RFC 7831
Application Bridging for Federated Access Beyond Web (ABFAB) Architecture
RFC 7831 🎀 Application Bridging for Federated Access Beyond Web (ABFAB) Architecture
Okay bestie, here's the title block tea ☕ This is RFC 7831 from the Internet Engineering Task Force (IETF). It's Informational, which means it's published for info and isn't a Standards Track spec 🙈 It came out in May 2016 with ISSN 2070-1721. The authors are J. Howlett (Jisc), S. Hartman (Painless Security), H. Tschofenig (ARM Ltd.), and J. Schaad (August Cellars). This header doesn't list anything it obsoletes or updates, so no drama there 💅
Abstract 📝
Okay so over the last decade there's been a lot of federated access management work 🌈 Most of it focuses on two use cases: network access and web-based access. The tea is that the solutions for those two have been popping up with really few building blocks in common 😭
This memo describes an architecture that extends the security mechanisms people already use for both federated and non-federated access management, including the Remote Authentication Dial-In User Service (RADIUS), the Generic Security Service Application Program Interface (GSS-API), the Extensible Authentication Protocol (EAP), and the Security Assertion Markup Language (SAML). It's mainly about federated access management for non-web-based services, and it's built to scale to huge numbers of Identity Providers, Relying Parties, and federations, iykyk 🦋
Status of This Memo 🌸
This document is not an Internet Standards Track specification. It's published for informational purposes ✨
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community, and it has gotten public review and been approved for publication by the Internet Engineering Steering Group (IESG). Not every document the IESG approves is a candidate for any level of Internet Standard, so check out Section 2 of RFC 5741 for the details 🧐
Info about this document's current status, any errata, and how to send feedback lives at http://www.rfc-editor.org/info/rfc7831 💜
Copyright Notice 🔥
Copyright (c) 2016 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please read these carefully, because they spell out your rights and restrictions for this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions, and they're provided without warranty as described in the Simplified BSD License 🫶
Table of Contents 🗺️
1. Introduction ....................................................4
1.1. Terminology ................................................5
1.1.1. Channel Binding .....................................6
1.2. An Overview of Federation ..................................8
1.3. Challenges for Contemporary Federation ....................11
1.4. An Overview of ABFAB-Based Federation .....................11
1.5. Design Goals ..............................................14
2. Architecture ...................................................15
2.1. Relying Party to Identity Provider ........................16
2.1.1. AAA, RADIUS, and Diameter ..........................17
2.1.2. Discovery and Rules Determination ..................19
2.1.3. Routing and Technical Trust ........................20
2.1.4. AAA Security .......................................21
2.1.5. SAML Assertions ....................................22
2.2. Client to Identity Provider ...............................24
2.2.1. Extensible Authentication Protocol (EAP) ...........24
2.2.2. EAP Channel Binding ................................26
2.3. Client to Relying Party ...................................26
2.3.1. GSS-API ............................................27
2.3.2. Protocol Transport .................................28
2.3.3. Re-authentication ..................................29
3. Application Security Services ..................................29
3.1. Authentication ............................................29
3.2. GSS-API Channel Binding ...................................31
3.3. Host-Based Service Names ..................................32
3.4. Additional GSS-API Services ...............................33
4. Privacy Considerations .........................................34
4.1. Entities and Their Roles ..................................35
4.2. Privacy Aspects of ABFAB Communication Flows ..............36
4.2.1. Client to RP .......................................36
4.2.2. Client to IdP (via Federation Substrate) ...........37
4.2.3. IdP to RP (via Federation Substrate) ...............38
4.3. Relationship between User and Entities ....................39
4.4. Accounting Information ....................................39
4.5. Collection and Retention of Data and Identifiers ..........39
4.6. User Participation ........................................40
5. Security Considerations ........................................40
6. References .....................................................41
6.1. Normative References ......................................41
6.2. Informative References ....................................42
Acknowledgments ...................................................46
Authors' Addresses ................................................461. Introduction ✨
Okay, so there have been a ton of security mechanisms deployed on the Internet to manage access to all kinds of resources 🌐 Over the last decade they've been generalized and scaled up through things like the Simple Authentication and Security Layer (SASL) with the Generic Security Server Application Program Interface (GSS-API), known as the GS2 family [RFC5801]; the Security Assertion Markup Language (SAML) [OASIS.saml-core-2.0-os]; and the Authentication, Authorization, and Accounting (AAA) architecture, as embodied in RADIUS [RFC2865] and Diameter [RFC6733] 💅
A Relying Party (RP) is the entity that manages access to some resource. The entity requesting access to that resource is often called the client 🙋 Lots of security mechanisms are basically an exchange of information between these two, so the RP is able to decide whether the client is authorized or not. Main character energy for the RP, honestly 👑
Some security mechanisms let the RP delegate parts of the access management decision to an entity called the Identity Provider (IdP). This delegation needs technical signaling, trust, and a shared understanding of semantics between the RP and IdP. All of that is usually managed inside a relationship called a "federation". That's why this style of access management is described as "federated access management" 🦋
Federated access management has evolved over the last decade through specs like SAML [OASIS.saml-core-2.0-os], OpenID (http://www.openid.net), OAuth [RFC6749], and WS-Trust [WS-TRUST]. The benefits of federated access management include:
Single or simplified sign-on:
An Internet service can delegate access management, and the responsibilities that come with it (like identity management and credentialing), to an organization that already has a long-term relationship with the client. This is super attractive because RPs often don't want those responsibilities 🙈 The client also needs fewer credentials, which is very desirable 🔑
Data minimization and user participation:
Honestly, an RP often doesn't need to know a client's identity to reach an access management decision. It's frequently enough for the RP to know specific attributes about the client, like whether they're affiliated with a particular organization or have a certain role or entitlement. Sometimes the RP only needs a pseudonym of the client 🫣
Before the IdP releases attributes to the RP, it checks configuration and policy to see if those attributes are allowed to be released. Right now, the client doesn't directly participate in this decision.
Provisioning:
Sometimes an RP needs, or would like, to know more about a client than just an affiliation or a pseudonym. For example, an RP may want the client's email address or name 📧 Some federated access management technologies let the IdP supply this info, either when the RP asks for it or unprompted 🌸
This memo describes the Application Bridging for Federated Access Beyond web (ABFAB) architecture. This architecture is about federated access management, mainly for non-web-based services 🎀 It makes use of extensions to the commonly used security mechanisms for both federated and non-federated access management, including RADIUS, the Generic Security Service (GSS), the Extensible Authentication Protocol (EAP), and SAML. The architecture should be extended to use Diameter in the future. It's designed to scale to large numbers of IdPs, RPs, and federations, so it's built to be a whole vibe at any size 🌈✨
1.1. Terminology 📚💅
okay so first things first, this doc borrows its identity management and privacy vocab from [RFC6973] 🫶 the big words we're using are "identity provider", "relying party", "identifier", "pseudonymity", "unlinkability", and "anonymity". learn these and the rest of the doc slides way smoother ✨
in this architecture, the IdP is made up of three parts: an EAP server, a RADIUS server, and, optionally, a SAML Assertion service 🎀
we also use "Network Access Identifier" (NAI) as defined in [RFC7542] 🦋 an NAI has a realm identifier, which is tied to a AAA server (and therefore an IdP), plus a username, which belongs to one specific client of that IdP. it's giving two-part identity, very cute 🧚
ngl some people have said this doc reads kinda inconsistent, and I get why 😭 the standards we're referencing all call the same thing by different names. so the vibe is: we use either the ABFAB term or whatever term the standard under discussion uses, depending on what fits. to keep you from getting lost, here's Table 1, a mapping of all those different names 💜 (items marked "N/A" (not applicable) mean there's no name for that entity in that protocol 🙈)
+----------+-----------+--------------------+-----------------------+
| Protocol | Client | Relying Party | Identity Provider |
+----------+-----------+--------------------+-----------------------+
| ABFAB | N/A | Relying Party (RP) | Identity Provider |
| | | | (IdP) |
| | | | |
| | Initiator | Acceptor | N/A |
| | | | |
| | Client | Server | N/A |
| | | | |
| SAML | Subject | Service provider | Issuer |
| | | | |
| GSS-API | Initiator | Acceptor | N/A |
| | | | |
| EAP | EAP peer | EAP authenticator | EAP server |
| | | | |
| AAA | N/A | AAA client | AAA server |
| | | | |
| RADIUS | user | NAS | N/A |
| | | | |
| | N/A | RADIUS client | RADIUS server |
+----------+-----------+--------------------+-----------------------+
Table 1: Terminology1.1.1. Channel Binding 🔗✨
okay bestie, "channel binding" shows up in this doc in two different contexts, and it means something different in each one 😭 so let's sort it out 🌸
first up, EAP channel binding. this one is used to implement GSS-API naming semantics 💖 the peer sends a set of attributes over to the EAP server, either as part of the EAP conversation or as part of a secure association protocol. the attributes also get passed along in the back-end protocol from the EAP authenticator to the EAP server. then the EAP server checks that the attributes are consistent and sends the confirmation back to the peer 🫶 in this document, if we just say "channel binding" with no extra label, we mean EAP channel binding. no cap, that's the default 💅
next is GSS-API channel binding. it protects against man-in-the-middle attacks when GSS-API is used for authentication inside of some tunnel, and it's similar to a thing in EAP called "cryptographic binding" 🦋 the way it works is that each side derives a cryptographic value from the tunnel itself, then uses that value to prove to the other side that it knows it 🔐 main character energy for sure 👑
if you want the full breakdown of how these two differ, go read [RFC5056] 📖 but the highlights are:
- GSS-API channel binding specifies that there is nobody between the client and the EAP authenticator. 🙈
- EAP channel binding allows the client to know certain EAP authenticator attributes, like the EAP authenticator's name. 🎀
now, here's the thing 🥹 people usually think of channel binding together with mutual authentication. it's so common that, without any additional qualification, channel binding should be assumed to imply mutual authentication ✨ in GSS-API, if there's no mutual authentication, only the acceptor has authenticated the initiator. and in EAP, if there's no mutual authentication, only the EAP server has authenticated the peer. one-sided is lowkey weird, but it's sometimes actually useful! 🧚
for example, picture a user who wants to access a protected resource for a shared whiteboard in a conference room 🍓 the whiteboard is the acceptor, so it knows the initiator is authorized to give it a presentation. the user can check that the whiteboard got the right presentation just by looking at it 👀 (the presentation should not be confidential in this case, fr fr 💀)
if channel binding is used without mutual authentication, it's basically a request to disclose the resource in the context of one particular channel 🌈 this kind of authentication is kinda similar to a holder-of-key SAML Assertion. but here's the tea 🫢 even though it's not happening in the protocol itself, mutual authentication is still happening in the overall system, because the user can visually authenticate the content. and that matches every use of channel binding without protocol-level mutual authentication that we've found so far 🎀💜
1.2. An Overview of Federation 🌸
okay so, in the last section we met these three entities:
- the client,
- the IdP, and
- the RP.
