that in some cases a cell has been left empty; in these cases 240 there is no name that represents the entity. 242 1.1.1. Channel Binding 244 This document uses the term channel binding with two different 245 meanings. 247 EAP channel binding is used to provide GSS-API naming semantics. EAP 248 channel binding sends a set of attributes from the peer to the EAP 249 server either as part of the EAP conversation or as part of a secure 250 association protocol. In addition, attributes are sent in the 251 backend protocol from the EAP authenticator to the EAP server. The 252 EAP server confirms the consistency of these attributes and provides 253 the confirmation back to the peer. In this document, channel binding 254 without qualification refers to EAP channel binding. 256 GSS-API channel binding provides protection against man-in-the-middle 257 attacks when GSS-API is used for authentication inside of some 258 tunnel; it is similar to a facility called cryptographic binding in 259 EAP. The binding works by each side deriving a cryptographic value 260 from the tunnel itself and then using that cryptographic value to 261 prove to the other side that it knows the value. 263 See [RFC5056] for a discussion of the differences between these two 264 facilities. However, the difference can be summarized as GSS-API 265 channel binding says that there is nobody between the client and the 266 EAP authenticator while EAP channel binding allows the client to have 267 knowledge about attributes of the EAP authenticator (such as its 268 name). 270 Typically when considering both EAP and GSS-API channel binding, 271 people think of channel binding in combination with mutual 272 authentication. This is sufficiently common that without additional 273 qualification channel binding should be assumed to imply mutual 274 authentication. In GSS-API, without mutualtion only the acceptor has 275 authenticated the initator. Similarly in EAP, only the EAP server 276 has authenticated the peer. That's sometimes useful. Consider for 277 example a user who wishes to access a protected resource for a shared 278 whiteboard in a conference room. The whiteboard is the acceptor; it 279 knows that the initiator is authorized to give it a presentation and 280 the user can validate the whitebord got the correct presentation by 281 visual means. (The presention should not be confiduatal in this 282 case.) If channel binding is used without mutual authentication, it 283 is effectively a request to disclose the resource in the context of a 284 particular channel. Such an authentication would be similar in 285 concept to a holder-of-key SAML assertion. However, also note that 286 while it is not happening in the protocol, mutual authentication is 287 happening in the overall system: the user is able to visually 288 authenticate the content. This is consistent with all uses of 289 channel binding without protocol level mutual authentication found so 290 far. 292 1.2. An Overview of Federation 294 In the previous section we introduced the following entities: 296 o the Client, 298 o the Identity Provider, and 300 o the Relying Party. 302 The final entity that needs to be introduced is the Individual. An 303 Individual is a human being that is using the Client. In any given 304 situation, an Individual may or may not exist. Clients can act 305 either as front ends for Individuals or they may be independent 306 entities that are setup and allowed to run autonomously. An example 307 of such an entity can be found in the trust routing protocol where 308 the routers use ABFAB to authenticate to each other. 310 These entities and their relationships are illustrated graphically in 311 Figure 1. 313 ,----------\ ,---------\ 314 | Identity | Federation | Relying | 315 | Provider + <-------------------> + Party | 316 `----------' '---------' 317 < 318 \ 319 \ Authentication 320 \ 321 \ 322 \ 323 \ 324 \ +---------+ 325 \ | | O 326 v| Client | \|/ Individual 327 | | | 328 +---------+ / \ 330 Figure 1: Entities and their Relationships 332 The relationships between the entities in Figure 1 are: 334 Federation 336 The Identity Provider and the Relying Parties are part of a 337 Federation. The relationship may be direct (they have an explicit 338 trust relationship) or transitive (the trust relationship is 339 mediated by one or more entities). The federation relationship is 340 governed by a federation agreement. Within a single federation, 341 there may be multiple Identity Providers as well as multiple 342 Relying Parties. 344 Authentication 346 There is a direct relationship between the Client and the Identity 347 Provider by which the entities trust and can securely authenticate 348 each other. 350 A federation agreement typically encompasses operational 351 specifications and legal rules: 353 Operational Specifications: 355 These include the technical specifications (e.g. protocols used to 356 communicate between the three parties), process standards, 357 policies, identity proofing, credential and authentication 358 algorithm requirements, performance requirements, assessment and 359 audit criteria, etc. The goal of operational specifications is to 360 provide enough definition that the system works and 361 interoperability is possible. 363 Legal Rules: 365 The legal rules take the legal framework into consideration and 366 provide contractual obligations for each entity. The rules define 367 the responsibilities of each party and provide further 368 clarification of the operational specifications. These legal 369 rules regulate the operational specifications, make operational 370 specifications legally binding to the participants, define and 371 govern the rights and responsibilities of the participants. The 372 legal rules may, for example, describe liability for losses, 373 termination rights, enforcement mechanisms, measures of damage, 374 dispute resolution, warranties, etc. 376 The Operational Specifications can demand the usage of a 377 sophisticated technical infrastructure, including requirements on the 378 message routing intermediaries, to offer the required technical 379 functionality. In other environments, the Operational Specifications 380 require fewer technical components in order to meet the required 381 technical functionality. 383 The Legal Rules include many non-technical aspects of federation, 384 such as business practices and legal arrangements, which are outside 385 the scope of the IETF. The Legal Rules can still have an impact on 386 the architectural setup or on how to ensure the dynamic establishment 387 of trust. 389 While a federation agreement is often discussed within the context of 390 formal relationships, such as between an enterprise and an employee 391 or a government and a citizen, a federation agreement does not have 392 to require any particular level of formality. For an IdP and a 393 Client, it is sufficient for a relationship to be established by 394 something as simple as using a web form and confirmation email. For 395 an IdP and an RP, it is sufficient for the IdP to publish contact 396 information along with a public key and for the RP to use that data. 397 Within the framework of ABFAB, it will generally be required that a 398 mechanism exists for the IdP to be able to trust the identity of the 399 RP, if this is not present then the IdP cannot provide the assurances 400 to the client that the identity of the RP has been established. 402 The nature of federation dictates that there is some form of 403 relationship between the identity provider and the relying party. 404 This is particularly important when the relying party wants to use 405 information obtained from the identity provider for access management 406 decisions and when the identity provider does not want to release 407 information to every relying party (or only under certain 408 conditions). 410 While it is possible to have a bilateral agreement between every IdP 411 and every RP; on an Internet scale this setup requires the 412 introduction of the multi-lateral federation concept, as the 413 management of such pair-wise relationships would otherwise prove 414 burdensome. 416 The IdP will typically have a long-term relationship with the Client. 417 This relationship typically involves the IdP positively identifying 418 and credentialing the Client (for example, at time of employment 419 within an organization). When dealing with individuals, this process 420 is called identity proofing [NIST-SP.800-63]. The relationship will 421 often be instantiated within an agreement between the IdP and the 422 Client (for example, within an employment contract or terms of use 423 that stipulates the appropriate use of credentials and so forth). 425 The nature and quality of the relationship between the Client and the 426 IdP is an important contributor to the level of trust that an RP may 427 attribute to an assertion describing a Client made by an IdP. This 428 is sometimes described as the Level of Assurance [NIST-SP.800-63]. 430 Federation does not require an a priori relationship or a long-term 431 relationship between the RP and the Client; it is this property of 432 federation that yields many of its benefits. However, federation 433 does not preclude the possibility of a pre-existing relationship 434 between the RP and the Client, nor that they may use the introduction 435 to create a new long-term relationship independent of the federation. 437 Finally, it is important to reiterate that in some scenarios there 438 might indeed be an Individual behind the Client and in other cases 439 the Client may be autonomous. 441 1.3. Challenges for Contemporary Federation 443 As the number of federated services has proliferated, the role of the 444 individual can become ambiguous in certain circumstances. For 445 example, a school might provide online access for a student's grades 446 to their parents for review, and to the student's teacher for 447 modification. A teacher who is also a parent must clearly 448 distinguish her role upon access. 450 Similarly, as the number of federations proliferates, it becomes 451 increasingly difficult to discover which identity provider(s) a user 452 is associated with. This is true for both the web and non-web case, 453 but is particularly acute for the latter as many non-web 454 authentication systems are not semantically rich enough on their own 455 to allow for such ambiguities. For instance, in the case of an email 456 provider, the use of SMTP and IMAP protocols do not have the ability 457 for the server to get additional information, beyond the clients NAI, 458 in order to provide additional input to decide between multiple 459 federations it may be associated with. However, the building blocks 460 do exist to add this functionality. 462 1.4. An Overview of ABFAB-based Federation 464 The previous section described the general model of federation, and 465 the application of access management within the federation. This 466 section provides a brief overview of ABFAB in the context of this 467 model. 