bestie, there's one more entity to meet: the Individual 🧚. An Individual is a human being who is using the client. In any given situation, an Individual may or may not exist. Clients can act as front ends for Individuals, or they can be independent entities that get set up and allowed to run on their own 🤖. An example of that is the Trust Router Protocol (https://www.ietf.org/proceedings/86/slides/slides-86-rtgarea-0.pdf), where the routers use ABFAB to authenticate to each other. iconic, right? 💖
These entities and their relationships are shown in Figure 1 ✨
,----------\ ,---------\
| Identity | Federation | Relying |
| Provider + <--------------------> + Party |
`----------' '---------'
<
\
\ Authentication
\
\
\
\
\ +---------+
\ | | O
v| Client | \|/ Individual
| | |
+---------+ / \
Figure 1: Entities and Their Relationshipsso the relationships between the entities in Figure 1 go like this 👑
Federation
The IdP and the RPs are part of a federation 🦋. The relationship may be direct (they have an explicit trust relationship) or transitive (the trust relationship is mediated by one or more entities). The federation relationship is governed by a federation agreement. Within a single federation, there may be multiple IdPs as well as multiple RPs. the vibes are very "more the merrier" 🫶
Authentication
There is a direct relationship between the client and the IdP. This relationship provides the means by which they trust each other and can securely authenticate each other 🔐
a federation agreement usually covers two big things: operational specifications and legal rules 📜
Operational Specifications:
The goal of operational specifications is to provide enough definition that the system works and interoperability is possible ✨. These include the technical specifications (e.g., protocols used to communicate between the three parties), process standards, policies, identity proofing, credential and authentication algorithm requirements, performance requirements, assessment and audit criteria, etc. it's a whole list, bestie 😭
Legal Rules:
The legal rules take the legal framework into consideration and provide contractual obligations for each entity 💅. The rules define the responsibilities of each party and provide further clarification of the operational specifications. These legal rules regulate the operational specifications, make operational specifications legally binding to the participants, and define and govern the rights and responsibilities of the participants. The legal rules may, for example, describe liability for losses, termination rights, enforcement mechanisms, measures of damage, dispute resolution, warranties, etc. not me reading "warranties" and going "wait, that's a lot" 💀
The operational specifications can demand the usage of a specific technical infrastructure, including requirements on the message routing intermediaries, to offer the required technical functionality. In other environments, the operational specifications require fewer technical components in order to meet the required technical functionality 🌈
The legal rules include many non-technical aspects of federation, such as business practices and legal arrangements, which are outside the scope of the IETF 🙈. The legal rules can still have an impact on the architectural setup or on how to ensure the dynamic establishment of trust, so it's not totally separate, lowkey 🧠
While a federation agreement is often discussed in the context of formal relationships, such as between an enterprise and an employee or between a government and a citizen, a federation agreement does not have to require any particular level of formality 🌸. For an IdP and a client, it is sufficient for a relationship to be established by something as simple as using a web form and confirmation email 📧. For an IdP and an RP, it is sufficient for the IdP to publish contact information along with a public key and for the RP to use that data. Within the framework of ABFAB, it will generally be required that a mechanism exist for the IdP to be able to trust the identity of the RP; if this is not present, then the IdP cannot provide the assurances to the client that the identity of the RP has been established. so the trust has to be able to flow both ways, bestie 🔁
The nature of federation dictates that there exists some form of relationship between the IdP and the RP. This is particularly important when the RP wants to use information obtained from the IdP for access management decisions and when the IdP does not want to release information to every RP (or only under certain conditions) 🎀
While it is possible to have a bilateral agreement between every IdP and every RP, on an Internet scale, this setup requires the introduction of the multilateral federation concept, as the management of such pair-wise relationships would otherwise prove burdensome 🕸️. Nobody wants to manage a million one-on-one deals, honestly 😩
The IdP will typically have a long-term relationship with the client. This relationship typically involves the IdP positively identifying and credentialing the client (for example, at the time of employment within an organization). When dealing with Individuals, this process is called "identity proofing" [NIST-SP.800-63-2]. The relationship will often be instantiated within an agreement between the IdP and the client (for example, within an employment contract or terms of use that stipulate the appropriate use of credentials and so forth) 📝
The nature and quality of the relationship between the client and the IdP are important contributors to the level of trust that an RP may assign to an assertion describing a client made by an IdP. This is sometimes described as the level of assurance [NIST-SP.800-63-2] 🥇
Federation does not require an a priori relationship or a long-term relationship between the RP and the client; it is this property of federation that yields many of its benefits ✨. However, federation does not preclude the possibility of a pre-existing relationship between the RP and the client, or the possibility that the RP and client may use the introduction to create a new long-term relationship independent of the federation 💜
Finally, it is important to reiterate that in some scenarios there might indeed be an Individual behind the client, and in other cases the client may be autonomous 🤖🧚
1.3. Challenges for Contemporary Federation 🦋
As federated IdPs and RPs (services) proliferate, the role of an Individual can become ambiguous in certain circumstances 🌀. For example, a school might provide online access for a student's grades to their parents for review and to the student's teacher for modification. A teacher who is also a parent must clearly distinguish their role upon access. this is a whole plot twist, bestie 😭
Similarly, as federations proliferate, it becomes increasingly difficult to discover which IdP(s) a user is associated with. This is true for both the web and non-web case, but is particularly acute for the latter, as many non-web authentication systems are not semantically rich enough on their own to allow for such ambiguities 🙃. For instance, in the case of an email provider, SMTP and IMAP do not have the ability for the server to request information from the client, beyond the client NAI, that the server would then use to decide between the multiple federations it is associated with. However, the building blocks do exist to add this functionality 🛠️✨
1.4. An Overview of ABFAB-Based Federation ✨
okay so the last bit was all about the big federation picture and who gets access to what 💅 this part is a quick tour of where ABFAB fits in that model 🦋
picture a client trying to connect to a server so it can get some data or do a transaction. before anything else, they have to mutually authenticate 🫶 here's the ABFAB play-by-play (there's a visual in Figure 2 if you're a visual girly 🎀):
- Client configuration: the client gets an NAI from the IdP 🪪 it also gets any keys, certificates, passwords, or other secret and public info it needs to run the EAP protocols with the IdP 🔐
- Authentication mechanism selection: the client is set up to use the GSS-EAP GSS-API mechanism for authentication/authorization 🧩
- Client provides an NAI to RP: the client opens a transport to the RP and starts GSS-EAP authentication 🚪 the RP sends back an EAP request (nested in GSS-EAP) asking for the client's name. the client replies with an EAP response using an NAI name form that, at a minimum, has the realm portion of its full NAI 🌸
- Discovery of federated IdP: the RP uses preconfigured info or a federation proxy to figure out which IdP to use, based on policy and the realm portion of the client's NAI 🔍 this one gets a deeper dive below (Section 2.1.2) 👀
- Request from RP to IdP: once the RP knows who the IdP is, it (or its agent) sends a RADIUS request to the IdP 📨 the RADIUS Access-Request wraps the EAP response. at this point the RP probably has no idea who the client is 🙈 it sends its own identity to the IdP in AAA attributes, and it may also send a SAML request in an AAA attribute. the AAA network then checks that the identity the RP claimed is valid ✅
- IdP begins EAP with the client: the IdP sends the client an EAP message with the EAP method to use 🎯 the IdP should not ask for the client's name again in this message 🙅♀️ but clients still need to be able to handle it anyway. in this case the IdP must accept a realm only, which is how the client's name stays hidden from the RP 🤫 the methods that are available and appropriate get covered below (Section 2.2.1) 📚
- EAP is run: a whole bunch of EAP messages go back and forth between the client (EAP peer) and the IdP (EAP server) until the result of the authentication protocol is decided 🏓 how many messages and what's in them depends on the EAP method picked. if the IdP can't authenticate the client, it sends an EAP failure message to the RP 💀 as part of the EAP method, the client sends an EAP channel-binding message to the IdP (Section 2.2.2) 🔗 in that message, the client says, among other things, which RP it's trying to authenticate to. the IdP checks the channel-binding data from the client against the data the RP gave it over AAA. if the bindings don't match, the IdP sends an EAP failure message to the RP 🚨
- Successful EAP authentication: now the IdP (EAP server) and the client (EAP peer) have mutually authenticated each other 🥰 as a result, the client and the IdP both hold two cryptographic keys: a Master Session Key (MSK) and an Extended MSK (EMSK) 🗝️✨ at this point the client has a level of assurance about the RP's identity, based on the name checking the IdP did, using the RP naming info from the AAA framework and from the client (through the channel-binding data) 💖
- Local IdP policy check: the IdP checks its local policy to see whether the RP and client are allowed to do a given transaction/service and, if so, which attributes (if any) get released to the RP 📋 if the IdP gets a policy failure, it sends an EAP failure message to the RP and client 🙅♀️ (the RP already did its policy checks during discovery, fyi 💅)
- IdP provides the RP with the MSK: the IdP sends a success result EAP to the RP, along with an optional set of AAA attributes tied to the client (usually as one or more SAML Assertions) 🎁 the EAP MSK gets sent back to the RP too 🔑
- RP processes results: when the RP gets the result from the IdP, it should have enough info to either grant or refuse a resource Access-Request 🧠 it might also have info that links the client to specific authorization identities. if it needs more attributes from the IdP, it can make a new SAML request to the IdP 📬 then it applies those results in whatever way makes sense for the app 🛠️
- RP returns results to client: once the RP has a response, it must tell the client what happened 📣 if everything went well, everyone is authenticated and the app carries on with the right authorization levels 🎉 the client can then finish authenticating the RP by using the EAP MSK value 🤝
Here's the whole thing as a diagram, because it's giving main character energy 🎬
Relying Client Identity
Party Provider
| (1) | Client configuration
| | |
|<-----(2)----->| | Mechanism selection
| | |
|<-----(3)-----<| | NAI transmitted to RP
| | |
|<=====(4)====================>| IdP Discovery
| | |
|>=====(5)====================>| Access-Request from RP to IdP
| | |
| |< - - (6) - -<| EAP method to client
| | |
| |< - - (7) - ->| EAP exchange to authenticate
| | | client
| | |
| | (8 & 9) Local policy check
| | |
|<====(10)====================<| Results to RP
| | |
(11) | | RP processes results
| | |
|>----(12)----->| | Results to client
Legend:
-----: Between client and RP
=====: Between RP and IdP
- - -: Between client and IdP (via RP)
Figure 2: ABFAB Authentication Steps1.5. Design Goals 🎯
okay now the vision board 🪄 our key design goals are:
- Each party in a transaction will be authenticated, although perhaps not identified, and the client will be authorized for access to a specific resource 🪪✨
- The means of authentication is decoupled from the application protocol so as to allow for multiple authentication methods with minimal changes to the application 🔀
- The architecture requires no sharing of long-term private keys between clients and RPs 🔐🙅♀️
- The system will scale to large numbers of IdPs, RPs, and users 📈🌍
- The system will be designed primarily for non-web-based authentication 🛜
- The system will build upon existing standards, components, and operational practices 🧱
designing new three-party authentication and authorization protocols is genuinely hard 😭 it's so easy to get cryptographic flaws in there. getting people to actually deploy them is even harder 💀 a lot of the attention on federated access has gone to the web, but this document is all about the non-web world, so it focuses on protocols where HTTP isn't used 🌿 even with everyone layering every protocol on top of HTTP these days, there are still plenty of protocols that don't use HTTP-based transports 🙌 a lot of those are missing a native authentication and authorization framework of the style shown in Figure 1 🧐
2. Architecture 🏗️
we already introduced the federated access architecture, with the illustration of all the actors that need to interact 🎭 this section goes deeper into how to support non-web-based apps and explains why the design decisions were made 🧠 the main theme here is reusing building blocks that are already deployed and rearranging them in a fresh way 🔄✨
even though this architecture assumes updates to the RP, the client, and the IdP, those changes are kept to a minimum 🤏 a mechanism that can show deployment benefits (like easy updates to existing software, low implementation effort, etc.) is preferred 🏆 there may also be a need to specify multiple mechanisms to cover the range of deployment scenarios 🌈
there are a bunch of ways to encapsulate EAP into an application protocol 🎀 for easy integration with a wide range of non-web-based application protocols, GSS-API was chosen 💜 the technical spec for GSS-EAP can be found in [RFC7055] 📖
the architecture is made of several building blocks, as shown in Figure 3 🧩 in the sections that follow, we talk about the data flow between each entity, the protocols used for that data flow, and some of the trade-offs made in choosing the protocols ⚖️
+--------------+
| Identity |
| Provider |
| (IdP) |
+-^----------^-+
* EAP o RADIUS
* o
--v----------v--
/// \\\
// \\
| Federation |
| Substrate |
\\ //
\\\ ///
--^----------^--
* EAP o RADIUS
* o
+-------------+ +-v----------v--+
| | | |
| Client | EAP/EAP Method | Relying Party |
| Application |<****************>| (RP) |
| | GSS-API | |
| |<---------------->| |
| | Application | |
| | Protocol | |
| |<================>| |
+-------------+ +---------------+
Legend:
<****>: Client-to-IdP Exchange
<---->: Client-to-RP Exchange
<oooo>: RP-to-IdP Exchange
<====>: Protocol through which GSS-API/GS2 exchanges are tunneled
Figure 3: ABFAB Protocol Instantiation2.1. Relying Party to Identity Provider ✨
okay so the RP and the IdP talk to each other through the Federation Substrate 💖. this comms channel has a few big jobs:
- Establishing the trust relationship between the RP and the IdP.
- Determining the rules governing the relationship.
- Conveying authentication packets from the client to the IdP and back.
- Providing the means of establishing a trust relationship between the RP and the client.
- Providing a means for the RP to obtain attributes about the client from the IdP.
the ABFAB working group chose the AAA framework for the messages going between the RP and IdP 🦋. and honestly, AAA slays at all of these jobs, like this:
- The AAA backbone supplies the trust relationship between the RP and the IdP.
- The agreements governing a specific AAA backbone contain the rules governing the relationships within the AAA federation.
- A method exists for carrying EAP packets within RADIUS [RFC3579] and Diameter [RFC4072].
- The use of EAP channel binding [RFC6677] along with the core ABFAB protocol provide the pieces necessary to establish the identities of the RP and the client, while EAP provides the cryptographic methods for the RP and the client to validate that they are talking to each other 🔐.
- A method exists for carrying SAML packets within RADIUS [RFC7833]; this method allows the RP to query attributes about the client from the IdP.
protocols that support the same framework but do different routing are expected to show up in the future 🌈. one of them is the Trust Router, which is set up to build a trusted point-to-point channel on the fly. you can peek at the slides here: https://www.ietf.org/proceedings/86/slides/slides-86-rtgarea-0.pdf 👀.
2.1.1. AAA, RADIUS, and Diameter 🎀
the AAA framework with RADIUS [RFC2865] and Diameter [RFC6733] has been a total success for network access authentication, deployment-wise 🙌. quick mapping so you can follow Figure 1: the IdP is the AAA server, the RP is the AAA client, and the building blocks of a federation are AAA proxies, relays, and redirect agents (especially when third parties run them, like AAA brokers and clearinghouses) 💅. for network access, the front end (the path between the end host and the AAA client) is handled by link-layer protocols that forward authentication exchanges back and forth. a big real-world example is eduroam, which lives at https://www.eduroam.org 🍓.
ABFAB can ride on federation agreements that already exist, so those agreements just need to get expanded to cover the ABFAB architecture ✨. the AAA framework has already handled some of the problems we talked about, like:
- It already has a method for routing requests based on a domain.
- It already has an extensible architecture allowing for new attributes to be defined and transported.
- Pre-existing relationships can be reused.
okay, the astute reader is going to ask: RADIUS and Diameter are super similar, so why not just pick one? 🤔 well, they live in different places. RADIUS is everywhere in enterprise and university networks, and it's also used by operators of fixed networks. Diameter is the one deployed by mobile network operators 📱. another big difference is that RADIUS today is mostly carried over UDP. the choice of which protocol to deploy is left to implementers 💜. the protocol defines all the necessary new AAA attributes as RADIUS attributes, and a future document could define those same AAA attributes for a Diameter environment. we also note that there are proxies that convert from RADIUS to Diameter and back, which means both can live in a single Federation Substrate 🧚.
through the integrity-protection mechanisms in the AAA framework, the IdP can establish technical trust that messages are really coming from the appropriate RP 🫶. every interaction is tied to one federation at the policy level. the legal or business relationship decides what statements the IdP is trusted to make and how the RP is supposed to interpret them. the AAA framework also lets the RP, or elements between the RP and IdP, make statements about the RP itself.
the AAA framework is the transport for attributes. statements about the client made by the IdP, statements about the RP, and other info all travel as attributes 🎁.
one demand the AAA substrate makes of the upper layers is that they must properly identify the endpoints of the communication 😤. the AAA client at the RP has to be able to figure out where to send each RADIUS or Diameter message. without this, the RP would have to work out the client's identity on its own, with no help from an IdP 🙈. this architecture uses the Network Access Identifier (NAI), where the realm component tells you which IdP to talk to [RFC7542]. the NAI is represented and consumed by the GSS-API layer as GSS_C_NT_USER_NAME, as specified in [RFC2743]. the GSS-API EAP mechanism puts the NAI in the EAP Response/Identity message 📨.
and at the time of this writing, no profiles for using Diameter have been created 😭.
2.1.2. Discovery and Rules Determination 🔍
the RP can use the AAA protocols to talk to the IdP, but it might have multiple Federation Substrates to pick from 💅. the RP has a list of criteria it uses to choose which federation to go with. the federation that gets selected must
- be able to communicate with the IdP.
- match the business rules and technical policies required for the RP security requirements.
the RP needs to discover which federation will be used to contact the IdP 🗺️. the first selection criterion during discovery is the name of the IdP being contacted. the second is the set of business rules and technical policies governing the relationship, which is called "rules determination" 👑. the RP also needs to establish technical trust in its communications with the IdP.
rules determination covers a broad range of decisions about the exchange 🌸. one of them is whether the given RP is allowed to talk to the IdP through a given federation at all, so rules determination includes the basic authorization decision. other factors count too, like what policies govern releasing information about the client to the RP, and what policies govern how the RP is allowed to use that information. rules determination is ultimately a business function, but it hugely affects the technical exchanges 💀. the protocols need to communicate the result of authorization. when multiple sets of rules are possible, the protocol must disambiguate which set of rules is in play. some rules have technical enforcement mechanisms; for example, in some federations, intermediaries validate the information being communicated within the federation ✨.
at the time of this writing, no protocol mechanism has been specified that lets an AAA client figure out whether an AAA proxy will actually be able to route AAA requests to a specific IdP 🥺. AAA routing is affected by business rules and technical policies that can get really complicated, so for now, route selection is based on manual configuration 🛠️.