469 In this example, a client is attempting to connect to a server in 470 order to either get access to some data or perform some type of 471 transaction. In order for the client to mutually authenticate with 472 the server, the following steps are taken in an ABFAB architecture: 474 1. Client Configuration: The Client Application is configured with 475 an NAI assigned by the IdP. It is also configured with any 476 keys, certificates, passwords or other secret and public 477 information needed to run the EAP protocols between it and the 478 IdP. 480 2. Authentication mechanism selection: The GSS-EAP GSS-API 481 mechanism is selected for authentication/authorization. 483 3. Client provides an NAI to RP: The client application sets up a 484 transport to the RP and begins the GSS-EAP authentication. In 485 response, the RP sends an EAP request message (nested in the 486 GSS-EAP protocol) asking for the Client's name. The Client 487 sends an EAP response with an NAI name form that, at a minimum, 488 contains the realm portion of its full NAI. 490 4. Discovery of federated IdP: The RP uses pre-configured 491 information or a federation proxy to determine what IdP to use 492 based on policy and the realm portion of the provided Client 493 NAI. This is discussed in detail below (Section 2.1.2). 495 5. Request from Relying Party to IdP: Once the RP knows who the IdP 496 is, it (or its agent) will send a RADIUS/Diameter request to the 497 IdP. The RADIUS/Diameter access request encapsulates the EAP 498 response. At this stage, the RP will likely have no idea who 499 the client is. The RP sends its identity to the IdP in AAA 500 attributes, and it may send a SAML Attribute Request in a AAA 501 attribute. The AAA network checks that the identity claimed by 502 the RP is valid. 504 6. IdP begins EAP with the client: The IdP sends an EAP message to 505 the client with an EAP method to be used. The IdP SHOULD NOT 506 re-request the clients name in this message, but clients need to 507 be able to handle it. In this case the IdP MUST accept a realm 508 only in order to protect the client's name from the RP. The 509 available and appropriate methods are discussed below in this 510 memo (Section 2.2.1). 512 7. The EAP protocol is run: A bunch of EAP messages are passed 513 between the client (EAP peer) and the IdP (EAP server), until 514 the result of the authentication protocol is determined. The 515 number and content of those messages depends on the EAP method 516 selected. If the IdP is unable to authenticate the client, the 517 IdP sends an EAP failure message to the RP. As part of the EAP 518 protocol, the client sends a channel bindings EAP message to the 519 IdP (Section 2.2.2). In the channel binding message the client 520 identifies, among other things, the RP to which it is attempting 521 to authenticate. The IdP checks the channel binding data from 522 the client with that provided by the RP via the AAA protocol. 523 If the bindings do not match the IdP sends an EAP failure 524 message to the RP. 526 8. Successful EAP Authentication: At this point, the IdP (EAP 527 server) and client (EAP peer) have mutually authenticated each 528 other. As a result, the client and the IdP hold two 529 cryptographic keys: a Master Session Key (MSK), and an Extended 530 MSK (EMSK). At this point the client has a level of assurance 531 about the identity of the RP based on the name checking the IdP 532 has done using the RP naming information from the AAA framework 533 and from the client (by the channel binding data). 535 9. Local IdP Policy Check: At this stage, the IdP checks local 536 policy to determine whether the RP and client are authorized for 537 a given transaction/service, and if so, what if any, attributes 538 will be released to the RP. If the IdP gets a policy failure, 539 it sends an EAP failure message to the RP.[[Should this be an 540 EAP failure to the client as well?]] (The RP will have done its 541 policy checks during the discovery process.) 543 10. IdP provides the RP with the MSK: The IdP sends a positive 544 result EAP to the RP, along with an optional set of AAA 545 attributes associated with the client (usually as one or more 546 SAML assertions). In addition, the EAP MSK is returned to the 547 RP. 549 11. RP Processes Results: When the RP receives the result from the 550 IdP, it should have enough information to either grant or refuse 551 a resource access request. It may have information that 552 associates the client with specific authorization identities. 553 If additional attributes are needed from the IdP the RP may make 554 a new SAML Request to the IdP. It will apply these results in 555 an application-specific way. 557 12. RP returns results to client: Once the RP has a response it must 558 inform the client application of the result. If all has gone 559 well, all are authenticated, and the application proceeds with 560 appropriate authorization levels. The client can now complete 561 the authentication of the RP by the use of the EAP MSK value. 563 An example communication flow is given below: 565 Relying Client Identity 566 Party App Provider 568 | (1) | Client Configuration 569 | | | 570 |<-----(2)----->| | Mechanism Selection 571 | | | 572 |<-----(3)-----<| | NAI transmitted to RP 573 | | | 574 |<=====(4)====================>| Discovery 575 | | | 576 |>=====(5)====================>| Access request from RP to IdP 577 | | | 578 | |< - - (6) - -<| EAP method to Client 579 | | | 580 | |< - - (7) - ->| EAP Exchange to authenticate 581 | | | Client 582 | | | 583 | | (8 & 9) Local Policy Check 584 | | | 585 |<====(10)====================<| IdP Assertion to RP 586 | | | 587 (11) | | RP processes results 588 | | | 589 |>----(12)----->| | Results to client app. 591 ----- = Between Client App and RP 592 ===== = Between RP and IdP 593 - - - = Between Client App and IdP 595 1.5. Design Goals 597 Our key design goals are as follows: 599 o Each party of a transaction will be authenticated, although 600 perhaps not identified, and the client will be authorized for 601 access to a specific resource. 603 o Means of authentication is decoupled so as to allow for multiple 604 authentication methods. 606 o The architecture requires no sharing of long term private keys 607 between clients and servers. 609 o The system will scale to large numbers of identity providers, 610 relying parties, and users. 612 o The system will be designed primarily for non-Web-based 613 authentication. 615 o The system will build upon existing standards, components, and 616 operational practices. 618 Designing new three party authentication and authorization protocols 619 is hard and fraught with risk of cryptographic flaws. Achieving 620 widespread deployment is even more difficult. A lot of attention on 621 federated access has been devoted to the Web. This document instead 622 focuses on a non-Web-based environment and focuses on those protocols 623 where HTTP is not used. Despite the increased excitement for 624 layering every protocol on top of HTTP there are still a number of 625 protocols available that do not use HTTP-based transports. Many of 626 these protocols are lacking a native authentication and authorization 627 framework of the style shown in Figure 1. 629 2. Architecture 631 We have already introduced the federated access architecture, with 632 the illustration of the different actors that need to interact, but 633 did not expand on the specifics of providing support for non-Web 634 based applications. This section details this aspect and motivates 635 design decisions. The main theme of the work described in this 636 document is focused on re-using existing building blocks that have 637 been deployed already and to re-arrange them in a novel way. 639 Although this architecture assumes updates to the relying party, the 640 client application, and the Identity Provider, those changes are kept 641 at a minimum. A mechanism that can demonstrate deployment benefits 642 (based on ease of update of existing software, low implementation 643 effort, etc.) is preferred and there may be a need to specify 644 multiple mechanisms to support the range of different deployment 645 scenarios. 647 There are a number of ways for encapsulating EAP into an application 648 protocol. For ease of integration with a wide range of non-Web based 649 application protocols the usage of the GSS-API was chosen. A 650 description of the technical specification can be found in 651 [I-D.ietf-abfab-gss-eap]. 653 The architecture consists of several building blocks, which is shown 654 graphically in Figure 2. In the following sections, we discuss the 655 data flow between each of the entities, the protocols used for that 656 data flow and some of the trade-offs made in choosing the protocols. 658 +--------------+ 659 | Identity | 660 | Provider | 661 | (IdP) | 662 +-^----------^-+ 663 * EAP o RADIUS/ 664 * o Diameter 665 --v----------v-- 666 /// \\\ 667 // \\ 668 | Federation | 669 | Substrate | 670 \\ // 671 \\\ /// 672 --^----------^-- 673 * EAP o RADIUS/ 674 * o Diameter 675 +-------------+ +-v----------v--+ 676 | | | | 677 | Client | EAP/EAP Method | Relying Party | 678 | Application |<****************>| (RP) | 679 | | GSS-API | | 680 | |<---------------->| | 681 | | Application | | 682 | | Protocol | | 683 | |<================>| | 684 +-------------+ +---------------+ 686 Legend: 688 <****>: Client-to-IdP Exchange 689 <---->: Client-to-RP Exchange 690 : RP-to-IdP Exchange 691 <====>: Protocol through which GSS-API/GS2 exchanges are tunneled 693 Figure 2: ABFAB Protocol Instantiation 695 2.1. Relying Party to Identity Provider 697 Communications between the Relying Party and the Identity Provider is 698 done by the federation substrate. This communication channel is 699 responsible for: 701 o Establishing the trust relationship between the RP and the IdP. 703 o Determining the rules governing the relationship. 705 o Conveying authentication packets from the client to the IdP and 706 back. 708 o Providing the means of establishing a trust relationship between 709 the RP and the client. 711 o Providing a means for the RP to obtain attributes about the client 712 from the IdP. 714 The ABFAB working group has chosen the AAA framework for the messages 715 transported between the RP and IdP. The AAA framework supports the 716 requirements stated above as follows: 718 o The AAA backbone supplies the trust relationship between the RP 719 and the IdP. 721 o The agreements governing a specific AAA backbone contains the 722 rules governing the relationships within the AAA federation. 724 o A method exists for carrying EAP packets within RADIUS [RFC3579] 725 and Diameter [RFC4072]. 727 o The use of EAP channel binding [RFC6677] along with the core ABFAB 728 protocol provide the pieces necessary to establish the identities 729 of the RP and the client, while EAP provides the cryptographic 730 methods for the RP and the client to validate they are talking to 731 each other. 