2.1.3. Routing and Technical Trust
Several approaches to having messages routed through the Federation Substrate are possible. These routing methods can most easily be classified based on the mechanism for technical trust that is used. The choice of technical trust mechanism constrains how rules determination is implemented. Regardless of what deployment strategy is chosen, it is important that the technical trust mechanism be able to validate the identities of both parties to the exchange. The trust mechanism must ensure that the entity acting as the IdP for a given NAI is permitted to be the IdP for that realm and that any service name claimed by the RP is permitted to be claimed by that entity. Here are the categories of technical trust determination:
AAA Proxy: The simplest model is that an RP is a AAA client and can send the request directly to a AAA proxy. The hop-by-hop integrity protection of the AAA fabric provides technical trust. An RP can submit a request directly to the correct federation. Alternatively, a federation disambiguation fabric can be used. Such a fabric takes information about what federations the RP is part of and what federations the IdP is part of, and it routes a message to the appropriate federation. The routing of messages across the fabric, plus attributes added to requests and responses, together provide rules determination. For example, when a disambiguation fabric routes a message to a given federation, that federation's rules are chosen. Name validation is enforced as messages travel across the fabric. The entities near the RP confirm its identity and validate names it claims. The fabric routes the message towards the appropriate IdP, validating the name of the IdP in the process. The routing can be statically configured. Alternatively, a routing protocol could be developed to exchange reachability information about a given IdP and to apply policy across the AAA fabric. Such a routing protocol could flood naming constraints to the appropriate points in the fabric.
Trust Broker: Instead of routing messages through AAA proxies, some trust broker could establish keys between entities near the RP and entities near the IdP. The advantage of this approach is efficiency of message handling. Fewer entities are needed to be involved for each message. Security may be improved by sending individual messages over fewer hops. Rules determination involves decisions made by trust brokers about what keys to grant. Also, associated with each credential is context about rules and about other aspects of technical trust, including names that may be claimed. A routing protocol similar to the one for AAA proxies is likely to be useful to trust brokers in flooding rules and naming constraints.
Global Credential: A global credential such as a public key and certificate in a public key infrastructure can be used to establish technical trust. A directory or distributed database such as the Domain Name System is used by the RP to discover the endpoint to contact for a given NAI. Either the database or certificates can provide a place to store information about rules determination and naming constraints. Provided that no intermediates are required (or appear to be required) and that the RP and IdP are sufficient to enforce and determine rules, rules determination is reasonably simple. However, applying certain rules is likely to be quite complex. For example, if multiple sets of rules are possible between an IdP and RP, confirming that the correct set is used may be difficult. This is particularly true if intermediates are involved in making the decision. Also, to the extent that directory information needs to be trusted, rules determination may be more complex.
Real-world deployments are likely to be mixtures of these basic approaches. For example, it will be quite common for an RP to route traffic to a AAA proxy within an organization. That proxy could then use any of the above three methods to get closer to the IdP. It is also likely that, rather than being directly reachable, the IdP may have a proxy on the edge of its organization. Federations will likely provide a traditional AAA proxy interface even if they also provide another mechanism for increased efficiency or security.
2.1.4. AAA Security
For the AAA framework, there are two different places where security needs to be examined. The first is the security that is in place for the links in the AAA backbone being used. The second are the nodes that form the AAA backbone.
The default link security for RADIUS is showing its age, as it uses MD5 and a shared secret to both obfuscate passwords and provide integrity on the RADIUS messages. While some EAP methods include the ability to protect the client authentication credentials, the MSK returned from the IdP to the RP is protected only by RADIUS security. In many environments, this is considered to be insufficient, especially as not all attributes are obfuscated and can thus leak information to a passive eavesdropper. The use of RADIUS with Transport Layer Security (TLS) [RFC6614] and/or Datagram Transport Layer Security (DTLS) [RFC7360] addresses these attacks. The same level of security is included in the base Diameter specifications.
2.1.5. SAML Assertions
For the traditional use of AAA frameworks, i.e., granting access to a network, an affirmative response from the IdP is sufficient. In the ABFAB world, the RP may need to get significantly more additional information about the client before granting access. ABFAB therefore has a requirement that it can transport an arbitrary set of attributes about the client from the IdP to the RP.
The Security Assertion Markup Language (SAML) [OASIS.saml-core-2.0-os] was designed in order to carry an extensible set of attributes about a subject. Since SAML is extensible in the attribute space, ABFAB has no immediate needs to update the core SAML specifications for our work. It will be necessary to update IdPs that need to return SAML Assertions to RPs and for both the IdP and the RP to implement a new SAML profile designed to carry SAML Assertions in AAA. The new profile can be found in [RFC7833]. As SAML statements will frequently be large, RADIUS servers and clients that deal with SAML statements will need to implement [RFC7499].
There are several issues that need to be highlighted:
o The security of SAML Assertions.
o Namespaces and mapping of SAML attributes.
o Subject naming of entities.
o Making multiple queries about the subject(s).
o Level of assurance for authentication.
SAML Assertions have an optional signature that can be used to protect and provide the origination of the assertion. These signatures are normally based on asymmetric key operations and require that the verifier be able to check not only the cryptographic operation but also the binding of the originator's name and the public key. In a federated environment, it will not always be possible for the RP to validate the binding; for this reason, the technical trust established in the federation is used as an alternate method of validating the origination and integrity of the SAML Assertion.
Attributes in a SAML Assertion are identified by a name string. The name string is either assigned by the SAML issuer context or scoped by a namespace (for example, a URI or object identifier (OID)). This means that the same attribute can have different name strings used to identify it. In many cases, but not all, the federation agreements will determine what attributes and names can be used in a SAML statement. This means that the RP needs to map from the SAML issuer or federation name, type, and semantic to the name, type, and semantics that the policies of the RP are written in. In other cases, the Federation Substrate, in the form of proxies, will modify the SAML Assertions in transit to do the necessary name, type, and value mappings as the assertion crosses boundaries in the federation. If the proxies are modifying the SAML Assertion, then they will remove any signatures on the SAML Assertion, as changing the content of the SAML Assertion would invalidate the signature. In this case, the technical trust is the required mechanism for validating the integrity of the assertion. (The proxy could re-sign the SAML Assertion, but the same issues of establishing trust in the proxy would still exist.) Finally, the attributes may still be in the namespace of the originating IdP. When this occurs, the RP will need to get the required mapping operations from the federation agreements and do the appropriate mappings itself.
[RFC7833] has defined a new SAML name format that corresponds to the NAI name form defined by [RFC7542]. This allows for easy name matching in many cases, as the name form in the SAML statement and the name form used in RADIUS or Diameter will be the same. In addition to the NAI name form, [RFC7833] also defines a pair of implicit name forms corresponding to the client and the client's machine. These implicit name forms are based on the Identity-Type enumeration defined in the Tunnel Extensible Authentication Protocol (TEAP) specification [RFC7170]. If the name form returned in a SAML statement is not based on the NAI, then it is a requirement on the EAP server that it validate that the subject of the SAML Assertion, if any, is equivalent to the subject identified by the NAI used in the RADIUS or Diameter session.
RADIUS has the ability to deal with multiple SAML queries for those EAP servers that follow [RFC5080]. In this case, a State attribute will always be returned with the Access-Accept. The EAP client can then send a new Access-Request with the State attribute and the new SAML request. Multiple SAML queries can then be done by making a new Access-Request, using the State attribute returned in the last Access-Accept to link together the different RADIUS sessions.
Some RPs need to ensure that specific criteria are met during the authentication process. This need is met by using levels of assurance. A level of assurance is communicated to the RP from the EAP server by using a SAML Authentication Request, using the Authentication Profile described in [RFC7833]. When crossing boundaries between different federations, (1) the policy specified will need to be shared between the two federations, (2) the policy will need to be mapped by the proxy server on the boundary, or (3) the proxy server on the boundary will need to supply information to the EAP server so that the EAP server can do the required mapping. If this mapping is not done, then the EAP server will not be able to enforce the desired level of assurance, as it will not understand the policy requirements.
2.2. Client to Identity Provider 🦋
Okay so bestie, when we look at the convo between the client and the IdP, there are two big things that need sorting out:
- The client and the IdP need to mutually authenticate each other. 🔐
- The client and the IdP need to mutually agree on the identity of the RP. 🤝
ABFAB picked EAP for the mutual authentication part and helped cook up some new EAP channel-binding docs for working out the RP's identity. 🍓 [RFC6677] lays out a framework for the channel-binding mechanism, and it lets the IdP check the identity of the RP that the AAA framework gave it against the identity the client provided. Iconic. ✨
2.2.1. Extensible Authentication Protocol (EAP) 💅
Okay, real talk: traditional web federation never describes how a client talks to an IdP to authenticate, so that conversation is just not standardized. 🙈 That causes a bunch of problems. Since nothing is standardized, it's hard for machines to figure out which entity is doing the authenticating, and so which credentials to use and where to type them into the auth form. Humans are way better at handling this stuff. Browsers also make it hard to roll out stronger forms of auth beyond plain usernames and passwords that the server knows. And sometimes the auth interface pops up before the client has really checked that it's talking to the right server. 😭 If you hand control of that interface to a potential attacker, the system gets less secure and phishing becomes a whole thing. Not me being dramatic about it, it's just true. 💀
So it's really desirable to pick some standardized approach for how the client's end host talks to the IdP. There are a bunch of requirements this approach has to meet, which we'll get into below. 🌸
Experience has taught us one key security and scalability requirement: it's important that the RP does not get possession of the client's long-term secret. 🔑 Besides a valuable secret getting exposed, a synchronization problem can pop up when the client changes keys with the IdP. Yikes. 🙅♀️
There's also no single authentication mechanism that gets used everywhere, so the authentication framework must allow for flexible integration of authentication mechanisms. 🧚 Like, some IdPs want hardware tokens and others use passwords. A service provider wants to support both of those, plus other methods from IdPs that nobody has even seen yet. Main character energy for flexibility. 👑
Good news: these requirements can be met using standardized and successfully deployed tech, namely the EAP framework [RFC3748]. Figure 3 illustrates the integration graphically. 🌈
EAP is an end-to-end framework. It gives two-way communication between a peer (i.e., client or Individual) through the EAP authenticator (i.e., RP) to the back end (i.e., IdP). That's exactly, and conveniently, the communication path that federated identity needs. 😍 EAP support is already built into AAA systems (see [RFC3579] and [RFC4072]), but a few challenges are still hanging around:
- First up, how do you carry EAP payloads from the end host to the RP? 🤔
- Next, you have to verify the statements the RP made to the client, confirm those statements match the ones made to the IdP, and confirm that all of it lines up with the federation and any federation-specific policy or configuration. Lots of checking, bestie. 📋
- Last but not least, choosing which IdP to use for which service. 🎯
The EAP method used for ABFAB needs to meet these requirements:
- It needs to provide mutual authentication of the client and IdP. 🫶
- It needs to support channel binding. 🔗
As of this writing, the only EAP method that meets these criteria is TEAP [RFC7170]. It can be used alone (if client certificates are used) or with an inner EAP method that does mutual authentication. ✨
2.2.2. EAP Channel Binding 🦋
Okay, so EAP channel binding gets mixed up with a GSS-API facility that's also called "channel binding." 🙃 GSS-API channel binding protects against man-in-the-middle attacks when GSS-API is used for authentication inside some tunnel. It's kind of similar to a facility called "cryptographic binding" in EAP. If you want the breakdown of how these two differ, check out [RFC5056]. 📚
In theory, the client knows the name of the RP it tried to connect to. But if an attacker has intercepted the protocol, the client and the IdP need to be able to detect that situation. 👀 A general overview of the problem, along with a recommended way to deal with the channel-binding issues, is in [RFC6677]. 💜
Since [RFC6677] came out, a number of possible attacks were found. Methods to address those attacks are outlined in [RFC7029]. Security is an ongoing glow-up. 💖
2.3. Client to Relying Party 🌸
The last set of interactions to think about is the one between the client and the RP. This one is kinda the most complex, since at least part of it is outside the scope of the ABFAB work. 😤 The interactions between these parties include:
- Running the protocol that implements the service the RP provides and the client wants. 🍰
- Authenticating the client to the RP and the RP to the client. 🪪
- Providing the necessary security services the service protocol needs, beyond authentication. 🛡️
- Dealing with client re-authentication where it's desired. 🔄
2.3.1. GSS-API 🧚♀️
One of the remaining layers is responsible for integrating federated authentication with the application. Applications have adopted a bunch of different approaches to security, so multiple strategies for integrating federated authentication with applications might be needed. 🌈 To that end, we start with a strategy that integrates with a large number of application protocols. 💪
Lots of applications support GSS-API [RFC2743], including Secure Shell (SSH) [RFC4462], NFS [RFC7530], DNS [RFC3645], and several non-IETF applications. Lots of others, like IMAP, SMTP, the Extensible Messaging and Presence Protocol (XMPP), and the Lightweight Directory Access Protocol (LDAP), support the Simple Authentication and Security Layer (SASL) [RFC4422] framework. 📬 These two approaches work together nicely: by creating a GSS-API mechanism, SASL integration is also handled. Basically, using a GSS-API mechanism with SASL just means putting some headers in front of the mechanism's messages and constraining certain GSS-API options. Easy. 😌
GSS-API is specified in terms of an abstract set of operations that can be mapped into a programming language to form an API. When people first meet GSS-API, they focus on it as an API. 🙋♀️ But from the perspective of authentication for non-web applications, GSS-API should be thought of as a protocol as well as an API. As a protocol, it consists of abstract operations, such as the initial context exchange, which has two sub-operations (GSS_Init_sec_context and GSS_Accept_sec_context) [RFC2743]. An application defines which abstract operations it's going to use and where the messages from those operations fit into the application architecture. A GSS-API mechanism defines what actual protocol messages come out of that abstract message for a given abstract operation. Since this work focuses on one particular GSS-API mechanism, we mostly focus on protocol elements rather than the API view. 🧩
The API view of GSS-API has real value too! Since the abstract operations are well defined, the information a mechanism gets from the application is well defined. The set of assumptions the application is allowed to make is also generally well defined. 📝 So an application protocol that supports GSS-API or SASL is very likely to be usable with a new approach to authentication, including the authentication mechanism defined in this document, with no required modifications. ✅ In some cases, support for a new authentication mechanism has been added through plugin interfaces without modifying the application at all. Even when modifications are required, they can often be limited to supporting a new naming and authorization model. For example, this work focuses on privacy. An application that assumes it will always get an identifier for the client will need to be modified to support anonymity, unlinkability, or pseudonymity. 🙈🎭
So, we use GSS-API and SASL because a bunch of the application protocols we want to federate support these strategies for security integration. 🫶 What does that mean from a protocol standpoint, and how does it relate to other layers? It means we need to design a concrete GSS-API mechanism. We chose a GSS-API mechanism that encapsulates EAP authentication. So GSS-API (and SASL) encapsulates EAP between the end host and the service, and the AAA framework encapsulates EAP between the RP and the IdP. 🪄 The GSS-API mechanism also includes rules about how initiators and services are named, plus per-message security and other facilities required by the applications we want to support. Okay, the plan is giving. 💖
2.3.2. Protocol Transport 🚚
The transport of data between the client and the RP is not provided by GSS-API. GSS-API creates and consumes messages, but it doesn't provide the transport itself. Instead, the protocol using GSS-API needs to provide the transport. 📦 In many cases HTTP or HTTPS is used for this, but other transports are perfectly acceptable too. The core GSS-API document [RFC2743] gives some details on what requirements exist. 📖
We also want to highlight the following:
- The transport does not need to provide either confidentiality or integrity. Once GSS-EAP has finished negotiation, GSS-API can be used to provide both of those services. 🔒 If the negotiation process itself needs protection from eavesdroppers, then the transport would need to provide the necessary services. 👂
- The transport needs to provide reliable transport of the messages. 💯
- The transport needs to ensure that tokens are delivered in order during the negotiation process. 🔢
- GSS-API messages need to be delivered atomically. If the transport breaks up a message, it must also reassemble the message before delivery. 🧃
2.3.3. Re-authentication 🔁
There are times when the RP will want the client to re-authenticate itself. 🙋 These include very long sessions, where the original authentication is time limited, or cases where a different authentication is required in order to complete an operation. GSS-EAP doesn't have any mechanism for the server to start a re-authentication, since all authentication operations start from the client. 😬 So if a protocol using GSS-EAP needs server-initiated re-authentication, a request from the server to the client for the re-authentication to start needs to be placed in the protocol. 📨
Clients can reuse the existing secure connection that GSS-API established, and run the new authentication over that connection by calling GSS_Init_sec_context. At this point, a full re-authentication will be done. 🔐✨
3. Application Security Services ✨
okay so bestie, the big goal here is to plug federated authentication into application protocols people already use, ideally with the existing implementations too 🫶 and while we're doing that, we have to keep up with the best security practices for every technology involved. this section lists the security services and properties the EAP GSS-API mechanism needs to hit those goals. ngl, you could read it as stuff that only applies to that one mechanism, but future app integration strategies are very likely to need similar services, so these will probably get expanded as new strategies show up 💅
3.1. Authentication
GSS-API has an optional security service called "mutual authentication" 🔐 It means that on top of the initiator giving its (potentially anonymous or pseudonymous) identity to the acceptor, the acceptor also confirms its own identity to the initiator. the naming is kinda confusing in ABFAB specifically, so just know that we still say mutual authentication is provided when the acceptor's identity is strongly authenticated to an anonymous initiator 🙈
the thing is, [RFC2743] never really spells out what mutual authentication means. so in this document we define it ourselves, like this:
- If a target name is configured for the initiator, then the initiator trusts that the supplied target name describes the acceptor. This implies that (1) appropriate cryptographic exchanges took place for the initiator to make such a trust decision and (2) after evaluating the results of these exchanges, the initiator's policy trusts that the target name is accurate.
- If no target name is configured for the initiator, then the initiator trusts that the acceptor name, supplied by the acceptor, correctly names the entity it is communicating with.