733 o A method exists for carrying SAML packets within RADIUS 734 [I-D.ietf-abfab-aaa-saml] and Diameter (work in progress) which 735 allows the RP to query attributes about the client from the IdP. 737 Future protocols that support the same framework but do different 738 routing may be used in the future. One such effort is to setup a 739 framework that creates a trusted point-to-point channel on the fly. 741 2.1.1. AAA, RADIUS and Diameter 743 Interestingly, for network access authentication the usage of the AAA 744 framework with RADIUS [RFC2865] and Diameter [RFC3588] was quite 745 successful from a deployment point of view. To map to the 746 terminology used in Figure 1 to the AAA framework the IdP corresponds 747 to the AAA server, the RP corresponds to the AAA client, and the 748 technical building blocks of a federation are AAA proxies, relays and 749 redirect agents (particularly if they are operated by third parties, 750 such as AAA brokers and clearing houses). The front-end, i.e. the 751 end host to AAA client communication, is in case of network access 752 authentication offered by link layer protocols that forward 753 authentication protocol exchanges back-and-forth. An example of a 754 large scale RADIUS-based federation is EDUROAM [2]. 756 By using the AAA framework, ABFAB gets a lot of mileage as many of 757 the federation agreements already exist and merely need to be 758 expanded to cover the ABFAB additions. The AAA framework has already 759 addressed some of the problems outlined above. For example, 761 o It already has a method for routing requests based on a domain. 763 o It already has an extensible architecture allowing for new 764 attributes to be defined and transported. 766 o Pre-existing relationships can be re-used. 768 The astute reader will notice that RADIUS and Diameter have 769 substantially similar characteristics. Why not pick one? RADIUS and 770 Diameter are deployed in different environments. RADIUS can often be 771 found in enterprise and university networks, and is also in use by 772 fixed network operators. Diameter, on the other hand, is deployed by 773 mobile operators. Another key difference is that today RADIUS is 774 largely transported upon UDP. We leave as a deployment decision, 775 which protocol will be appropriate. The protocol defines all the 776 necessary new AAA attributes as RADIUS attributes. A future document 777 would define the same AAA attributes for a Diameter environment. We 778 also note that there exist proxies which convert from RADIUS to 779 Diameter and back. This makes it possible for both to be deployed in 780 a single federation substrate. 782 Through the integrity protection mechanisms in the AAA framework, the 783 identity provider can establish technical trust that messages are 784 being sent by the appropriate relying party. Any given interaction 785 will be associated with one federation at the policy level. The 786 legal or business relationship defines what statements the identity 787 provider is trusted to make and how these statements are interpreted 788 by the relying party. The AAA framework also permits the relying 789 party or elements between the relying party and identity provider to 790 make statements about the relying party. 792 The AAA framework provides transport for attributes. Statements made 793 about the client by the identity provider, statements made about the 794 relying party and other information are transported as attributes. 796 One demand that the AAA substrate makes of the upper layers is that 797 they must properly identify the end points of the communication. It 798 must be possible for the AAA client at the RP to determine where to 799 send each RADIUS or Diameter message. Without this requirement, it 800 would be the RP's responsibility to determine the identity of the 801 client on its own, without the assistance of an IdP. This 802 architecture makes use of the Network Access Identifier (NAI), where 803 the IdP is indicated by the realm component [I-D.ietf-radext-nai]. 804 The NAI is represented and consumed by the GSS-API layer as 805 GSS_C_NT_USER_NAME as specified in [RFC2743]. The GSS-API EAP 806 mechanism includes the NAI in the EAP Response/Identity message. 808 2.1.2. Discovery and Rules Determination 810 While we are using the AAA protocols to communicate with the IdP, the 811 RP may have multiple federation substrates to select from. The RP 812 has a number of criteria that it will use in selecting which of the 813 different federations to use: 815 o The federation selected must be able to communicate with the IdP. 817 o The federation selected must match the business rules and 818 technical policies required for the RP security requirements. 820 The RP needs to discover which federation will be used to contact the 821 IdP. The first selection criteria used during discovery is going to 822 be the name of the IdP to be contacted. The second selection 823 criteria used during discovery is going to be the set of business 824 rules and technical policies governing the relationship; this is 825 called rules determination. The RP also needs to establish technical 826 trust in the communications with the IdP. 828 Rules determination covers a broad range of decisions about the 829 exchange. One of these is whether the given RP is permitted to talk 830 to the IdP using a given federation at all, so rules determination 831 encompasses the basic authorization decision. Other factors are 832 included, such as what policies govern release of information about 833 the client to the RP and what policies govern the RP's use of this 834 information. While rules determination is ultimately a business 835 function, it has significant impact on the technical exchanges. The 836 protocols need to communicate the result of authorization. When 837 multiple sets of rules are possible, the protocol must disambiguate 838 which set of rules are in play. Some rules have technical 839 enforcement mechanisms; for example in some federations 840 intermediaries validate information that is being communicated within 841 the federation. 843 At the time of writing no protocol mechanism has been specified to 844 allow a AAA client to determine whether a AAA proxy will indeed be 845 able to route AAA requests to a specific IdP. The AAA routing is 846 impacted by business rules and technical policies that may be quite 847 complex and at the present time, the route selection is based on 848 manual configuration. 850 2.1.3. Routing and Technical Trust 852 Several approaches to having messages routed through the federation 853 substrate are possible. These routing methods can most easily be 854 classified based on the mechanism for technical trust that is used. 855 The choice of technical trust mechanism constrains how rules 856 determination is implemented. Regardless of what deployment strategy 857 is chosen, it is important that the technical trust mechanism be able 858 to validate the identities of both parties to the exchange. The 859 trust mechanism must ensure that the entity acting as IdP for a given 860 NAI is permitted to be the IdP for that realm, and that any service 861 name claimed by the RP is permitted to be claimed by that entity. 862 Here are the categories of technical trust determination: 864 AAA Proxy: 865 The simplest model is that an RP is an AAA client and can send the 866 request directly to an AAA proxy. The hop-by-hop integrity 867 protection of the AAA fabric provides technical trust. An RP can 868 submit a request directly to the correct federation. 869 Alternatively, a federation disambiguation fabric can be used. 870 Such a fabric takes information about what federations the RP is 871 part of and what federations the IdP is part of and routes a 872 message to the appropriate federation. The routing of messages 873 across the fabric plus attributes added to requests and responses 874 provides rules determination. For example, when a disambiguation 875 fabric routes a message to a given federation, that federation's 876 rules are chosen. Name validation is enforced as messages travel 877 across the fabric. The entities near the RP confirm its identity 878 and validate names it claims. The fabric routes the message 879 towards the appropriate IdP, validating the name of the IdP in the 880 process. The routing can be statically configured. Alternatively 881 a routing protocol could be developed to exchange reachability 882 information about a given IdP and to apply policy across the AAA 883 fabric. Such a routing protocol could flood naming constraints to 884 the appropriate points in the fabric. 886 Trust Broker: 887 Instead of routing messages through AAA proxies, some trust broker 888 could establish keys between entities near the RP and entities 889 near the IdP. The advantage of this approach is efficiency of 890 message handling. Fewer entities are needed to be involved for 891 each message. Security may be improved by sending individual 892 messages over fewer hops. Rules determination involves decisions 893 made by trust brokers about what keys to grant. Also, associated 894 with each credential is context about rules and about other 895 aspects of technical trust including names that may be claimed. A 896 routing protocol similar to the one for AAA proxies is likely to 897 be useful to trust brokers in flooding rules and naming 898 constraints. 900 Global Credential: 902 A global credential such as a public key and certificate in a 903 public key infrastructure can be used to establish technical 904 trust. A directory or distributed database such as the Domain 905 Name System is used by the RP to discover the endpoint to contact 906 for a given NAI. Either the database or certificates can provide 907 a place to store information about rules determination and naming 908 constraints. Provided that no intermediates are required (or 909 appear to be required) and that the RP and IdP are sufficient to 910 enforce and determine rules, rules determination is reasonably 911 simple. However applying certain rules is likely to be quite 912 complex. For example if multiple sets of rules are possible 913 between an IdP and RP, confirming the correct set is used may be 914 difficult. This is particularly true if intermediates are 915 involved in making the decision. Also, to the extent that 916 directory information needs to be trusted, rules determination may 917 be more complex. 919 Real-world deployments are likely to be mixtures of these basic 920 approaches. For example, it will be quite common for an RP to route 921 traffic to a AAA proxy within an organization. That proxy could then 922 use any of the three methods to get closer to the IdP. It is also 923 likely that rather than being directly reachable, the IdP may have a 924 proxy on the edge of its organization. Federations will likely 925 provide a traditional AAA proxy interface even if they also provide 926 another mechanism for increased efficiency or security. 928 2.1.4. AAA Security 930 For the AAA framework there are two different places where security 931 needs to be examined. The first is the security that is in place for 932 the links in the AAA backbone being used. The second is the nodes 933 that the backbone consists of. 935 The default link security for RADIUS is showing its age as it uses 936 MD5 and a shared secret to both obfuscate passwords and to provide 937 integrity on the RADIUS messages. While some EAP methods have 938 designed in the ability to protect the client authentication 939 credentials, the MSK returned from the IDP to the RP is protected 940 only by the RADIUS security. In many environments this is considered 941 to be insufficient, especially as not all attributes are obfuscated 942 and can thus leak information to a passive eavesdropper. The use of 943 RADIUS with TLS [RFC6614] and/or DTLS [I-D.ietf-radext-dtls] 944 addresses these attacks. The same level of security is included in 945 the base Diameter specifications. 947 2.1.5. SAML Assertions 948 For the traditional use of AAA frameworks, network access, the only 949 requirement that was necessary to grant access was an affirmative 950 response from the IdP. In the ABFAB world, the RP may need to get 951 additional information about the client before granting access. 952 ABFAB therefore has a requirement that it can transport an arbitrary 953 set of attributes about the client from the IdP to the RP. 955 Security Assertions Markup Language (SAML) [OASIS.saml-core-2.0-os] 956 was designed in order to carry an extensible set of attributes about 957 a subject. Since SAML is extensible in the attribute space, ABFAB 958 has no immediate needs to update the core SAML specifications for our 959 work. It will be necessary to update IdPs that need to return SAML 960 assertions to RPs and for both the IdP and the RP to implement a new 961 SAML profile designed to carry SAML assertions in AAA. The new 962 profile can be found in RFCXXXX [I-D.ietf-abfab-aaa-saml]. As SAML 963 statements will frequently be large, RADIUS servers and clients that 964 deal with SAML statements will need to implement RFC XXXX 965 [I-D.perez-radext-radius-fragmentation] 967 There are several issues that need to be highlighted: 969 o The security of SAML assertions. 971 o Namespaces and mapping of SAML attributes. 973 o Subject naming of entities. 975 o Making multiple queries about the subject(s). 977 o Level of Assurance for authentication. 979 SAML assertions have an optional signature that can be used to 980 protect and provide origination of the assertion. These signatures 981 are normally based on asymmetric key operations and require that the 982 verifier be able to check not only the cryptographic operation, but 983 also the binding of the originators name and the public key. In a 984 federated environment it will not always be possible for the RP to 985 validate the binding, for this reason the technical trust established 986 in the federation is used as an alternate method of validating the 987 origination and integrity of the SAML Assertion. 989 Attributes placed in SAML assertions can have different namespaces 990 assigned to the same name. In many, but not all, cases the 991 federation agreements will determine what attributes can be used in a 992 SAML statement. This means that the RP needs to map from the 993 federation names, types and semantics into the ones that the policies 994 of the RP are written in. In other cases the federation substrate 995 may modify the SAML assertions in transit to do the necessary 996 namespace, naming and semantic mappings as the assertion crosses the 997 different boundaries in the federation. If the proxies are modifying 998 the SAML Assertion, then they will obviously remove any signatures as 999 they would no longer validate. In this case the technical trust is 1000 the required mechanism for validating the integrity of the assertion. 1001 Finally, the attributes may still be in the namespace of the 1002 originating IdP. When this occurs the RP will need to get the 1003 required mapping operations from the federation agreements and do the 1004 appropriate mappings itself. 1006 The RADIUS SAML RFC [I-D.ietf-abfab-aaa-saml] has defined a new SAML 1007 name format that corresponds to the NAI name form defined by RFC XXXX 1008 [I-D.ietf-radext-nai]. This allows for easy name matching in many 1009 cases as the name form in the SAML statement and the name form used 1010 in RADIUS or Diameter will be the same. In addition to the NAI name 1011 form, the document also defines a pair of implicit name forms 1012 corresponding to the Client and the Client's machine. These implicit 1013 name forms are based on the Identity-Type enumeration defined in TEAP 1014 [I-D.ietf-emu-eap-tunnel-method]. If the name form returned in a 1015 SAML statement is not based on the NAI, then it is a requirement on 1016 the EAP server that it validate that the subject of the SAML 1017 assertion, if any, is equivalent to the subject identified by the NAI 1018 used in the RADIUS or Diameter session. 1020 RADIUS has the ability to deal with multiple SAML queries for those 1021 EAP Servers which follow RFC 5080 [RFC5080]. In this case a State 1022 attribute will always be returned with the Access-Accept. The EAP 1023 client can then send a new Access-Request with the State attribute 1024 and the new SAML Request Multiple SAML queries can then be done by 1025 making a new Access-Request using the State attribute returned in the 1026 last Access-Accept to link together the different RADIUS sessions. 1028 Some RPs need to ensure that specific criteria are met during the 1029 authentication process. This need is met by using Levels of 1030 Assurance. The way a Level of Assurance is communicated to the RP 1031 from the EAP server is by the use of a SAML Authentication Request 1032 using the Authentication Profile from RFC XXX 1033 [I-D.ietf-abfab-aaa-saml] When crossing boundaries between different 1034 federations, either the policy specified will need to be shared 1035 between the two federations, the policy will need to be mapped by the 1036 proxy server on the boundary or the proxy server on the boundary will 1037 need to supply information the EAP server so that it can do the 1038 required mapping. If this mapping is not done, then the EAP server 1039 will not be able to enforce the desired Level of Assurance as it will 1040 not understand the policy requirements. 1042 2.2. Client To Identity Provider 1043 Looking at the communications between the client and the IdP, the 1044 following items need to be dealt with: 1046 o The client and the IdP need to mutually authenticate each other. 1048 o The client and the IdP need to mutually agree on the identity of 1049 the RP. 1051 ABFAB selected EAP for the purposes of mutual authentication and 1052 assisted in creating some new EAP channel binding documents for 1053 dealing with determining the identity of the RP. A framework for the 1054 channel binding mechanism has been defined in RFC 6677 [RFC6677] that 1055 allows the IdP to check the identity of the RP provided by the AAA 1056 framework with that provided by the client. 1058 2.2.1. Extensible Authentication Protocol (EAP) 1060 Traditional web federation does not describe how a client interacts 1061 with an identity provider for authentication. As a result, this 1062 communication is not standardized. There are several disadvantages 1063 to this approach. Since the communication is not standardized, it is 1064 difficult for machines to correctly enter their credentials with 1065 different authentications, where Individuals can correctly identify 1066 the entire mechanism on the fly. The use of browsers for 1067 authentication restricts the deployment of more secure forms of 1068 authentication beyond plaintext username and password known by the 1069 server. In a number of cases the authentication interface may be 1070 presented before the client has adequately validated they are talking 1071 to the intended server. By giving control of the authentication 1072 interface to a potential attacker, the security of the system may be 1073 reduced and phishing opportunities introduced. 1075 As a result, it is desirable to choose some standardized approach for 1076 communication between the client's end-host and the identity 1077 provider. There are a number of requirements this approach must 1078 meet. 1080 Experience has taught us one key security and scalability 1081 requirement: it is important that the relying party not get 1082 possession of the long-term secret of the client. Aside from a 1083 valuable secret being exposed, a synchronization problem can develop 1084 when the client changes keys with the IdP. 1086 Since there is no single authentication mechanism that will be used 1087 everywhere there is another associated requirement: The 1088 authentication framework must allow for the flexible integration of 1089 authentication mechanisms. For instance, some IdPs require hardware 1090 tokens while others use passwords. A service provider wants to 1091 provide support for both authentication methods, and other methods 1092 from IdPs not yet seen. 1094 These requirements can be met by utilizing standardized and 1095 successfully deployed technology, namely by the Extensible 1096 Authentication Protocol (EAP) framework [RFC3748]. Figure 2 1097 illustrates the integration graphically. 1099 EAP is an end-to-end framework; it provides for two-way communication 1100 between a peer (i.e. client or individual) through the EAP 1101 authenticator (i.e., relying party) to the back-end (i.e., identity 1102 provider). Conveniently, this is precisely the communication path 1103 that is needed for federated identity. Although EAP support is 1104 already integrated in AAA systems (see [RFC3579] and [RFC4072]) 1105 several challenges remain: 1107 o The first is how to carry EAP payloads from the end host to the 1108 relying party. 1110 o Another is to verify statements the relying party has made to the 1111 client, confirm these statements are consistent with statements 1112 made to the identity provider and confirm all of the above are 1113 consistent with the federation and any federation-specific policy 1114 or configuration. 1116 o Another challenge is choosing which identity provider to use for 1117 which service. 1119 The EAP method used for ABFAB needs to meet the following 1120 requirements: 1122 o It needs to provide mutual authentication of the client and IdP. 1124 o It needs to support channel binding. 1126 As of this writing, the only EAP method that meets these criteria is 1127 TEAP [I-D.ietf-emu-eap-tunnel-method] either alone (if client 1128 certificates are used) or with an inner EAP method that does mutual 1129 authentication. 1131 2.2.2. EAP Channel Binding 1133 EAP channel binding is easily confused with a facility in GSS-API 1134 also called channel binding. GSS-API channel binding provides 1135 protection against man-in-the-middle attacks when GSS-API is used as 1136 authentication inside some tunnel; it is similar to a facility called 1137 cryptographic binding in EAP. See [RFC5056] for a discussion of the 1138 differences between these two facilities. 