- Both the initiator and acceptor have the same key material for per-message keys, and both parties have confirmed that they actually have the key material. In EAP terms, there is a protected indication of success.
mutual auth is a big deal for defending against certain flavors of phishing 🎣 clients kinda assume that if someone asks for their credentials, then successfully getting access means they're talking to the party they expected. without mutual authentication, the server could "grant access" no matter what credentials you hand over, which is so not it. mutual auth matches that user intuition way better 💜
so it's important that the GSS-EAP mechanism implements mutual authentication. an initiator needs to be able to request it. when it's requested, only EAP methods that can actually provide the necessary service can be used, and the right steps have to be taken to provide mutual authentication. yes, a wider set of EAP methods could be supported if we didn't require mutual authentication, but the call was made that the client always needs the ability to ask for it 🦋 and in some cases the IdP and the RP won't support mutual authentication at all. the good news is the client will always be able to detect that and make an appropriate security decision ✨
the AAA infrastructure may hide the initiator's identity from the GSS-API acceptor, which gives anonymity between the two. right now, whether the identity gets disclosed is decided by EAP server policy, not by anything the initiator indicates. also, initiators are unlikely to be able to tell whether anonymous communication will actually be provided 😭 for that reason, initiators are unlikely to set the anonymous return flag from GSS_Init_sec_context (Section 2.2.1 of [RFC2743]) 🌸
3.2. GSS-API Channel Binding
[RFC5056] defines a concept of channel binding that's used to prevent man-in-the-middle attacks 🕵️♀️ it works by taking a cryptographic value from the transport security layer and checking that both sides of the GSS-API conversation know that same value. Transport Layer Security (TLS) [RFC5246] is the most common transport security layer used for this 💖
this part needs to be stressed, bestie: channel binding as described in [RFC5056] (also called "GSS-API channel binding" when GSS-API is involved) is not the same thing as EAP channel binding 🙅♀️ GSS-API channel binding is for detecting man-in-the-middle attacks. EAP channel binding is for mutual authentication and acceptor naming checks. see [RFC7055] for the details, and [RFC5056] has a more detailed rundown of how the two facilities differ 🎀
TLS can give you both encryption and integrity on the channel, and it's common to pair SASL and GSS-API with those other security services 🔥
one of the perks of TLS is that a client can validate the name of the server 👑 but that validation depends on a couple of things. the TLS session needs to be using certificates and can't be an anonymous session, and the client and the TLS server need to share a common trust point for the certificate used to validate the server. TLS does provide its own server authentication, but there are lots of situations where, for policy or usability reasons, that authentication just isn't checked. when it is checked, and the trust infrastructure behind TLS authentication is different from the one behind GSS-API mutual authentication, then confirming the endpoints using both trust infrastructures is likely to boost security 🧚 if the endpoints of the GSS-API authentication differ from the endpoints of the lower layer, that's a strong sign something's wrong, like a man-in-the-middle attack 😱 channel binding gives you a way to check whether those endpoints are the same.
the GSS-EAP mechanism needs to support channel binding. when an application provides channel-binding data, the mechanism needs to confirm that it's the same on both sides, consistent with the GSS-API specification ✨
3.3. Host-Based Service Names
IETF security mechanisms usually take a host name and maybe a service, typed in by a user, and then make some trust decision about whether the remote party is the intended one 🍓 that decision can be made with certificates, preconfigured key information, or a previous leap of trust. GSS-API has a relatively flexible naming convention, but most IETF applications that use GSS-API (including SSH, NFS, IMAP, LDAP, and XMPP) have chosen a more restricted naming convention based on the host name 💅 so the GSS-EAP mechanism needs to support host-based service names in order to work with existing IETF protocols 🫶
using host-based service names creates a pretty tricky trust delegation problem 😵💫 who gets to decide whether a particular host name maps to a specific entity? a few possible solutions have been looked at:
- The Public Key Infrastructure (PKI) used by the web has chosen to have a number of trust anchors (root certificate authorities), each of which can map any host name to a public key.
- A number of GSS-API mechanisms, such as Kerberos [RFC1964], have split the problem into two parts. [RFC1964] introduced a new concept called a realm; the realm is responsible for host mapping within itself. The mechanism then decides what realm is responsible for a given name. This is the approach adopted by ABFAB.
GSS-EAP defines a host naming convention that takes into account the host name, the realm, the service, and the service parameters 👑 an example of a GSS-API service name is xmpp/foo@example.com. that identifies the XMPP service on the host foo in the realm example.com. any of the components except for the service name can be omitted from a name, and when one is omitted, a local default is used for that component 🦋
there's no requirement that realm names map to Fully Qualified Domain Names (FQDNs) within DNS, but in practice they normally do 🌈 that's handy because it lets the realm portion of service names and the portion of NAIs be the same. it also allows DNS to be used for locating the host of a service while the transport channel between the client and the RP is being set up 💜
the application is responsible for figuring out which server it's going to talk to. GSS-API can help confirm that the server is the desired one, but it can't determine the name of the server to use 🙈 it's also up to the application to decide how much of the service-identifying information needs to be validated by the ABFAB system. the info that needs validating is used to construct the service name passed into the GSS-EAP mechanism, and what gets validated depends on (1) what information the client provided and (2) what information is considered significant. if the client only cares about getting a specific service, it doesn't need to validate the host and realm providing that service ✨
applications can look up information about service providers in DNS. Service Records (SRVs) [RFC2782] and Naming Authority Pointer (NAPTR) [RFC3401] records help find a host that provides a service, but whether DNSSEC is needed on those queries depends on how the info will be used 🔐 if the host name returned won't be validated by EAP channel binding because only the service is being validated, then DNSSEC [RFC4033] is not required. however, if the host name is going to be validated by EAP channel binding, then DNSSEC needs to be used to make sure the correct host name gets validated. in general, if the info returned from a DNS query is going to be validated, it needs to be obtained in a secure manner 🎀
one more issue for host-based service names: they don't work ideally when different instances of a service run on different ports 🙃 if the services are equivalent, it doesn't matter. but if there are substantial differences in the quality of the service, that information needs to be part of the validation process. and if the info being validated is just a host name without a port, that's not going to be a successful strategy 💀
3.4. Additional GSS-API Services ✨
okay so GSS-API gives you per-message security services 🔒 that can cover confidentiality and/or integrity 💅. Some IETF protocols, like NFS and SSH, totally take advantage of these, so GSS-EAP needs to support them too 🫶. Just like with mutual authentication 🤝, per-message security services limit which EAP methods you can use to the ones that generate a Master Session Key (MSK) 👑. Any EAP method that produces an MSK can support per-message security services as described in [RFC2743] ✨.
GSS-API also gives you a pseudorandom function 🎲. It spits out a pseudorandom sequence, using the shared session key as the seed for the bytes it generates. So the initiator and the acceptor can both run the same algorithm and end up with the same key value 🔑. An application can use this to generate keys or other shared secrets for use in other places in the protocol 🧚.
This is kinda similar in concept to the mechanism (formerly known as "TLS Extractors") described in [RFC5705] 🦋. No current IETF protocols require this feature yet, but non-IETF protocols are expected to take advantage of it in the near future 👀. Plus, a bunch of protocols have found the mechanism in [RFC5705] useful, so it's highly probable that IETF protocols may also start using this feature too 💖.
4. Privacy Considerations 🕵️
ABFAB is built for federated authentication and for sending identity information securely between entities, so obviously privacy needs serious care 🙈. The potential for privacy violations is real, bestie 😭.
This section looks at the privacy-related information in this document, summing up the entities involved in ABFAB communications and how much exposure each one has to identity information. For these privacy considerations, we use terminology and ideas from [RFC6973] 📚.
Note that ABFAB's architecture is built on several existing technologies and protocols at its core, and detailed privacy discussion of those topics is not covered here 🫣. This section instead focuses on the privacy considerations specifically related to the overall architecture and usage of ABFAB 💜.
+--------+ +---------------+ +--------------+
| Client | <---> | RP | <---> | AAA Client |
+--------+ +---------------+ +--------------+
^
|
v
+---------------+ +----------------+
| SAML Server | | AAA Proxy |
+---------------+ | (or Proxies) |
^ +----------------+
| ^
| |
v v
+------------+ +---------------+ +--------------+
| EAP Server | <---> | IdP | <---> | AAA Server |
+------------+ +---------------+ +--------------+
Figure 4: Entities and Data Flow4.1. Entities and Their Roles
Sorting the ABFAB entities in Figure 4 into the taxonomy from [RFC6973] is kinda complicated 🥴, because each entity's role changes during the different phases of ABFAB communications. The three main phases that matter are the client-to-RP communication phase, the client-to-IdP (via the Federation Substrate) communication phase, and the IdP-to-RP (via the Federation Substrate) communication phase 🌸.
In the client-to-RP communication phase, we have:
- Initiator: Client.
- Observers: Client, RP.
- Recipient: RP.
In the client-to-IdP (via the Federation Substrate) communication phase, we have:
- Initiator: Client.
- Observers: Client, RP, AAA Client, AAA Proxy (or Proxies), AAA Server, IdP.
- Recipient: IdP.
In the IdP-to-RP (via the Federation Substrate) communication phase, we have:
- Initiator: RP.
- Observers: IdP, AAA Server, AAA Proxy (or Proxies), AAA Client, RP.
- Recipient: IdP.
Eavesdroppers and attackers can lurk on any or all of the communication links between the entities shown in Figure 4 👀🔥.
The various entities in the system might also collude, or be coerced into colluding 😬. Some of the big collusions to look at are:
- If two RPs are colluding, they have the information available to both nodes. This can be analyzed as if a single RP were offering multiple services.
- If an RP and a AAA proxy are colluding, then the trust of the system is broken, as the RP would be able to lie about its own identity to the IdP. There is no known way to deal with this situation 💀.
- If multiple AAA proxies are colluding, they can be treated as a single node for analysis.
The Federation Substrate is made up of all of the AAA entities 🧩. In some cases, the AAA proxies may not exist, since the AAA client can talk directly to the AAA server. Specifications such as the Trust Router Protocol (https://www.ietf.org/proceedings/86/slides/slides-86-rtgarea-0.pdf) and RADIUS dynamic discovery [RFC7585] can be used to shorten the path between the AAA client and the AAA server, which stops these AAA proxies from being observers 🦋. However, even in these circumstances, there may still be AAA proxies in the path 🫠.
In Figure 4, the IdP has been divided into multiple logical pieces. In actual implementations, these pieces will frequently be tightly coupled 🧸. The links between these pieces give attackers and eavesdroppers the greatest opportunity to acquire information. But since they're all under the control of a single entity, they're also the easiest to lock down tight 🔐✨.
4.2. Privacy Aspects of ABFAB Communication Flows
The ABFAB architecture uses a few different types of data and identifiers 📦. The best way to understand them, and their potential privacy impacts, is to look at each phase of communication in ABFAB 🍓.
4.2.1. Client to RP
The flow of data between the client and the RP is split into two parts 🎀. The first part is all of the data exchanged as part of the ABFAB authentication process. The second part is all of the data exchanged after the authentication process has finished.