1140 The client knows, in theory, the name of the RP that it attempted to 1141 connect to, however in the event that an attacker has intercepted the 1142 protocol, the client and the IdP need to be able to detect this 1143 situation. A general overview of the problem along with a 1144 recommended way to deal with the channel binding issues can be found 1145 in RFC 6677 [RFC6677]. 1147 Since that document was published, a number of possible attacks were 1148 found and methods to address these attacks have been outlined in 1149 [I-D.ietf-emu-crypto-bind]. 1151 2.3. Client to Relying Party 1153 The final set of interactions between the parties to consider are 1154 those between the client and the RP. In some ways this is the most 1155 complex set since at least part of it is outside the scope of the 1156 ABFAB work. The interactions between these parties include: 1158 o Running the protocol that implements the service that is provided 1159 by the RP and desired by the client. 1161 o Authenticating the client to the RP and the RP to the client. 1163 o Providing the necessary security services to the service protocol 1164 that it needs beyond authentication. 1166 o Deal with client re-authentication where desired. 1168 2.3.1. GSS-API 1170 One of the remaining layers is responsible for integration of 1171 federated authentication into the application. There are a number of 1172 approaches that applications have adopted for security. So, there 1173 may need to be multiple strategies for integration of federated 1174 authentication into applications. However, we have started with a 1175 strategy that provides integration to a large number of application 1176 protocols. 1178 Many applications such as SSH [RFC4462], NFS [RFC2203], DNS [RFC3645] 1179 and several non-IETF applications support the Generic Security 1180 Services Application Programming Interface [RFC2743]. Many 1181 applications such as IMAP, SMTP, XMPP and LDAP support the Simple 1182 Authentication and Security Layer (SASL) [RFC4422] framework. These 1183 two approaches work together nicely: by creating a GSS-API mechanism, 1184 SASL integration is also addressed. In effect, using a GSS-API 1185 mechanism with SASL simply requires placing some headers on the front 1186 of the mechanism and constraining certain GSS-API options. 1188 GSS-API is specified in terms of an abstract set of operations which 1189 can be mapped into a programming language to form an API. When 1190 people are first introduced to GSS-API, they focus on it as an API. 1191 However, from the prospective of authentication for non-web 1192 applications, GSS-API should be thought of as a protocol as well as 1193 an API. When looked at as a protocol, it consists of abstract 1194 operations such as the initial context exchange, which includes two 1195 sub-operations (gss_init_sec_context and gss_accept_sec_context). An 1196 application defines which abstract operations it is going to use and 1197 where messages produced by these operations fit into the application 1198 architecture. A GSS-API mechanism will define what actual protocol 1199 messages result from that abstract message for a given abstract 1200 operation. So, since this work is focusing on a particular GSS-API 1201 mechanism, we generally focus on protocol elements rather than the 1202 API view of GSS-API. 1204 The API view of GSS-API does have significant value as well, since 1205 the abstract operations are well defined, the set of information that 1206 a mechanism gets from the application is well defined. Also, the set 1207 of assumptions the application is permitted to make is generally well 1208 defined. As a result, an application protocol that supports GSS-API 1209 or SASL is very likely to be usable with a new approach to 1210 authentication including this one with no required modifications. In 1211 some cases, support for a new authentication mechanism has been added 1212 using plugin interfaces to applications without the application being 1213 modified at all. Even when modifications are required, they can 1214 often be limited to supporting a new naming and authorization model. 1215 For example, this work focuses on privacy; an application that 1216 assumes it will always obtain an identifier for the client will need 1217 to be modified to support anonymity, unlinkability or pseudonymity. 1219 So, we use GSS-API and SASL because a number of the application 1220 protocols we wish to federate support these strategies for security 1221 integration. What does this mean from a protocol standpoint and how 1222 does this relate to other layers? This means we need to design a 1223 concrete GSS-API mechanism. We have chosen to use a GSS-API 1224 mechanism that encapsulates EAP authentication. So, GSS-API (and 1225 SASL) encapsulates EAP between the end-host and the service. The AAA 1226 framework encapsulates EAP between the relying party and the identity 1227 provider. The GSS-API mechanism includes rules about how initiators 1228 and services are named as well as per-message security and other 1229 facilities required by the applications we wish to support. 1231 2.3.2. Protocol Transport 1233 The transport of data between the client and the relying party is not 1234 provided by GSS-API. GSS-API creates and consumes messages, but it 1235 does not provide the transport itself, instead the protocol using 1236 GSS-API needs to provide the transport. In many cases HTTP or HTTPS 1237 is used for this transport, but other transports are perfectly 1238 acceptable. The core GSS-API document [RFC2743] provides some 1239 details on what requirements exist. 1241 In addition we highlight the following: 1243 o The transport does not need to provide either privacy or 1244 integrity. After GSS-EAP has finished negotiation, GSS-API can be 1245 used to provide both services. If the negotiation process itself 1246 needs protection from eavesdroppers then the transport would need 1247 to provide the necessary services. 1249 o The transport needs to provide reliable transport of the messages. 1251 o The transport needs to ensure that tokens are delivered in order 1252 during the negotiation process. 1254 o GSS-API messages need to be delivered atomically. If the 1255 transport breaks up a message it must also reassemble the message 1256 before delivery. 1258 2.3.3. Reauthentication 1260 There are circumstances where the server will want to have the client 1261 reauthenticate itself. These include very long sessions, where the 1262 original authentication is time limited or cases where in order to 1263 complete an operation a different authentication is required. GSS- 1264 EAP does not have any mechanism for the server to initiate a 1265 reauthentication as all authentication operations start from the 1266 client. If a protocol using GSS-EAP needs to support 1267 reauthentication that is initiated by the server, then a request from 1268 the server to the client for the reauthentiction to start needs to be 1269 placed in the protocol. 1271 Clients can re-use the existing secure connection established by GSS- 1272 API to run the new authentication in by calling GSS_Init_sec_context. 1273 At this point a full reauthentication will be done. 1275 3. Application Security Services 1277 One of the key goals is to integrate federated authentication into 1278 existing application protocols and where possible, existing 1279 implementations of these protocols. Another goal is to perform this 1280 integration while meeting the best security practices of the 1281 technologies used to perform the integration. This section describes 1282 security services and properties required by the EAP GSS-API 1283 mechanism in order to meet these goals. This information could be 1284 viewed as specific to that mechanism. However, other future 1285 application integration strategies are very likely to need similar 1286 services. So, it is likely that these services will be expanded 1287 across application integration strategies if new application 1288 integration strategies are adopted. 1290 3.1. Authentication 1292 GSS-API provides an optional security service called mutual 1293 authentication. This service means that in addition to the initiator 1294 providing (potentially anonymous or pseudonymous) identity to the 1295 acceptor, the acceptor confirms its identity to the initiator. 1296 Especially for the ABFAB context, this service is confusingly named. 1297 We still say that mutual authentication is provided when the identity 1298 of an acceptor is strongly authenticated to an anonymous initiator. 1300 RFC 2743, unfortunately, does not explicitly talk about what mutual 1301 authentication means. Within this document we therefore define it 1302 as: 1304 o If a target name is configured for the initiator, then the 1305 initiator trusts that the supplied target name describes the 1306 acceptor. This implies both that appropriate cryptographic 1307 exchanges took place for the initiator to make such a trust 1308 decision, and that after evaluating the results of these 1309 exchanges, the initiator's policy trusts that the target name is 1310 accurate. 1312 o If no target name is configured for the initiator, then the 1313 initiator trusts that the acceptor name, supplied by the acceptor, 1314 correctly names the entity it is communicating with. 1316 o Both the initiator and acceptor have the same key material for 1317 per-message keys and both parties have confirmed they actually 1318 have the key material. In EAP terms, there is a protected 1319 indication of success. 1321 Mutual authentication is an important defense against certain aspects 1322 of phishing. Intuitively, clients would like to assume that if some 1323 party asks for their credentials as part of authentication, 1324 successfully gaining access to the resource means that they are 1325 talking to the expected party. Without mutual authentication, the 1326 server could "grant access" regardless of what credentials are 1327 supplied. Mutual authentication better matches this user intuition. 1329 It is important, therefore, that the GSS-EAP mechanism implement 1330 mutual authentication. That is, an initiator needs to be able to 1331 request mutual authentication. When mutual authentication is 1332 requested, only EAP methods capable of providing the necessary 1333 service can be used, and appropriate steps need to be taken to 1334 provide mutual authentication. While a broader set of EAP methods 1335 could be supported by not requiring mutual authentication, it was 1336 decided that the client needs to always have the ability to request 1337 it. In some cases the IdP and the RP will not support mutual 1338 authentication, however the client will always be able to detect this 1339 and make an appropriate security decision. 1341 The AAA infrastructure MAY hide the initiator's identity from the 1342 GSS-API acceptor, providing anonymity between the initiator and the 1343 acceptor. At this time, whether the identity is disclosed is 1344 determined by EAP server policy rather than by an indication from the 1345 initiator. Also, initiators are unlikely to be able to determine 1346 whether anonymous communication will be provided. For this reason, 1347 initiators are unlikely to set the anonymous return flag from 1348 GSS_Init_Sec_context. 1350 3.2. GSS-API Channel Binding 1352 [RFC5056] defines a concept of channel binding which is used prevent 1353 man-in-the-middle attacks. The channel binding works by taking a 1354 cryptographic value from the transport security and checks that both 1355 sides of the GSS-API conversation know this value. Transport Layer 1356 Security (TLS) is the most common transport security layer used for 1357 this purpose. 1359 It needs to be stressed that RFC 5056 channel binding (also called 1360 GSS-API channel binding when GSS-API is involved) is not the same 1361 thing as EAP channel binding. GSS-API channel binding is used for 1362 detecting Man-In-The-Middle attacks. EAP channel binding is used for 1363 mutual authentication and acceptor naming checks. Details are 1364 discussed in the mechanisms specification [I-D.ietf-abfab-gss-eap]. 1365 A fuller description of the differences between the facilities can be 1366 found in RFC 5056 [RFC5056]. 1368 The use of TLS can provide both encryption and integrity on the 1369 channel. It is common to provide SASL and GSS-API with these other 1370 security services. 1372 One of the benefits that the use of TLS provides, is that client has 1373 the ability to validate the name of the server. However this 1374 validation is predicated on a couple of things. The TLS sessions 1375 needs to be using certificates and not be an anonymous session. The 1376 client and the TLS server need to share a common trust point for the 1377 certificate used in validating the server. TLS provides its own 1378 server authentication. However there are a variety of situations 1379 where this authentication is not checked for policy or usability 1380 reasons. When the TLS authentication is checked, if the trust 1381 infrastructure behind the TLS authentication is different from the 1382 trust infrastructure behind the GSS-API mutual authentication then 1383 confirming the end-points using both trust infrastructures is likely 1384 to enhance security. If the endpoints of the GSS-API authentication 1385 are different than the endpoints of the lower layer, this is a strong 1386 indication of a problem such as a man-in-the-middle attack. Channel 1387 binding provides a facility to determine whether these endpoints are 1388 the same. 1390 The GSS-EAP mechanism needs to support channel binding. When an 1391 application provides channel binding data, the mechanism needs to 1392 confirm this is the same on both sides consistent with the GSS-API 1393 specification. 1395 3.3. Host-Based Service Names 1397 IETF security mechanisms typically take a host name and perhaps a 1398 service, entered by a user, and make some trust decision about 1399 whether the remote party in the interaction is the intended party. 1400 This decision can be made by the use of certificates, pre-configured 1401 key information or a previous leap of trust. GSS-API has defined a 1402 relatively flexible name convention, however most of the IETF 1403 applications that use GSS-API (including SSH, NFS, IMAP, LDAP and 1404 XMPP) have chosen to use a more restricted naming convention based on 1405 the host name. The GSS-EAP mechanism needs to support host-based 1406 service names in order to work with existing IETF protocols. 1408 The use of host-based service names leads to a challenging trust 1409 delegation problem. Who is allowed to decide whether a particular 1410 host name maps to a specific entity? Possible solutions to this 1411 problem have been looked at. 1413 o The public-key infrastructure (PKI) used by the web has chosen to 1414 have a number of trust anchors (root certificate authorities) each 1415 of which can map any host name to a public key. 1417 o A number of GSS-API mechanisms, such as Kerberos [RFC1964], have 1418 split the problem into two parts. A new concept called a realm is 1419 introduced, the realm is responsible for host mapping within that 1420 realm. The mechanism then decides what realm is responsible for a 1421 given name. This is the approach adopted by ABFAB. 1423 GSS-EAP defines a host naming convention that takes into account the 1424 host name, the realm, the service and the service parameters. An 1425 example of GSS-API service name is "xmpp/foo@example.com". This 1426 identifies the XMPP service on the host foo in the realm example.com. 1427 Any of the components, except for the service name may be omitted 1428 from a name. When omitted, then a local default would be used for 1429 that component of the name. 1431 While there is no requirement that realm names map to Fully Qualified 1432 Domain Names (FQDN) within DNS, in practice this is normally true. 1433 Doing so allows for the realm portion of service names and the 1434 portion of NAIs to be the same. It also allows for the use of DNS in 1435 locating the host of a service while establishing the transport 1436 channel between the client and the relying party. 1438 It is the responsibility of the application to determine the server 1439 that it is going to communicate with; GSS-API has the ability to help 1440 confirm that the server is the desired server but not to determine 1441 the name of the server to use. It is also the responsibility of the 1442 application to determine how much of the information identifying the 1443 service needs to be validated by the ABFAB system. The information 1444 that needs to be validated is used to build up the service name 1445 passed into the GSS-EAP mechanism. What information is to be 1446 validated will depend on both what information was provided by the 1447 client, and what information is considered significant. If the 1448 client only cares about getting a specific service, then the host and 1449 realm that provides the service does not need to be validated. 1451 Applications may retrieve information about providers of services 1452 from DNS. Service Records (SRV) and Naming Authority Pointer (NAPTR) 1453 records are used to help find a host that provides a service; however 1454 the necessity of having DNSSEC on the queries depends on how the 1455 information is going to be used. If the host name returned is not 1456 going to be validated by EAP channel binding, because only the 1457 service is being validated, then DNSSEC is not required. However, if 1458 the host name is going to be validated by EAP channel binding then 1459 DNSSEC needs to be use to ensure that the correct host name is 1460 validated. In general, if the information that is returned from the 1461 DNS query is to be validated, then it needs to be obtained in a 1462 secure manner. 1464 Another issue that needs to be addressed for host-based service names 1465 is that they do not work ideally when different instances of a 1466 service are running on different ports. If the services are 1467 equivalent, then it does not matter. However if there are 1468 substantial differences in the quality of the service that 1469 information needs to be part of the validation process. If one has 1470 just a host name and not a port in the information being validated, 1471 then this is not going to be a successful strategy. 1473 3.4. Additional GSS-API Services 1475 GSS-API provides per-message security services that can provide 1476 confidentiality and/or integrity. Some IETF protocols such as NFS 1477 and SSH take advantage of these services. As a result GSS-EAP needs 1478 to support these services. As with mutual authentication, per- 1479 message services will limit the set of EAP methods that can be used 1480 to those that generate a Master Session Key (MSK). Any EAP method 1481 that produces an MSK is able to support per-message security services 1482 described in [RFC2743]. 1484 GSS-API provides a pseudo-random function. This function generates a 1485 pseudo-random sequence using the shared session key as the seed for 1486 the bytes generated. This provides an algorithm that both the 1487 initiator and acceptor can run in order to arrive at the same key 1488 value. The use of this feature allows for an application to generate 1489 keys or other shared secrets for use in other places in the protocol. 1490 In this regards, it is similar in concept to the TLS extractor (RFC 1491 5705 [RFC5705].). While no current IETF protocols require this, non- 1492 IETF protocols are expected to take advantage of this in the near 1493 future. Additionally, a number of protocols have found the TLS 1494 extractor to be useful in this regards so it is highly probable that 1495 IETF protocols may also start using this feature. 1497 4. Privacy Considerations 1499 ABFAB, as an architecture designed to enable federated authentication 1500 and allow for the secure transmission of identity information between 1501 entities, obviously requires careful consideration around privacy and 1502 the potential for privacy violations. 1504 This section examines the privacy related information presented in 1505 this document, summarizing the entities that are involved in ABFAB 1506 communications and what exposure they have to identity information. 1507 In discussing these privacy considerations in this section, we use 1508 terminology and ideas from [I-D.iab-privacy-considerations]. 1510 Note that the ABFAB architecture uses at its core several existing 1511 technologies and protocols; detailed privacy discussion around these 1512 is not examined. This section instead focuses on privacy 1513 considerations specifically related to overall architecture and usage 1514 of ABFAB. 1516 +--------+ +---------------+ +--------------+ 1517 | Client | <---> | RP | <---> | AAA Client | 1518 +--------+ +---------------+ +--------------+ 1519 ^ 1520 | 1521 v 1522 +---------------+ +--------------+ 1523 | SAML Server | | AAA Proxy(s) | 1524 +---------------+ +--------------+ 1525 ^ ^ 1526 | | 1527 v v 1528 +------------+ +---------------+ +--------------+ 1529 | EAP Server | <---> | IdP | <---> | AAA Server | 1530 +------------+ +---------------+ +--------------+ 1532 Figure 3: Entities and Data Flow 1534 4.1. Entities and their roles 1536 Categorizing the ABFAB entities shown in the Figure 3 according to 1537 the taxonomy of terms from [I-D.iab-privacy-considerations] the 1538 entities shown in Figure 3 is somewhat complicated as during the 1539 various phases of ABFAB communications the roles of each entity 1540 changes. The three main phases of relevance are the Client to RP 1541 communication phase, the Client to IdP (via the Federation Substrate) 1542 phase, and the IdP to RP (via the Federation Substrate) phase. 1544 In the Client to RP communication phase, we have: 1546 Initiator: Client. 1548 Observers: Client, RP. 1550 Recipient: RP. 1552 In the Client to IdP (via the Federation Substrate) communication 1553 phase, we have: 1555 Initiator: Client. 