During the initial communication phase, the client sends an NAI (see [RFC7542]) to the RP 📨. Many EAP methods (but not all) let the client disclose an NAI to the RP in a form that includes only a realm component during this phase. That's the minimum amount of identity information ABFAB needs to work. It just indicates an IdP that the principal has a relationship with 🫶. EAP methods that don't allow this will necessarily also reveal an identifier for the principal in the IdP realm (for example, a username) 🙈.
The data shared during the initial communication phase may be protected by a channel protocol such as TLS. This prevents information from leaking to passive eavesdroppers 🛡️. However, an active attacker may still be able to set itself up as a man-in-the-middle 😱. The client may not be able to validate the certificates (if any) provided by the service, so it defers checking the identity of the RP until the ABFAB authentication protocol completes, using EAP channel binding rather than certificates 🔗.
The data exchanged after the authentication process can have privacy and authentication using the GSS-API services 💅. If the overall application protocol allows re-authentication, then the same privacy implications discussed in the previous paragraphs apply 🔁.
4.2.2. Client to IdP (via Federation Substrate)
This phase includes a secure TLS tunnel set up between the client and the IdP via the RP and the Federation Substrate 🚇. The RP starts the process using the realm information the client gave it. Once the tunnel is set up, it's used to send credentials to the IdP so the client can authenticate 🔑.
Various operational information is carried between the RP and the IdP over the AAA infrastructure, for example using RADIUS headers 📡. No end-to-end security is provided by AAA, which means all AAA entities on the path between the RP and IdP can eavesdrop on this information 👂. Some of it may form identifiers or explicit identity information:
- The RP knows the IP address of the client. It's possible that the RP could choose to expose this IP address by including it in a RADIUS header (for example, using the Calling-Station-Id). This is a privacy consideration to take into account for the application protocol 🧐.
- The EAP MSK is transported between the IdP and the RP over the AAA infrastructure, for example through RADIUS headers. This is a particularly important privacy consideration 😭, because any AAA proxy that has access to the EAP MSK can decrypt and eavesdrop on any traffic encrypted using that EAP MSK (that is, all communications between the client and RP). This problem can be mitigated if the application protocol sets up a secure tunnel between the client and the RP and performs a cryptographic binding between the tunnel and the EAP MSK 🔐.
- Related to the bullet point above, the AAA server has access to the material necessary to derive the session key. So the AAA server can observe any traffic encrypted between the client and RP 👀. This "feature" was chosen as a simplification and to make performance faster. If it was decided that this trade-off was not desirable for privacy and security reasons, then extensions to ABFAB that make use of techniques such as Diffie-Hellman key exchange would mitigate this 🦋.
The choice of EAP method has other potential privacy implications too 💭. For example, if the EAP method in use does not support mutual authentication, then there are no guarantees that the IdP is who it claims to be. That means the full NAI, including a username and a realm, might get sent to any entity masquerading as a particular IdP 😱🎭.
Note that ABFAB has not specified any AAA accounting requirements 🧾. Implementations that use the accounting portion of AAA should consider privacy appropriately when designing this aspect 💖.
4.2.3. IdP to RP (via Federation Substrate) 🦋
okay so this phase is where the IdP tells the RP whether the user's authentication worked or flopped 😬, and optionally it can also send over identity info about the principal ✨
it's the same deal as the last flow (client to IdP): all the operation info travels between the IdP and RP over the AAA infrastructure, and the same privacy considerations apply 🔒. BUT here's the spicy part 👀. In this flow, explicit identity info about the authenticated principal can get sent from the IdP to the RP. That info can ride along in RADIUS headers, or it can go via SAML [RFC7833] 💅. It can include protocol-specific identifiers, like SAML NameIDs, plus arbitrary attribute info about the principal 🧚. What gets released is controlled by policy on the IdP, so the IdP has the final say 👑.
when it's sent in RADIUS headers, every AAA entity on the path between the RP and IdP can eavesdrop 😭, unless extra protection is added (like TLS for RADIUS [RFC6614]). but when the info is sent using SAML as specified in [RFC7833], confidentiality of the information should be guaranteed 🔐, because [RFC7833] requires the use of TLS for RADIUS. that's the iconic move 💖.
4.3. Relationship between User and Entities
- Between user and IdP - The IdP is an entity the user will have a direct relationship with, created when the organization that operates the entity provisioned and exchanged the user's credentials. Privacy and data protection guarantees may form a part of this relationship 🫶.
- Between user and RP - The RP is an entity the user may or may not have a direct relationship with, depending on the service in question 🤷♀️. Some services may only be offered to those users where such a direct relationship exists (for particularly sensitive services, for example), while some may not require this and would instead be satisfied with basic federation trust guarantees between themselves and the IdP. This may well include the option that the user stays anonymous with respect to the RP (though, obviously, never anonymous to the IdP) 🎭. If attempting to preserve privacy via data minimization (Section 1), then the only attribute information about Individuals exposed to the RP should be attribute information that is strictly necessary for the operation of the service 🌸.
- Between user and Federation Substrate - The user is highly likely to have no knowledge of, or relationship with, any entities involved with the Federation Substrate (not that the IdP and/or RP may, however) 🙈. Knowledge of attribute information about Individuals for these entities is not necessary, so such information should be protected in a way that prevents the possibility of access to this information 🔐.
4.4. Accounting Information
alongside the core authentication and authorization that happen in AAA communications, accounting information about resource consumption may be delivered as part of the accounting exchange during the lifetime of the granted application session 📊✨.
4.5. Collection and Retention of Data and Identifiers
okay so in cases where RPs don't need to identify a particular Individual when that Individual wants to use their service, the ABFAB architecture makes anonymous or pseudonymous access possible 🦋. Which means data and identifiers other than pseudonyms and unlinkable attribute information don't need to be stored and retained 🧹.
BUT if RPs do need the ability to identify a particular Individual (e.g., so they can link this identity info to a particular account in their service, or because identity info is required for audit purposes) 📋, the service will need to collect and store that info, and keep it for as long as they need it. De-provisioning those accounts and info is out of scope for ABFAB 🙅♀️, but for privacy protection, it is obvious that any identifiers collected should be deleted once they're no longer needed 🗑️💜.
4.6. User Participation
in the ABFAB architecture, users are, by their very nature, active participants in sharing their identifiers 🙋♀️, since they start the communications exchange every single time they want to access a server ✨. They are not, however, involved in controlling the information about them that gets sent from the IdP to the RP for authorization purposes 😤. Instead, that's under the control of policy on the IdP. Because of how the AAA communication flows work, the current ABFAB architecture has no place for gaining user consent before information is released from the IdP to the RP 💀.
5. Security Considerations 🔥
This document describes the architecture for Application Bridging for Federated Access Beyond web (ABFAB), so security is the main focus, bestie 🛡️. Many of the items that are security considerations have already been discussed in Section 4 ("Privacy Considerations"), so make sure you read that section too 📖.
There are a LOT of places in this document where TLS is used 🔐. In some places (e.g., client to RP), anonymous connections can be used, but it is very important that TLS connections within the AAA infrastructure, and between the client and the IdP, be fully authenticated and, if using certificates, that revocation be checked as well 😱. When using anonymous connections between the client and the RP, all messages and data exchanged between those two entities will be visible to an active attacker 👀💀. In situations where the client is not yet on the network, the status_request extension [RFC6066] can be used to obtain revocation-checking data inside of the TLS protocol 🧚. Clients also need to get the trust anchor for the IdP configured correctly in order to prevent attacks. This is a difficult problem in general, and it's going to be even more difficult for kiosk environments 🖥️😭.
Selection of the EAP methods to be permitted by clients and IdPs is important 🎀. The use of a tunneling method such as TEAP [RFC7170] allows other EAP methods to be used while hiding the contents of those EAP exchanges from the RP and the AAA framework 🫣. When considering inner EAP methods, the considerations outlined in [RFC7029] about binding the inner and outer EAP methods need to be taken into account 🧩. Finally, one wants to have the ability to support channel binding in those cases where the client needs to validate that it is talking to the correct RP 🎯.
In places where SAML statements are used, RPs will generally be unable to validate signatures on the SAML statement 🙃. That's either because the signature has been stripped off by the IdP, or because the RP is unable to validate the binding between the signer, the key used to sign, and the realm represented by the IdP. For these reasons, it is required that IdPs do the necessary trust checking on the SAML statements, and that RPs can trust the AAA infrastructure to keep the SAML statements valid 🫶.
When a pseudonym is generated as a unique long-term identifier for a client by an IdP, care must be taken in the algorithm so that it cannot easily be reverse-engineered by the service provider 🕵️♀️. If it can be reverse-engineered, then the service provider can consult an oracle to determine whether a given unique long-term identifier is associated with a different known identifier 😳🚨.
6. References 📚
6.1. Normative References
[RFC2743] Linn, J., "Generic Security Service Application Program
Interface Version 2, Update 1", RFC 2743,
DOI 10.17487/RFC2743, January 2000,
<http://www.rfc-editor.org/info/rfc2743>.
[RFC2865] Rigney, C., Willens, S., Rubens, A., and W. Simpson,
"Remote Authentication Dial In User Service (RADIUS)",
RFC 2865, DOI 10.17487/RFC2865, June 2000,
<http://www.rfc-editor.org/info/rfc2865>.
[RFC3579] Aboba, B. and P. Calhoun, "RADIUS (Remote Authentication
Dial In User Service) Support For Extensible
Authentication Protocol (EAP)", RFC 3579,
DOI 10.17487/RFC3579, September 2003,
<http://www.rfc-editor.org/info/rfc3579>.
[RFC3748] Aboba, B., Blunk, L., Vollbrecht, J., Carlson, J., and H.
Levkowetz, Ed., "Extensible Authentication Protocol
(EAP)", RFC 3748, DOI 10.17487/RFC3748, June 2004,
<http://www.rfc-editor.org/info/rfc3748>.