1557 Observers: Client, RP, AAA Client, AAA Proxy(s), AAA Server, IdP. 1559 Recipient: IdP 1561 In the IdP to Relying party (via the Federation Substrate) 1562 communication phase, we have: 1564 Initiator: RP. 1566 Observers: IdP, AAA Server, AAA Proxy(s), AAA Client, RP. 1568 Recipient: IdP 1570 Eavesdroppers and Attackers can reside on any communication link 1571 between entities in Figure 3. 1573 The Federation Substrate consists of all of the AAA entities. In 1574 some cases the AAA Proxies entities may not exist as the AAA Client 1575 can talk directly to the AAA Server. Specifications such as the 1576 Trust Router Protocol and RADIUS dynamic discovery 1577 [I-D.ietf-radext-dynamic-discovery] can be used to shorten the path 1578 between the AAA client and the AAA server (and thus stop these AAA 1579 Proxies from being Observers); however even in these circumstances 1580 there may be AAA Proxies in the path. 1582 In Figure 3 the IdP has been divided into multiple logical pieces, in 1583 actual implementations these pieces will frequently be tightly 1584 coupled. The links between these pieces provide the greatest 1585 opportunity for attackers and eavesdroppers to acquire information, 1586 however, as they are all under the control of a single entity they 1587 are also the easiest to have tightly secured. 1589 4.2. Privacy Aspects of ABFAB Communication Flows 1591 In the ABFAB architecture, there are a few different types of data 1592 and identifiers in use. The best way to understand them, and the 1593 potential privacy impacts of them, is to look at each phase of 1594 communication in ABFAB. 1596 4.2.1. Client to RP 1598 The flow of data between the client and the RP is divided into two 1599 parts. The first part consists of all of the data exchanged as part 1600 of the ABFAB authentication process. The second part consists of all 1601 of the data exchanged after the authentication process has been 1602 finished. 1604 During the initial communications phase, the client sends an NAI (see 1605 [I-D.ietf-radext-nai]) to the RP. Many EAP methods (but not all) 1606 allow for the client to disclose an NAI to RP the in a form that 1607 includes only a realm component during this communications phase. 1608 This is the minimum amount of identity information necessary for 1609 ABFAB to work - it indicates an IdP that the principal has a 1610 relationship with. EAP methods that do not allow this will 1611 necessarily also reveal an identifier for the principal in the IdP 1612 realm (e.g. a username). 1614 The data shared during the initial communication phase may be 1615 protected by a channel protocol such as TLS. This will prevent the 1616 leak of information to passive eavesdroppers, however an active 1617 attacker may still be able to setup as a man-in-the-middle. The 1618 client may not be able to validate the certificates (if any) provided 1619 by the service, deferring the check of the identity of the RP until 1620 the completion of the ABFAB authentication protocol (i.e., using EAP 1621 channel binding). 1623 The data exchanged after the authentication process can have privacy 1624 and authentication using the GSS-API services. If the overall 1625 application protocol allows for the process of re-authentication, 1626 then the same privacy implications as discussed in previous 1627 paragraphs apply. 1629 4.2.2. Client to IdP (via Federation Substrate) 1631 This phase sees a secure TLS tunnel initiated between the Client and 1632 the IdP via the RP and federation substrate. The process is 1633 initiated by the RP using the realm information given to it by the 1634 client. Once set up, the tunnel is used to send credentials to IdP 1635 to authenticate. 1637 Various operational information is transported between RP and IdP, 1638 over the AAA infrastructure, for example using RADIUS headers. As no 1639 end-to-end security is provided by AAA, all AAA entities on the path 1640 between the RP and IdP have the ability to eavesdrop on this 1641 information unless additional security measures are taken (such as 1642 the use of TLS for RADIUS [I-D.ietf-radext-dtls]). Some of this 1643 information may form identifiers or explicit identity information: 1645 o The Relying Party knows the IP address of the Client. It is 1646 possible that the Relying Party could choose to expose this IP 1647 address by including it in a RADIUS header such as Calling Station 1648 ID. This is a privacy consideration to take into account of the 1649 application protocol. 1651 o The EAP MSK is transported between the IdP and the RP over the AAA 1652 infrastructure, for example through RADIUS headers. This is a 1653 particularly important privacy consideration, as any AAA Proxy 1654 that has access to the EAP MSK is able to decrypt and eavesdrop on 1655 any traffic encrypted using that EAP MSK (i.e. all communications 1656 between the Client and IdP). 1658 o Related to the above, the AAA server has access to the material 1659 necessary to derive the session key, thus the AAA server can 1660 observe any traffic encrypted between the Client and RP. This 1661 "feature" was" chosen as a simplification and to make performance 1662 faster; if it was decided that this trade-off was not desirable 1663 for privacy and security reasons, then extensions to ABFAB that 1664 make use of techniques such as Diffie-Helman key exchange would 1665 mitigate against this. 1667 The choice of EAP method used has other potential privacy 1668 implications. For example, if the EAP method in use does not support 1669 trust anchors to enable mutual authentication, then there are no 1670 guarantees that the IdP is who it claims to be, and thus the full NAI 1671 including a username and a realm might be sent to any entity 1672 masquerading as a particular IdP. 1674 Note that ABFAB has not specified any AAA accounting requirements. 1675 Implementations that use the accounting portion of AAA should 1676 consider privacy appropriately when designing this aspect. 1678 4.2.3. IdP to RP (via Federation Substrate) 1680 In this phase, the IdP communicates with the RP informing it as to 1681 the success or failure of authentication of the user, and optionally, 1682 the sending of identity information about the principal. 1684 As in the previous flow (Client to IdP), various operation 1685 information is transported between IdP and RP over the AAA 1686 infrastructure, and the same privacy considerations apply. However, 1687 in this flow, explicit identity information about the authenticated 1688 principal can be sent from the IdP to the RP. This information can 1689 be sent through RADIUS headers, or using SAML 1690 [I-D.ietf-abfab-aaa-saml]. This can include protocol specific 1691 identifiers, such as SAML NameIDs, as well as arbitrary attribute 1692 information about the principal. What information will be released 1693 is controlled by policy on the Identity Provider. As before, when 1694 sending this through RADIUS headers, all AAA entities on the path 1695 between the RP and IdP have the ability to eavesdrop unless 1696 additional security measures are taken (such as the use of TLS for 1697 RADIUS [I-D.ietf-radext-dtls]). When sending this using SAML, as 1698 specified in [I-D.ietf-abfab-aaa-saml], confidentiality of the 1699 information should however be guaranteed as [I-D.ietf-abfab-aaa-saml] 1700 requires the use of TLS for RADIUS. 1702 4.3. Relationship between User and Entities 1704 o Between User and IdP - the IdP is an entity the user will have a 1705 direct relationship with, created when the organization that 1706 operates the entity provisioned and exchanged the user's 1707 credentials. Privacy and data protection guarantees may form a 1708 part of this relationship. 1710 o Between User and RP - the RP is an entity the user may or may not 1711 have a direct relationship with, depending on the service in 1712 question. Some services may only be offered to those users where 1713 such a direct relationship exists (for particularly sensitive 1714 services, for example), while some may not require this and would 1715 instead be satisfied with basic federation trust guarantees 1716 between themselves and the IdP). This may well include the option 1717 that the user stays anonymous with respect to the RP (though 1718 obviously never to the IdP). If attempting to preserve privacy 1719 through the mitigation of data minimization, then the only 1720 attribute information about individuals exposed to the RP should 1721 be that which is strictly necessary for the operation of the 1722 service. 1724 o Between User and Federation substrate - the user is highly likely 1725 to have no knowledge of, or relationship with, any entities 1726 involved with the federation substrate (not that the IdP and/or RP 1727 may, however). Knowledge of attribute information about 1728 individuals for these entities is not necessary, and thus such 1729 information should be protected in such a way as to prevent access 1730 to this information from being possible. 1732 4.4. Accounting Information 1734 Alongside the core authentication and authorization that occurs in 1735 AAA communications, accounting information about resource consumption 1736 may be delivered as part of the accounting exchange during the 1737 lifetime of the granted application session. 1739 4.5. Collection and retention of data and identifiers 1741 In cases where Relying Parties are not required to identify a 1742 particular individual when an individual wishes to make use of their 1743 service, the ABFAB architecture enables anonymous or pseudonymous 1744 access. Thus data and identifiers other than pseudonyms and 1745 unlinkable attribute information need not be stored and retained. 1747 However, in cases where Relying Parties require the ability to 1748 identify a particular individual (e.g. so they can link this identity 1749 information to a particular account in their service, or where 1750 identity information is required for audit purposes), the service 1751 will need to collect and store such information, and to retain it for 1752 as long as they require. Deprovisioning of such accounts and 1753 information is out of scope for ABFAB, but obviously for privacy 1754 protection any identifiers collected should be deleted when they are 1755 no longer needed. 1757 4.6. User Participation 1759 In the ABFAB architecture, by its very nature users are active 1760 participants in the sharing of their identifiers as they initiate the 1761 communications exchange every time they wish to access a server. 1762 They are, however, not involved in control of the set of information 1763 related to them that transmitted from the IdP to RP for authorization 1764 purposes; rather, this is under the control of policy on the IdP. 1765 Due to the nature of the AAA communication flows, with the current 1766 ABFAB architecture there is no place for a process of gaining user 1767 consent for the information to be released from IdP to RP. 1769 5. Security Considerations 1771 This document describes the architecture for Application Bridging for 1772 Federated Access Beyond Web (ABFAB) and security is therefore the 1773 main focus. This section highlights the main communication channels 1774 and their security properties: 1776 Client-to-RP Channel: 1778 The channel binding material is provided by any certificates and 1779 the final message (i.e., a cryptographic token for the channel). 1780 Authentication may be provided by the RP to the client but a 1781 deployment without authentication at the TLS layer is possible as 1782 well. In addition, there is a channel between the GSS requestor 1783 and the GSS acceptor, but the keying material is provided by a 1784 "third party" to both entities. The client can derive keying 1785 material locally, but the RP gets the material from the IdP. In 1786 the absence of a transport that provides encryption and/or 1787 integrity, the channel between the client and the RP has no 1788 ability to have any cryptographic protection until the EAP 1789 authentication has been completed and the MSK is transferred from 1790 the IdP to the RP. 1792 RP-to-IdP Channel: 1794 The security of this communication channel is mainly provided by 1795 the functionality offered via RADIUS and Diameter. At the time of 1796 writing there are no end-to-end security mechanisms standardized 1797 and thereby the architecture has to rely on hop-by-hop security 1798 with trusted AAA entities or, as an alternative but possible 1799 deployment variant, direct communication between the AAA client to 1800 the AAA server. Note that the authorization result the IdP 1801 provides to the RP in the form of a SAML assertion may; however, 1802 be protected such that the SAML related components are secured 1803 end-to-end. 1805 The MSK is transported from the IdP to the RP over this channel. 1806 As no end-to-end security is provided by AAA, all AAA entities on 1807 the path between the RP and IdP have the ability to eavesdrop if 1808 no additional security measures are taken. One such measure is to 1809 use a transport between the client and the IdP that provides 1810 confidentiality. 1812 Client-to-IdP Channel: 1814 This communication interaction is accomplished with the help of 1815 EAP and EAP methods. The offered security protection will depend 1816 on the EAP method that is chosen but a minimum requirement is to 1817 offer mutual authentication, and key derivation. The IdP is 1818 responsible during this process to determine that the RP that is 1819 communication to the client over the RP-to-IdP channel is the same 1820 one talking to the IdP. This is accomplished via the EAP channel 1821 binding. 1823 Partial list of issues to be addressed in this section: Privacy, 1824 SAML, Trust Anchors, EAP Algorithm Selection, Diameter/RADIUS/AAA 1825 Issues, Naming of Entities, Protection of passwords, Channel Binding, 1826 End-point-connections (TLS), Proxy problems 1828 When a pseudonym is generated as a unique long term identifier for a 1829 client by an IdP, care MUST be taken in the algorithm that it cannot 1830 easily be reverse engineered by the service provider. If it can be 1831 reversed then the service provider can consult an oracle to determine 1832 if a given unique long term identifier is associated with a different 1833 known identifier. 1835 6. IANA Considerations 1837 This document does not require actions by IANA. 1839 7. Acknowledgments 1841 We would like to thank Mayutan Arumaithurai, Klaas Wierenga and Rhys 1842 Smith for their feedback. Additionally, we would like to thank Eve 1843 Maler, Nicolas Williams, Bob Morgan, Scott Cantor, Jim Fenton, Paul 1844 Leach, and Luke Howard for their feedback on the federation 1845 terminology question. 1847 Furthermore, we would like to thank Klaas Wierenga for his review of 1848 the pre-00 draft version. 1850 8. References 1852 8.1. Normative References 1854 [RFC2743] Linn, J., "Generic Security Service Application Program 1855 Interface Version 2, Update 1", RFC 2743, January 2000. 1857 [RFC2865] Rigney, C., Willens, S., Rubens, A., and W. Simpson, 1858 "Remote Authentication Dial In User Service (RADIUS)", RFC 1859 2865, June 2000. 1861 [RFC3588] Calhoun, P., Loughney, J., Guttman, E., Zorn, G., and J. 1862 Arkko, "Diameter Base Protocol", RFC 3588, September 2003. 1864 [RFC3748] Aboba, B., Blunk, L., Vollbrecht, J., Carlson, J., and H. 1865 Levkowetz, "Extensible Authentication Protocol (EAP)", RFC 1866 3748, June 2004. 1868 [RFC3579] Aboba, B. and P. Calhoun, "RADIUS (Remote Authentication 1869 Dial In User Service) Support For Extensible 1870 Authentication Protocol (EAP)", RFC 3579, September 2003. 1872 [RFC4072] Eronen, P., Hiller, T., and G. Zorn, "Diameter Extensible 1873 Authentication Protocol (EAP) Application", RFC 4072, 1874 August 2005. 1876 [I-D.ietf-abfab-gss-eap] 1877 Hartman, S. and J. Howlett, "A GSS-API Mechanism for the 1878 Extensible Authentication Protocol", draft-ietf-abfab-gss- 1879 eap-09 (work in progress), August 2012. 1881 [I-D.ietf-abfab-aaa-saml] 1882 Howlett, J. and S. Hartman, "A RADIUS Attribute, Binding, 1883 Profiles, Name Identifier Format, and Confirmation Methods 1884 for SAML", draft-ietf-abfab-aaa-saml-05 (work in 1885 progress), February 2013. 1887 [I-D.ietf-radext-nai] 1888 DeKok, A., "The Network Access Identifier", draft-ietf- 1889 radext-nai-02 (work in progress), January 2013. 1891 [RFC6677] Hartman, S., Clancy, T., and K. Hoeper, "Channel-Binding 1892 Support for Extensible Authentication Protocol (EAP) 1893 Methods", RFC 6677, July 2012. 1895 8.2. Informative References 1897 [RFC2903] de Laat, C., Gross, G., Gommans, L., Vollbrecht, J., and 1898 D. Spence, "Generic AAA Architecture", RFC 2903, August 1899 2000. 1901 [I-D.nir-tls-eap] 1902 Nir, Y., Sheffer, Y., Tschofenig, H., and P. Gutmann, "A 1903 Flexible Authentication Framework for the Transport Layer 1904 Security (TLS) Protocol using the Extensible 1905 Authentication Protocol (EAP)", draft-nir-tls-eap-13 (work 1906 in progress), December 2011. 1908 [I-D.ietf-oauth-v2] 1909 Hardt, D., "The OAuth 2.0 Authorization Framework", draft- 1910 ietf-oauth-v2-31 (work in progress), August 2012. 1912 [I-D.iab-privacy-considerations] 1913 Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 1914 Morris, J., Hansen, M., and R. Smith, "Privacy 1915 Considerations for Internet Protocols", draft-iab-privacy- 1916 considerations-03 (work in progress), July 2012. 1918 [I-D.perez-radext-radius-fragmentation] 1919 Perez-Mendez, A., Lopez, R., Pereniguez-Garcia, F., Lopez- 1920 Millan, G., Lopez, D., and A. DeKok, "Support of 1921 fragmentation of RADIUS packets", draft-perez-radext- 1922 radius-fragmentation-05 (work in progress), February 2013. 1924 [RFC4017] Stanley, D., Walker, J., and B. Aboba, "Extensible 1925 Authentication Protocol (EAP) Method Requirements for 1926 Wireless LANs", RFC 4017, March 2005. 1928 [RFC5106] Tschofenig, H., Kroeselberg, D., Pashalidis, A., Ohba, Y., 1929 and F. Bersani, "The Extensible Authentication Protocol- 1930 Internet Key Exchange Protocol version 2 (EAP-IKEv2) 1931 Method", RFC 5106, February 2008. 1933 [RFC1964] Linn, J., "The Kerberos Version 5 GSS-API Mechanism", RFC 1934 1964, June 1996. 1936 [RFC2203] Eisler, M., Chiu, A., and L. Ling, "RPCSEC_GSS Protocol 1937 Specification", RFC 2203, September 1997. 1939 [RFC3645] Kwan, S., Garg, P., Gilroy, J., Esibov, L., Westhead, J., 1940 and R. Hall, "Generic Security Service Algorithm for 1941 Secret Key Transaction Authentication for DNS (GSS-TSIG)", 1942 RFC 3645, October 2003. 1944 [RFC2138] Rigney, C., Rigney, C., Rubens, A., Simpson, W., and S. 1945 Willens, "Remote Authentication Dial In User Service 1946 (RADIUS)", RFC 2138, April 1997. 1948 [RFC4462] Hutzelman, J., Salowey, J., Galbraith, J., and V. Welch, 1949 "Generic Security Service Application Program Interface 1950 (GSS-API) Authentication and Key Exchange for the Secure 1951 Shell (SSH) Protocol", RFC 4462, May 2006. 1953 [RFC4422] Melnikov, A. and K. Zeilenga, "Simple Authentication and 1954 Security Layer (SASL)", RFC 4422, June 2006. 1956 [RFC5056] Williams, N., "On the Use of Channel Bindings to Secure 1957 Channels", RFC 5056, November 2007. 1959 [RFC5080] Nelson, D. and A. DeKok, "Common Remote Authentication 1960 Dial In User Service (RADIUS) Implementation Issues and 1961 Suggested Fixes", RFC 5080, December 2007. 1963 [RFC5705] Rescorla, E., "Keying Material Exporters for Transport 1964 Layer Security (TLS)", RFC 5705, March 2010. 1966 [RFC5801] Josefsson, S. and N. Williams, "Using Generic Security 1967 Service Application Program Interface (GSS-API) Mechanisms 1968 in Simple Authentication and Security Layer (SASL): The 1969 GS2 Mechanism Family", RFC 5801, July 2010. 1971 [RFC5849] Hammer-Lahav, E., "The OAuth 1.0 Protocol", RFC 5849, 1972 April 2010. 1974 [RFC6614] Winter, S., McCauley, M., Venaas, S., and K. Wierenga, 1975 "Transport Layer Security (TLS) Encryption for RADIUS", 1976 RFC 6614, May 2012. 1978 [OASIS.saml-core-2.0-os] 1979 Cantor, S., Kemp, J., Philpott, R., and E. Maler, 1980 "Assertions and Protocol for the OASIS Security Assertion 1981 Markup Language (SAML) V2.0", OASIS Standard saml- 1982 core-2.0-os, March 2005. 1984 [RFC2904] Vollbrecht, J., Calhoun, P., Farrell, S., Gommans, L., 1985 Gross, G., de Bruijn, B., de Laat, C., Holdrege, M., and 1986 D. Spence, "AAA Authorization Framework", RFC 2904, August 1987 2000. 1989 [I-D.ietf-emu-crypto-bind] 1990 Hartman, S., Wasserman, M., and D. Zhang, "EAP Mutual 1991 Cryptographic Binding", draft-ietf-emu-crypto-bind-03 1992 (work in progress), March 2013. 1994 [I-D.ietf-emu-eap-tunnel-method] 1995 Zhou, H., Cam-Winget, N., Salowey, J., and S. Hanna, 1996 "Tunnel EAP Method (TEAP) Version 1", draft-ietf-emu-eap- 1997 tunnel-method-05 (work in progress), February 2013. 1999 [I-D.ietf-radext-dtls] 2000 DeKok, A., "DTLS as a Transport Layer for RADIUS", draft- 2001 ietf-radext-dtls-03 (work in progress), January 2013. 2003 [I-D.ietf-radext-dynamic-discovery] 2004 Winter, S. and M. McCauley, "NAI-based Dynamic Peer 2005 Discovery for RADIUS/TLS and RADIUS/DTLS", draft-ietf- 2006 radext-dynamic-discovery-06 (work in progress), February 2007 2013. 2009 [WS-TRUST] 2010 Lawrence, K., Kaler, C., Nadalin, A., Goodner, M., Gudgin, 2011 M., Barbir, A., and H. Granqvist, "WS-Trust 1.4", OASIS 2012 Standard ws-trust-200902, February 2009, . 2015 [NIST-SP.800-63] 2016 Burr, W., Dodson, D., and W. Polk, "Electronic 2017 Authentication Guideline", NIST Special Publication 2018 800-63, April 2006. 2020 Authors' Addresses 2022 Josh Howlett 2023 JANET(UK) 2024 Lumen House, Library Avenue, Harwell 2025 Oxford OX11 0SG 2026 UK 2028 Phone: +44 1235 822363 2029 Email: Josh.Howlett@ja.net 2031 Sam Hartman 2032 Painless Security 2034 Email: hartmans-ietf@mit.edu 2035 Hannes Tschofenig 2036 Nokia Siemens Networks 2037 Linnoitustie 6 2038 Espoo 02600 2039 Finland 2041 Phone: +358 (50) 4871445 2042 Email: Hannes.Tschofenig@gmx.net 2043 URI: http://www.tschofenig.priv.at 2045 Eliot Lear 2046 Cisco Systems GmbH 2047 Richtistrasse 7 2048 Wallisellen, ZH CH-8304 2049 Switzerland 2051 Phone: +41 44 878 9200 2052 Email: lear@cisco.com 2054 Jim Schaad 2055 Soaring Hawk Consulting 2057 Email: ietf@augustcellars.com