[RFC4072] Eronen, P., Ed., Hiller, T., and G. Zorn, "Diameter
Extensible Authentication Protocol (EAP) Application",
RFC 4072, DOI 10.17487/RFC4072, August 2005,
<http://www.rfc-editor.org/info/rfc4072>.
[RFC6677] Hartman, S., Ed., Clancy, T., and K. Hoeper, "Channel-
Binding Support for Extensible Authentication Protocol
(EAP) Methods", RFC 6677, DOI 10.17487/RFC6677, July 2012,
<http://www.rfc-editor.org/info/rfc6677>.
[RFC7055] Hartman, S., Ed., and J. Howlett, "A GSS-API Mechanism for
the Extensible Authentication Protocol", RFC 7055,
DOI 10.17487/RFC7055, December 2013,
<http://www.rfc-editor.org/info/rfc7055>.
[RFC7542] DeKok, A., "The Network Access Identifier", RFC 7542,
DOI 10.17487/RFC7542, May 2015,
<http://www.rfc-editor.org/info/rfc7542>.
[RFC7833] Howlett, J., Hartman, S., and A. Perez-Mendez, Ed., "A
RADIUS Attribute, Binding, Profiles, Name Identifier
Format, and Confirmation Methods for the Security
Assertion Markup Language (SAML)", RFC 7833,
DOI 10.17487/RFC7833, May 2016,
<http://www.rfc-editor.org/info/rfc7833>.6.2. Informative References 📚🔍
Okay, so these are the "for your reading list" refs, bestie 📚 Nothing here is required to implement the spec, but they're the receipts behind the ideas. Here they are, in full:
- [NIST-SP.800-63-2] Burr, W., Dodson, D., Newton, E., Perlner, R., Polk, W., Gupta, S., and E. Nabbus, "Electronic Authentication Guideline", NIST Special Publication 800-63-2, August 2013,
<http://dx.doi.org/10.6028/NIST.SP.800-63-2>. - [OASIS.saml-core-2.0-os] Cantor, S., Kemp, J., Philpott, R., and E. Maler, "Assertions and Protocols for the OASIS Security Assertion Markup Language (SAML) V2.0", OASIS Standard saml-core-2.0-os, March 2005,
<http://docs.oasis-open.org/security/saml/v2.0/saml-core-2.0-os.pdf>. - [RFC1964] Linn, J., "The Kerberos Version 5 GSS-API Mechanism", RFC 1964, DOI 10.17487/RFC1964, June 1996,
<http://www.rfc-editor.org/info/rfc1964>. - [RFC2782] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for specifying the location of services (DNS SRV)", RFC 2782, DOI 10.17487/RFC2782, February 2000,
<http://www.rfc-editor.org/info/rfc2782>. - [RFC3401] Mealling, M., "Dynamic Delegation Discovery System (DDDS) Part One: The Comprehensive DDDS", RFC 3401, DOI 10.17487/RFC3401, October 2002,
<http://www.rfc-editor.org/info/rfc3401>. - [RFC3645] Kwan, S., Garg, P., Gilroy, J., Esibov, L., Westhead, J., and R. Hall, "Generic Security Service Algorithm for Secret Key Transaction Authentication for DNS (GSS-TSIG)", RFC 3645, DOI 10.17487/RFC3645, October 2003,
<http://www.rfc-editor.org/info/rfc3645>. - [RFC4033] Arends, R., Austein, R., Larson, M., Massey, D., and S. Rose, "DNS Security Introduction and Requirements", RFC 4033, DOI 10.17487/RFC4033, March 2005,
<http://www.rfc-editor.org/info/rfc4033>. - [RFC4422] Melnikov, A., Ed., and K. Zeilenga, Ed., "Simple Authentication and Security Layer (SASL)", RFC 4422, DOI 10.17487/RFC4422, June 2006,
<http://www.rfc-editor.org/info/rfc4422>. - [RFC4462] Hutzelman, J., Salowey, J., Galbraith, J., and V. Welch, "Generic Security Service Application Program Interface (GSS-API) Authentication and Key Exchange for the Secure Shell (SSH) Protocol", RFC 4462, DOI 10.17487/RFC4462, May 2006,
<http://www.rfc-editor.org/info/rfc4462>. - [RFC5056] Williams, N., "On the Use of Channel Bindings to Secure Channels", RFC 5056, DOI 10.17487/RFC5056, November 2007,
<http://www.rfc-editor.org/info/rfc5056>. - [RFC5080] Nelson, D. and A. DeKok, "Common Remote Authentication Dial In User Service (RADIUS) Implementation Issues and Suggested Fixes", RFC 5080, DOI 10.17487/RFC5080, December 2007,
<http://www.rfc-editor.org/info/rfc5080>. - [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security (TLS) Protocol Version 1.2", RFC 5246, DOI 10.17487/RFC5246, August 2008,
<http://www.rfc-editor.org/info/rfc5246>. - [RFC5705] Rescorla, E., "Keying Material Exporters for Transport Layer Security (TLS)", RFC 5705, DOI 10.17487/RFC5705, March 2010,
<http://www.rfc-editor.org/info/rfc5705>. - [RFC5801] Josefsson, S. and N. Williams, "Using Generic Security Service Application Program Interface (GSS-API) Mechanisms in Simple Authentication and Security Layer (SASL): The GS2 Mechanism Family", RFC 5801, DOI 10.17487/RFC5801, July 2010,
<http://www.rfc-editor.org/info/rfc5801>. - [RFC6066] Eastlake 3rd, D., "Transport Layer Security (TLS) Extensions: Extension Definitions", RFC 6066, DOI 10.17487/RFC6066, January 2011,
<http://www.rfc-editor.org/info/rfc6066>. - [RFC6614] Winter, S., McCauley, M., Venaas, S., and K. Wierenga, "Transport Layer Security (TLS) Encryption for RADIUS", RFC 6614, DOI 10.17487/RFC6614, May 2012,
<http://www.rfc-editor.org/info/rfc6614>. - [RFC6733] Fajardo, V., Ed., Arkko, J., Loughney, J., and G. Zorn, Ed., "Diameter Base Protocol", RFC 6733, DOI 10.17487/RFC6733, October 2012,
<http://www.rfc-editor.org/info/rfc6733>. - [RFC6749] Hardt, D., Ed., "The OAuth 2.0 Authorization Framework", RFC 6749, DOI 10.17487/RFC6749, October 2012,
<http://www.rfc-editor.org/info/rfc6749>. - [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., Morris, J., Hansen, M., and R. Smith, "Privacy Considerations for Internet Protocols", RFC 6973, DOI 10.17487/RFC6973, July 2013,
<http://www.rfc-editor.org/info/rfc6973>. - [RFC7029] Hartman, S., Wasserman, M., and D. Zhang, "Extensible Authentication Protocol (EAP) Mutual Cryptographic Binding", RFC 7029, DOI 10.17487/RFC7029, October 2013,
<http://www.rfc-editor.org/info/rfc7029>. - [RFC7170] Zhou, H., Cam-Winget, N., Salowey, J., and S. Hanna, "Tunnel Extensible Authentication Protocol (TEAP) Version 1", RFC 7170, DOI 10.17487/RFC7170, May 2014,
<http://www.rfc-editor.org/info/rfc7170>. - [RFC7360] DeKok, A., "Datagram Transport Layer Security (DTLS) as a Transport Layer for RADIUS", RFC 7360, DOI 10.17487/RFC7360, September 2014,
<http://www.rfc-editor.org/info/rfc7360>. - [RFC7499] Perez-Mendez, A., Ed., Marin-Lopez, R., Pereniguez-Garcia, F., Lopez-Millan, G., Lopez, D., and A. DeKok, "Support of Fragmentation of RADIUS Packets", RFC 7499, DOI 10.17487/RFC7499, April 2015,
<http://www.rfc-editor.org/info/rfc7499>. - [RFC7530] Haynes, T., Ed., and D. Noveck, Ed., "Network File System (NFS) Version 4 Protocol", RFC 7530, DOI 10.17487/RFC7530, March 2015,
<http://www.rfc-editor.org/info/rfc7530>. - [RFC7585] Winter, S. and M. McCauley, "Dynamic Peer Discovery for RADIUS/TLS and RADIUS/DTLS Based on the Network Access Identifier (NAI)", RFC 7585, DOI 10.17487/RFC7585, October 2015,
<http://www.rfc-editor.org/info/rfc7585>. - [WS-TRUST] Lawrence, K., Kaler, C., Nadalin, A., Goodner, M., Gudgin, M., Turner, D., Barbir, A., and H. Granqvist, "WS-Trust 1.4", OASIS Standard ws-trust-2012-04, April 2012,
<http://docs.oasis-open.org/ws-sx/ws-trust/v1.4/ws-trust.html>.
Acknowledgments 🫶
Huge thank-yous to Mayutan Arumaithurai, Klaas Wierenga, and Rhys Smith for their feedback 💖 Also to Eve Maler, Nicolas Williams, Bob Morgan, Scott Cantor, Jim Fenton, Paul Leach, and Luke Howard for their feedback on the federation terminology question 🌈
Klaas Wierenga also reviewed the first draft version of this document, and we're grateful for that 🎀 Eliot Lear gets a shoutout too for his work on the early draft versions 🧚
Authors' Addresses 💌
Josh Howlett
Jisc
Lumen House, Library Avenue, Harwell
Oxford OX11 0SG
United Kingdom
Phone: +44 1235 822363
Email: Josh.Howlett@ja.net
Sam Hartman
Painless Security
Email: hartmans-ietf@mit.edu
Hannes Tschofenig
ARM Ltd.
110 Fulbourn Road
Cambridge CB1 9NJ
United Kingdom
Email: Hannes.tschofenig@gmx.net
URI: http://www.tschofenig.priv.at
Jim Schaad
August Cellars
Email: ietf@augustcellars.comThat's the whole RFC, bestie! 🦋👑
📦 where this page comes from
- ietf: rfc7831 2016-05
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms