[RFCs/IDs] [Plain Text] [From draft-arkko-pppext-eap-aka]
INFORMATIONAL
Errata
Network Working Group J. Arkko
Request for Comments: 4187 Ericsson
Category: Informational H. Haverinen
Nokia
January 2006
Extensible Authentication Protocol Method for 3rd Generation
Authentication and Key Agreement (EAP-AKA)
Status of This Memo
This memo provides information for the Internet community. It does
not specify an Internet standard of any kind. Distribution of this
memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (2006).
IESG Note
The EAP-AKA protocol was developed by 3GPP. The documentation of
EAP-AKA is provided as information to the Internet community. While
the EAP WG has verified that EAP-AKA is compatible with EAP as
defined in RFC 3748, no other review has been done, including
validation of the security claims. The IETF has also not reviewed
the security of the underlying UMTS AKA algorithms.
Abstract
This document specifies an Extensible Authentication Protocol (EAP)
mechanism for authentication and session key distribution that uses
the Authentication and Key Agreement (AKA) mechanism. AKA is used in
the 3rd generation mobile networks Universal Mobile
Telecommunications System (UMTS) and CDMA2000. AKA is based on
symmetric keys, and typically runs in a Subscriber Identity Module,
which is a UMTS Subscriber Identity Module, USIM, or a (Removable)
User Identity Module, (R)UIM, similar to a smart card.
EAP-AKA includes optional identity privacy support, optional result
indications, and an optional fast re-authentication procedure.
Arkko & Haverinen Informational [Page 1]
RFC 4187 EAP-AKA Authentication January 2006
Table of Contents
1. Introduction and Motivation .....................................4
2. Terms and Conventions Used in This Document .....................5
3. Protocol Overview ...............................................9
4. Operation ......................................................15
4.1. Identity Management .......................................15
4.1.1. Format, Generation, and Usage of Peer Identities ...15
4.1.2. Communicating the Peer Identity to the Server ......21
4.1.3. Choice of Identity for the EAP-Response/Identity ...23
4.1.4. Server Operation in the Beginning of
EAP-AKA Exchange ...................................23
4.1.5. Processing of EAP-Request/AKA-Identity by
the Peer ...........................................24
4.1.6. Attacks against Identity Privacy ...................25
4.1.7. Processing of AT_IDENTITY by the Server ............26
4.2. Message Sequence Examples (Informative) ...................27
4.2.1. Usage of AT_ANY_ID_REQ .............................27
4.2.2. Fall Back on Full Authentication ...................28
4.2.3. Requesting the Permanent Identity 1 ................29
4.2.4. Requesting the Permanent Identity 2 ................30
4.2.5. Three EAP/AKA-Identity Round Trips .................30
5. Fast Re-Authentication .........................................32
5.1. General ...................................................32
5.2. Comparison to AKA .........................................33
5.3. Fast Re-Authentication Identity ...........................33
5.4. Fast Re-Authentication Procedure ..........................35
5.5. Fast Re-Authentication Procedure when Counter is
Too Small .................................................37
6. EAP-AKA Notifications ..........................................38
6.1. General ...................................................38
6.2. Result Indications ........................................39
6.3. Error Cases ...............................................40
6.3.1. Peer Operation .....................................41
6.3.2. Server Operation ...................................41
6.3.3. EAP-Failure ........................................42
6.3.4. EAP-Success ........................................42
7. Key Generation .................................................43
8. Message Format and Protocol Extensibility ......................45
8.1. Message Format ............................................45
8.2. Protocol Extensibility ....................................47
9. Messages .......................................................48
9.1. EAP-Request/AKA-Identity ..................................48
9.2. EAP-Response/AKA-Identity .................................48
9.3. EAP-Request/AKA-Challenge .................................49
9.4. EAP-Response/AKA-Challenge ................................49
9.5. EAP-Response/AKA-Authentication-Reject ....................50
9.6. EAP-Response/AKA-Synchronization-Failure ..................50
Arkko & Haverinen Informational [Page 2]
RFC 4187 EAP-AKA Authentication January 2006
9.7. EAP-Request/AKA-Reauthentication ..........................50
9.8. EAP-Response/AKA-Reauthentication .........................51
9.9. EAP-Response/AKA-Client-Error .............................52
9.10. EAP-Request/AKA-Notification .............................52
9.11. EAP-Response/AKA-Notification ............................52
10. Attributes ....................................................53
10.1. Table of Attributes ......................................53
10.2. AT_PERMANENT_ID_REQ ......................................54
10.3. AT_ANY_ID_REQ ............................................54
10.4. AT_FULLAUTH_ID_REQ .......................................54
10.5. AT_IDENTITY ..............................................55
10.6. AT_RAND ..................................................55
10.7. AT_AUTN ..................................................56
10.8. AT_RES ...................................................56
10.9. AT_AUTS ..................................................57
10.10. AT_NEXT_PSEUDONYM .......................................57
10.11. AT_NEXT_REAUTH_ID .......................................58
10.12. AT_IV, AT_ENCR_DATA, and AT_PADDING .....................58
10.13. AT_CHECKCODE ............................................60
10.14. AT_RESULT_IND ...........................................62
10.15. AT_MAC ..................................................63
10.16. AT_COUNTER ..............................................64
10.17. AT_COUNTER_TOO_SMALL ....................................64
10.18. AT_NONCE_S ..............................................65
10.19. AT_NOTIFICATION .........................................65
10.20. AT_CLIENT_ERROR_CODE ....................................66
11. IANA and Protocol Numbering Considerations ....................66
12. Security Considerations .......................................68
12.1. Identity Protection ......................................69
12.2. Mutual Authentication ....................................69
12.3. Flooding the Authentication Centre .......................69
12.4. Key Derivation ...........................................70
12.5. Brute-Force and Dictionary Attacks .......................70
12.6. Protection, Replay Protection, and Confidentiality .......70
12.7. Negotiation Attacks ......................................71
12.8. Protected Result Indications .............................72
12.9. Man-in-the-Middle Attacks ................................72
12.10. Generating Random Numbers ...............................73
13. Security Claims ...............................................73
14. Acknowledgements and Contributions ............................74
15. References ....................................................74
15.1. Normative References .....................................74
15.2. Informative References ...................................76
Appendix A. Pseudo-Random Number Generator .......................77
Arkko & Haverinen Informational [Page 3]
RFC 4187 EAP-AKA Authentication January 2006
1. Introduction and Motivation
This document specifies an Extensible Authentication Protocol (EAP)
mechanism for authentication and session key distribution that uses
the 3rd generation Authentication and Key Agreement mechanism,
specified for Universal Mobile Telecommunications System (UMTS) in
[TS33.102] and for CDMA2000 in [S.S0055-A]. UMTS and CDMA2000 are
global 3rd generation mobile network standards that use the same AKA
mechanism.
2nd generation mobile networks and 3rd generation mobile networks use
different authentication and key agreement mechanisms. The Global
System for Mobile communications (GSM) is a 2nd generation mobile
network standard, and EAP-SIM [EAP-SIM] specifies an EAP mechanism
that is based on the GSM authentication and key agreement primitives.
AKA is based on challenge-response mechanisms and symmetric
cryptography. AKA typically runs in a UMTS Subscriber Identity
Module (USIM) or a CDMA2000 (Removable) User Identity Module
((R)UIM). In this document, both modules are referred to as identity
modules. Compared to the 2nd generation mechanisms such as GSM AKA,
the 3rd generation AKA provides substantially longer key lengths and
mutual authentication.
The introduction of AKA inside EAP allows several new applications.
These include the following:
o The use of the AKA also as a secure PPP authentication method in
devices that already contain an identity module.
o The use of the 3rd generation mobile network authentication
infrastructure in the context of wireless LANs
o Relying on AKA and the existing infrastructure in a seamless way
with any other technology that can use EAP.
AKA works in the following manner:
o The identity module and the home environment have agreed on a
secret key beforehand. (The "home environment" refers to the home
operator's authentication network infrastructure.)
o The actual authentication process starts by having the home
environment produce an authentication vector, based on the secret
key and a sequence number. The authentication vector contains a
random part RAND, an authenticator part AUTN used for
authenticating the network to the identity module, an expected
result part XRES, a 128-bit session key for integrity check IK,
and a 128-bit session key for encryption CK.
Arkko & Haverinen Informational [Page 4]
RFC 4187 EAP-AKA Authentication January 2006
o The RAND and the AUTN are delivered to the identity module.
o The identity module verifies the AUTN, again based on the secret
key and the sequence number. If this process is successful (the
AUTN is valid and the sequence number used to generate AUTN is
within the correct range), the identity module produces an
authentication result RES and sends it to the home environment.
o The home environment verifies the correct result from the identity
module. If the result is correct, IK and CK can be used to
protect further communications between the identity module and the
home environment.
When verifying AUTN, the identity module may detect that the sequence
number the network uses is not within the correct range. In this
case, the identity module calculates a sequence number
synchronization parameter AUTS and sends it to the network. AKA
authentication may then be retried with a new authentication vector
generated using the synchronized sequence number.
For a specification of the AKA mechanisms and how the cryptographic
values AUTN, RES, IK, CK and AUTS are calculated, see [TS33.102] for
UMTS and [S.S0055-A] for CDMA2000.
In EAP-AKA, the EAP server node obtains the authentication vectors,
compares RES and XRES, and uses CK and IK in key derivation.
In the 3rd generation mobile networks, AKA is used for both radio
network authentication and IP multimedia service authentication
purposes. Different user identities and formats are used for these;
the radio network uses the International Mobile Subscriber Identifier
(IMSI), whereas the IP multimedia service uses the Network Access
Identifier (NAI) [RFC4282].
2. Terms and Conventions Used in This Document
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in [RFC2119].
The terms and abbreviations "authenticator", "backend authentication
server", "EAP server", "peer", "Silently Discard", "Master Session
Key (MSK)", and "Extended Master Session Key (EMSK)" in this document
are to be interpreted as described in [RFC3748].
This document frequently uses the following terms and abbreviations.
The AKA parameters are specified in detail in [TS33.102] for UMTS and
[S.S0055-A] for CDMA2000.
Arkko & Haverinen Informational [Page 5]
RFC 4187 EAP-AKA Authentication January 2006
AAA protocol
Authentication, Authorization and Accounting protocol
AKA
Authentication and Key Agreement
AuC
Authentication Centre. The mobile network element that can
authenticate subscribers in the mobile networks.
AUTN
AKA parameter. AUTN is an authentication value generated by
the AuC, which, together with the RAND, authenticates the
server to the peer, 128 bits.
AUTS
AKA parameter. A value generated by the peer upon
experiencing a synchronization failure, 112 bits.
EAP
Extensible Authentication Protocol [RFC3748]
Fast Re-Authentication
An EAP-AKA authentication exchange that is based on keys
derived upon a preceding full authentication exchange. The
3rd Generation AKA is not used in the fast re-authentication
procedure.
Fast Re-Authentication Identity
A fast re-authentication identity of the peer, including an
NAI realm portion in environments where a realm is used.
Used on re-authentication only.
Fast Re-Authentication Username
The username portion of fast re-authentication identity,
i.e., not including any realm portions.
Arkko & Haverinen Informational [Page 6]
RFC 4187 EAP-AKA Authentication January 2006
Full Authentication
An EAP-AKA authentication exchange that is based on the
3rd Generation AKA procedure.
GSM
Global System for Mobile communications.
NAI
Network Access Identifier [RFC4282]
Identity Module
Identity module is used in this document to refer to the
part of the mobile device that contains authentication and
key agreement primitives. The identity module may be an
integral part of the mobile device or it may be an application
on a smart card distributed by a mobile operator. USIM and
(R)UIM are identity modules.
Nonce
A value that is used at most once or that is never repeated
within the same cryptographic context. In general, a nonce can
be predictable (e.g., a counter) or unpredictable (e.g., a
random value). Because some cryptographic properties may
depend on the randomness of the nonce, attention should be paid
to whether a nonce is required to be random or not. In this
document, the term nonce is only used to denote random nonces,
and it is not used to denote counters.
Permanent Identity
The permanent identity of the peer, including an NAI realm
portion in environments where a realm is used. The permanent
identity is usually based on the IMSI. Used on full
authentication only.
Permanent Username
The username portion of permanent identity, i.e., not including
any realm portions.
Arkko & Haverinen Informational [Page 7]
RFC 4187 EAP-AKA Authentication January 2006
Pseudonym Identity
A pseudonym identity of the peer, including an NAI realm
portion in environments where a realm is used. Used on full
authentication only.
Pseudonym Username
The username portion of pseudonym identity, i.e., not including
any realm portions.
RAND
An AKA parameter. Random number generated by the AuC,
128 bits.
RES
Authentication result from the peer, which, together with
the RAND, authenticates the peer to the server,
128 bits.
(R)UIM
CDMA2000 (Removable) User Identity Module. (R)UIM is an
application that is resident on devices such as smart cards,
which may be fixed in the terminal or distributed by CDMA2000
operators (when removable).
SQN
An AKA parameter. Sequence number used in the authentication
process, 48 bits.
SIM
Subscriber Identity Module. The SIM is traditionally a smart
card distributed by a GSM operator.
SRES
The authentication result parameter in GSM, corresponds to
the RES parameter in 3G AKA, 32 bits.
Arkko & Haverinen Informational [Page 8]
RFC 4187 EAP-AKA Authentication January 2006
UAK
UIM Authentication Key, used in CDMA2000 AKA. Both the
identity module and the network can optionally generate the UAK
during the AKA computation in CDMA2000. UAK is not used in
this version of EAP-AKA.
UIM
Please see (R)UIM.
USIM
UMTS Subscriber Identity Module. USIM is an application that
is resident on devices such as smart cards distributed by UMTS
operators.
3. Protocol Overview
Figure 1 shows the basic, successful full authentication exchange in
EAP-AKA, when optional result indications are not used. The
authenticator typically communicates with an EAP server that is
located on a backend authentication server using an AAA protocol.
The authenticator shown in the figure is often simply relaying EAP
messages to and from the EAP server, but these backend AAA
communications are not shown. At the minimum, EAP-AKA uses two
roundtrips to authenticate and authorize the peer and generate
session keys. As in other EAP schemes, an identity request/response
message pair is usually exchanged first. On full authentication, the
peer's identity response includes either the user's International
Mobile Subscriber Identity (IMSI), or a temporary identity
(pseudonym) if identity privacy is in effect, as specified in
Section 4.1. (As specified in [RFC3748], the initial identity
request is not required, and MAY be bypassed in cases where the
network can presume the identity, such as when using leased lines,
dedicated dial-ups, etc. Please see Section 4.1.2 for specification
of how to obtain the identity via EAP AKA messages.)
After obtaining the subscriber identity, the EAP server obtains an
authentication vector (RAND, AUTN, RES, CK, IK) for use in
authenticating the subscriber. From the vector, the EAP server
derives the keying material, as specified in Section 6.4. The vector
may be obtained by contacting an Authentication Centre (AuC) on the
mobile network; for example, per UMTS specifications, several vectors
may be obtained at a time. Vectors may be stored in the EAP server
for use at a later time, but they may not be reused.
Arkko & Haverinen Informational [Page 9]
RFC 4187 EAP-AKA Authentication January 2006
In CDMA2000, the vector may include a sixth value called the User
Identity Module Authentication Key (UAK). This key is not used in
EAP-AKA.
Next, the EAP server starts the actual AKA protocol by sending an
EAP-Request/AKA-Challenge message. EAP-AKA packets encapsulate
parameters in attributes, encoded in a Type, Length, Value format.
The packet format and the use of attributes are specified in
Section 8. The EAP-Request/AKA-Challenge message contains a RAND
random number (AT_RAND), a network authentication token (AT_AUTN),
and a message authentication code (AT_MAC). The EAP-Request/
AKA-Challenge message MAY optionally contain encrypted data, which is
used for identity privacy and fast re-authentication support, as
described in Section 4.1. The AT_MAC attribute contains a message
authentication code covering the EAP packet. The encrypted data is
not shown in the figures of this section.
The peer runs the AKA algorithm (typically using an identity module)
and verifies the AUTN. If this is successful, the peer is talking to
a legitimate EAP server and proceeds to send the EAP-Response/
AKA-Challenge. This message contains a result parameter that allows
the EAP server, in turn, to authenticate the peer, and the AT_MAC
attribute to integrity protect the EAP message.
The EAP server verifies that the RES and the MAC in the EAP-Response/
AKA-Challenge packet are correct. Because protected success
indications are not used in this example, the EAP server sends the
EAP-Success packet, indicating that the authentication was
successful. (Protected success indications are discussed in
Section 6.2.) The EAP server may also include derived keying
material in the message it sends to the authenticator. The peer has
derived the same keying material, so the authenticator does not
forward the keying material to the peer along with EAP-Success.
Arkko & Haverinen Informational [Page 10]
RFC 4187 EAP-AKA Authentication January 2006
Peer Authenticator
| EAP-Request/Identity |
|<------------------------------------------------------|
| |
| EAP-Response/Identity |
| (Includes user's NAI) |
|------------------------------------------------------>|
| +------------------------------+
| | Server runs AKA algorithms, |
| | generates RAND and AUTN. |
| +------------------------------+
| EAP-Request/AKA-Challenge |
| (AT_RAND, AT_AUTN, AT_MAC) |
|<------------------------------------------------------|
+-------------------------------------+ |
| Peer runs AKA algorithms, | |
| verifies AUTN and MAC, derives RES | |
| and session key | |
+-------------------------------------+ |
| EAP-Response/AKA-Challenge |
| (AT_RES, AT_MAC) |
|------------------------------------------------------>|
| +--------------------------------+
| | Server checks the given RES, |
| | and MAC and finds them correct.|
| +--------------------------------+
| EAP-Success |
|<------------------------------------------------------|
Figure 1: EAP-AKA full authentication procedure
Arkko & Haverinen Informational [Page 11]
RFC 4187 EAP-AKA Authentication January 2006
Figure 2 shows how the EAP server rejects the Peer due to a failed
authentication.
Peer Authenticator
| EAP-Request/Identity |
|<------------------------------------------------------|
| |
| EAP-Response/Identity |
| (Includes user's NAI) |
|------------------------------------------------------>|
| +------------------------------+
| | Server runs AKA algorithms, |
| | generates RAND and AUTN. |
| +------------------------------+
| EAP-Request/AKA-Challenge |
| (AT_RAND, AT_AUTN, AT_MAC) |
|<------------------------------------------------------|
+-------------------------------------+ |
| Peer runs AKA algorithms, | |
| possibly verifies AUTN, and sends an| |
| invalid response | |
+-------------------------------------+ |
| EAP-Response/AKA-Challenge |
| (AT_RES, AT_MAC) |
|------------------------------------------------------>|
| +------------------------------------------+
| | Server checks the given RES and the MAC, |
| | and finds one of them incorrect. |
| +------------------------------------------+
| EAP-Request/AKA-Notification |
|<------------------------------------------------------|
| EAP-Response/AKA-Notification |
|------------------------------------------------------>|
| EAP-Failure |
|<------------------------------------------------------|
Figure 2: Peer authentication fails
Arkko & Haverinen Informational [Page 12]
RFC 4187 EAP-AKA Authentication January 2006
Figure 3 shows the peer rejecting the AUTN of the EAP server.
The peer sends an explicit error message (EAP-Response/
AKA-Authentication-Reject) to the EAP server, as usual in AKA when
AUTN is incorrect. This allows the EAP server to produce the same
error statistics that AKA generally produces in UMTS or CDMA2000.
Peer Authenticator
| EAP-Request/Identity |
|<------------------------------------------------------|
| EAP-Response/Identity |
| (Includes user's NAI) |
|------------------------------------------------------>|
| +------------------------------+
| | Server runs AKA algorithms, |
| | generates RAND and a bad AUTN|
| +------------------------------+
| EAP-Request/AKA-Challenge |
| (AT_RAND, AT_AUTN, AT_MAC) |
|<------------------------------------------------------|
+-------------------------------------+ |
| Peer runs AKA algorithms | |
| and discovers AUTN that can not be | |
| verified | |
+-------------------------------------+ |
| EAP-Response/AKA-Authentication-Reject |
|------------------------------------------------------>|
| EAP-Failure |
|<------------------------------------------------------|
Figure 3: Network authentication fails
The AKA uses shared secrets between the Peer and the Peer's home
operator, together with a sequence number, to actually perform an
authentication. In certain circumstances, shown in Figure 4, it is
possible for the sequence numbers to get out of sequence.
Arkko & Haverinen Informational [Page 13]
RFC 4187 EAP-AKA Authentication January 2006
Peer Authenticator
| EAP-Request/Identity |
|<------------------------------------------------------|
| EAP-Response/Identity |
| (Includes user's NAI) |
|------------------------------------------------------>|
| +------------------------------+
| | Server runs AKA algorithms, |
| | generates RAND and AUTN. |
| +------------------------------+
| EAP-Request/AKA-Challenge |
| (AT_RAND, AT_AUTN, AT_MAC) |
|<------------------------------------------------------|
+-------------------------------------+ |
| Peer runs AKA algorithms | |
| and discovers AUTN that contains an | |
| inappropriate sequence number | |
+-------------------------------------+ |
| EAP-Response/AKA-Synchronization-Failure |
| (AT_AUTS) |
|------------------------------------------------------>|
| +---------------------------+
| | Perform resynchronization |
| | Using AUTS and |
| | the sent RAND |
| +---------------------------+
| |
Figure 4: Sequence number synchronization
After the resynchronization process has taken place in the server and
AAA side, the process continues by the server side sending a new
EAP-Request/AKA-Challenge message.
In addition to the full authentication scenarios described above,
EAP-AKA includes a fast re-authentication procedure, which is
specified in Section 5. Fast re-authentication is based on keys
derived on full authentication. If the peer has maintained state
information for re-authentication and wants to use fast
re-authentication, then the peer indicates this by using a specific
fast re-authentication identity instead of the permanent identity or
a pseudonym identity.
Arkko & Haverinen Informational [Page 14]
RFC 4187 EAP-AKA Authentication January 2006
4. Operation
4.1. Identity Management
4.1.1. Format, Generation, and Usage of Peer Identities
4.1.1.1. General
In the beginning of EAP authentication, the Authenticator or the EAP
server usually issues the EAP-Request/Identity packet to the peer.
The peer responds with EAP-Response/Identity, which contains the
user's identity. The formats of these packets are specified in
[RFC3748].
Subscribers of mobile networks are identified with the International
Mobile Subscriber Identity (IMSI) [TS23.003]. The IMSI is a string
of not more than 15 digits. It is composed of a Mobile Country Code
(MCC) of 3 digits, a Mobile Network Code (MNC) of 2 or 3 digits, and
a Mobile Subscriber Identification Number (MSIN) of not more than 10
digits. MCC and MNC uniquely identify the GSM operator and help
identify the AuC from which the authentication vectors need to be
retrieved for this subscriber.
Internet AAA protocols identify users with the Network Access
Identifier (NAI) [RFC4282]. When used in a roaming environment, the
NAI is composed of a username and a realm, separated with "@"
(username@realm). The username portion identifies the subscriber
within the realm.
This section specifies the peer identity format used in EAP-AKA. In
this document, the term identity or peer identity refers to the whole
identity string that is used to identify the peer. The peer identity
may include a realm portion. "Username" refers to the portion of the
peer identity that identifies the user, i.e., the username does not
include the realm portion.
4.1.1.2. Identity Privacy Support
EAP-AKA includes optional identity privacy (anonymity) support that
can be used to hide the cleartext permanent identity and thereby make
the subscriber's EAP exchanges untraceable to eavesdroppers. Because
the permanent identity never changes, revealing it would help
observers to track the user. The permanent identity is usually based
on the IMSI, which may further help the tracking, because the same
identifier may be used in other contexts as well. Identity privacy
is based on temporary identities, or pseudonyms, which are equivalent
Arkko & Haverinen Informational [Page 15]
RFC 4187 EAP-AKA Authentication January 2006
to but separate from the Temporary Mobile Subscriber Identities
(TMSI) that are used on cellular networks. Please see Section 12.1
for security considerations regarding identity privacy.
4.1.1.3. Username Types in EAP-AKA Identities
There are three types of usernames in EAP-AKA peer identities:
(1) Permanent usernames. For example,
0123456789098765@myoperator.com might be a valid permanent identity.
In this example, 0123456789098765 is the permanent username.
(2) Pseudonym usernames. For example, 2s7ah6n9q@myoperator.com might
be a valid pseudonym identity. In this example, 2s7ah6n9q is the
pseudonym username.
(3) Fast re-authentication usernames. For example,
43953754@myoperator.com might be a valid fast re-authentication
identity. In this case, 43953754 is the fast re-authentication
username. Unlike permanent usernames and pseudonym usernames, fast
re-authentication usernames are one-time identifiers, which are not
re-used across EAP exchanges.
The first two types of identities are used only on full
authentication, and the last type only on fast re-authentication.
When the optional identity privacy support is not used, the
non-pseudonym permanent identity is used on full authentication. The
fast re-authentication exchange is specified in Section 5.
4.1.1.4. Username Decoration
In some environments, the peer may need to decorate the identity by
prepending or appending the username with a string, in order to
indicate supplementary AAA routing information in addition to the NAI
realm. (The usage of an NAI realm portion is not considered to be
decoration.) Username decoration is out of the scope of this
document. However, it should be noted that username decoration might
prevent the server from recognizing a valid username. Hence,
although the peer MAY use username decoration in the identities that
the peer includes in EAP-Response/Identity, and although the EAP
server MAY accept a decorated peer username in this message, the peer
or the EAP server MUST NOT decorate any other peer identities that
are used in various EAP-AKA attributes. Only the identity used in
EAP-Response/Identity may be decorated.
Arkko & Haverinen Informational [Page 16]
RFC 4187 EAP-AKA Authentication January 2006
4.1.1.5. NAI Realm Portion
The peer MAY include a realm portion in the peer identity, as per the
NAI format. The use of a realm portion is not mandatory.
If a realm is used, the realm MAY be chosen by the subscriber's home
operator and it MAY be a configurable parameter in the EAP-AKA peer
implementation. In this case, the peer is typically configured with
the NAI realm of the home operator. Operators MAY reserve a specific
realm name for EAP-AKA users. This convention makes it easy to
recognize that the NAI identifies an AKA subscriber. Such a reserved
NAI realm may be useful as a hint of the first authentication method
to use during method negotiation. When the peer is using a pseudonym
username instead of the permanent username, the peer selects the
realm name portion similarly to how it selects the realm portion when
using the permanent username.
If no configured realm name is available, the peer MAY derive the
realm name from the MCC and MNC portions of the IMSI. A RECOMMENDED
way to derive the realm from the IMSI, using the realm
3gppnetwork.org, will be specified in [TS23.003].
Some old implementations derive the realm name from the IMSI by
concatenating "mnc", the MNC digits of IMSI, ".mcc", the MCC digits
of IMSI, and ".owlan.org". For example, if the IMSI is
123456789098765, and the MNC is three digits long, then the derived
realm name is "mnc456.mcc123.owlan.org". As there are no DNS servers
running at owlan.org, these realm names can only be used with
manually configured AAA routing. New implementations SHOULD use the
mechanism specified in [TS23.003] instead of owlan.org.
The IMSI is a string of digits without any explicit structure, so the
peer may not be able to determine the length of the MNC portion. If
the peer is not able to determine whether the MNC is two or three
digits long, the peer MAY use a 3-digit MNC. If the correct length
of the MNC is two, then the MNC used in the realm name includes the
first digit of MSIN. Hence, when configuring AAA networks for
operators that have 2-digit MNC's, the network SHOULD also be
prepared for realm names with incorrect 3-digit MNC's.
4.1.1.6. Format of the Permanent Username
The non-pseudonym permanent username SHOULD be derived from the IMSI.
In this case, the permanent username MUST be of the format "0" |
IMSI, where the character "|" denotes concatenation. In other words,
the first character of the username is the digit zero (ASCII value 30
hexadecimal), followed by the IMSI. The IMSI is an ASCII string that
consists of not more than 15 decimal digits (ASCII values between 30
Arkko & Haverinen Informational [Page 17]
RFC 4187 EAP-AKA Authentication January 2006
and 39 hexadecimal), one character per IMSI digit, in the order as
specified in [TS23.003]. For example, a permanent username derived
from the IMSI 295023820005424 would be encoded as the ASCII string
"0295023820005424" (byte values in hexadecimal notation: 30 32 39 35
30 32 33 38 32 30 30 30 35 34 32 34)
The EAP server MAY use the leading "0" as a hint to try EAP-AKA as
the first authentication method during method negotiation, rather
than using, for example, EAP-SIM. The EAP-AKA server MAY propose
EAP-AKA even if the leading character was not "0".
Alternatively, an implementation MAY choose a permanent username that
is not based on the IMSI. In this case the selection of the
username, its format, and its processing is out of the scope of this
document. In this case, the peer implementation MUST NOT prepend any
leading characters to the username.
4.1.1.7. Generating Pseudonyms and Fast Re-Authentication Identities by
the Server
Pseudonym usernames and fast re-authentication identities are
generated by the EAP server. The EAP server produces pseudonym
usernames and fast re-authentication identities in an
implementation-dependent manner. Only the EAP server needs to be
able to map the pseudonym username to the permanent identity, or to
recognize a fast re-authentication identity.
EAP-AKA includes no provisions to ensure that the same EAP server
that generated a pseudonym username will be used on the
authentication exchange when the pseudonym username is used. It is
recommended that the EAP servers implement some centralized mechanism
to allow all EAP servers of the home operator to map pseudonyms
generated by other severs to the permanent identity. If no such
mechanism is available, then the EAP server, failing to understand a
pseudonym issued by another server, can request the peer to send the
permanent identity.
When issuing a fast re-authentication identity, the EAP server may
include a realm name in the identity that will cause the fast
re-authentication request to be forwarded to the same EAP server.
When generating fast re-authentication identities, the server SHOULD
choose a fresh, new fast re-authentication identity that is different
from the previous ones that were used after the same full
authentication exchange. A full authentication exchange and the
associated fast re-authentication exchanges are referred to here as
the same "full authentication context". The fast re-authentication
identity SHOULD include a random component. The random component
Arkko & Haverinen Informational [Page 18]
RFC 4187 EAP-AKA Authentication January 2006
works as a full authentication context identifier. A context-
specific fast re-authentication identity can help the server to
detect whether its fast re-authentication state information matches
the peer's fast re-authentication state information (in other words,
whether the state information is from the same full authentication
exchange). The random component also makes the fast re-
authentication identities unpredictable, so an attacker cannot
initiate a fast re-authentication exchange to get the server's
EAP-Request/AKA-Reauthentication packet.
Transmitting pseudonyms and fast re-authentication identities from
the server to the peer is discussed in Section 4.1.1.8. The
pseudonym is transmitted as a username, without an NAI realm, and the
fast re-authentication identity is transmitted as a complete NAI,
including a realm portion if a realm is required. The realm is
included in the fast re-authentication identity in order to allow the
server to include a server-specific realm.
Regardless of construction method, the pseudonym username MUST
conform to the grammar specified for the username portion of an NAI.
Also, the fast re-authentication identity MUST conform to the NAI
grammar. The EAP servers that the subscribers of an operator can use
MUST ensure that the pseudonym usernames and the username portions
used in fast re-authentication identities that they generate are
unique.
In any case, it is necessary that permanent usernames, pseudonym
usernames, and fast re-authentication usernames are separate and
recognizable from each other. It is also desirable that EAP-SIM and
EAP-AKA usernames be recognizable from each other as an aid to the
server when deciding which method to offer.
In general, it is the task of the EAP server and the policies of its
administrator to ensure sufficient separation of the usernames.
Pseudonym usernames and fast re-authentication usernames are both
produced and used by the EAP server. The EAP server MUST compose
pseudonym usernames and fast re-authentication usernames so that it
can recognize if an NAI username is an EAP-AKA pseudonym username or
an EAP-AKA fast re-authentication username. For instance, when the
usernames have been derived from the IMSI, the server could use
different leading characters in the pseudonym usernames and fast
re-authentication usernames (e.g., the pseudonym could begin with a
leading "2" character). When mapping a fast re-authentication
identity to a permanent identity, the server SHOULD only examine the
username portion of the fast re-authentication identity and ignore
the realm portion of the identity.
Arkko & Haverinen Informational [Page 19]
RFC 4187 EAP-AKA Authentication January 2006
Because the peer may fail to save a pseudonym username that was sent
in an EAP-Request/AKA-Challenge (for example, due to malfunction),
the EAP server SHOULD maintain, at least, the most recently used
pseudonym username in addition to the most recently issued pseudonym
username. If the authentication exchange is not completed
successfully, then the server SHOULD NOT overwrite the pseudonym
username that was issued during the most recent successful
authentication exchange.
4.1.1.8. Transmitting Pseudonyms and Fast Re-Authentication Identities
to the Peer
The server transmits pseudonym usernames and fast re-authentication
identities to the peer in cipher, using the AT_ENCR_DATA attribute.
The EAP-Request/AKA-Challenge message MAY include an encrypted
pseudonym username and/or an encrypted fast re-authentication
identity in the value field of the AT_ENCR_DATA attribute. Because
identity privacy support and fast re-authentication are optional to
implement, the peer MAY ignore the AT_ENCR_DATA attribute and always
use the permanent identity. On fast re-authentication (discussed in
Section 5), the server MAY include a new, encrypted fast re-
authentication identity in the EAP-Request/AKA-Reauthentication
message.
On receipt of the EAP-Request/AKA-Challenge, the peer MAY decrypt the
encrypted data in AT_ENCR_DATA; and if a pseudonym username is
included, the peer may use the obtained pseudonym username on the
next full authentication. If a fast re-authentication identity is
included, then the peer MAY save it together with other fast re-
authentication state information, as discussed in Section 5, for the
next fast re-authentication.
If the peer does not receive a new pseudonym username in the
EAP-Request/AKA-Challenge message, the peer MAY use an old pseudonym
username instead of the permanent username on next full
authentication. The username portions of fast re-authentication
identities are one-time usernames, which the peer MUST NOT re-use.
When the peer uses a fast re-authentication identity in an EAP
exchange, the peer MUST discard the fast re-authentication identity
and not re-use it in another EAP authentication exchange, even if the
authentication exchange was not completed.
4.1.1.9. Usage of the Pseudonym by the Peer
When the optional identity privacy support is used on full
authentication, the peer MAY use a pseudonym username received as
part of a previous full authentication sequence as the username
Arkko & Haverinen Informational [Page 20]
RFC 4187 EAP-AKA Authentication January 2006
portion of the NAI. The peer MUST NOT modify the pseudonym username
received in AT_NEXT_PSEUDONYM. However, as discussed above, the peer
MAY need to decorate the username in some environments by appending
or prepending the username with a string that indicates supplementary
AAA routing information.
When using a pseudonym username in an environment where a realm
portion is used, the peer concatenates the received pseudonym
username with the "@" character and an NAI realm portion. The
selection of the NAI realm is discussed above. The peer can select
the realm portion similarly, regardless of whether it uses the
permanent username or a pseudonym username.
4.1.1.10. Usage of the Fast Re-Authentication Identity by the Peer
On fast re-authentication, the peer uses the fast re-authentication
identity received as part of the previous authentication sequence. A
new fast re-authentication identity may be delivered as part of both
full authentication and fast re-authentication. The peer MUST NOT
modify the username part of the fast re-authentication identity
received in AT_NEXT_REAUTH_ID, except in cases when username
decoration is required. Even in these cases, the "root" fast
re-authentication username must not be modified, but it may be
appended or prepended with another string.
4.1.2. Communicating the Peer Identity to the Server
4.1.2.1. General
The peer identity MAY be communicated to the server with the
EAP-Response/Identity message. This message MAY contain the
permanent identity, a pseudonym identity, or a fast re-authentication
identity. If the peer uses the permanent identity or a pseudonym
identity, which the server is able to map to the permanent identity,
then the authentication proceeds as discussed in the overview of
Section 3. If the peer uses a fast re-authentication identity, and
if the fast re-authentication identity matches with a valid fast
re-authentication identity maintained by the server, then a fast
re-authentication exchange is performed, as described in Section 5.
The peer identity can also be transmitted from the peer to the server
using EAP-AKA messages instead of EAP-Response/Identity. In this
case, the server includes an identity requesting attribute
(AT_ANY_ID_REQ, AT_FULLAUTH_ID_REQ or AT_PERMANENT_ID_REQ) in the
EAP-Request/AKA-Identity message; and the peer includes the
AT_IDENTITY attribute, which contains the peer's identity, in the
EAP-Response/AKA-Identity message. The AT_ANY_ID_REQ attribute is a
general identity requesting attribute, which the server uses if it
Arkko & Haverinen Informational [Page 21]
RFC 4187 EAP-AKA Authentication January 2006
does not specify which kind of an identity the peer should return in
AT_IDENTITY. The server uses the AT_FULLAUTH_ID_REQ attribute to
request either the permanent identity or a pseudonym identity. The
server uses the AT_PERMANENT_ID_REQ attribute to request that the
peer send its permanent identity. The EAP-Request/AKA-Challenge,
EAP-Response/AKA-Challenge, or the packets used on fast re-
authentication may optionally include the AT_CHECKCODE attribute,
which enables the protocol peers to ensure the integrity of the
AKA-Identity packets. AT_CHECKCODE is specified in Section 10.13.
The identity format in the AT_IDENTITY attribute is the same as in
the EAP-Response/Identity packet (except that identity decoration is
not allowed). The AT_IDENTITY attribute contains a permanent
identity, a pseudonym identity, or a fast re-authentication identity.
Please note that only the EAP-AKA peer and the EAP-AKA server process
the AT_IDENTITY attribute and entities that pass through; EAP packets
do not process this attribute. Hence, the authenticator and other
intermediate AAA elements (such as possible AAA proxy servers) will
continue to refer to the peer with the original identity from the
EAP-Response/Identity packet unless the identity authenticated in the
AT_IDENTITY attribute is communicated to them in another way within
the AAA protocol.
4.1.2.2. Relying on EAP-Response/Identity Discouraged
The EAP-Response/Identity packet is not method specific; therefore,
in many implementations it may be handled by an EAP Framework. This
introduces an additional layer of processing between the EAP peer and
EAP server. The extra layer of processing may cache identity
responses or add decorations to the identity. A modification of the
identity response will cause the EAP peer and EAP server to use
different identities in the key derivation, which will cause the
protocol to fail.
For this reason, it is RECOMMENDED that the EAP peer and server use
the method-specific identity attributes in EAP-AKA, and the server is
strongly discouraged from relying upon the EAP-Response/Identity.
In particular, if the EAP server receives a decorated identity in
EAP-Response/Identity, then the EAP server MUST use the
identity-requesting attributes to request the peer to send an
unmodified and undecorated copy of the identity in AT_IDENTITY.
Arkko & Haverinen Informational [Page 22]
RFC 4187 EAP-AKA Authentication January 2006
4.1.3. Choice of Identity for the EAP-Response/Identity
If EAP-AKA peer is started upon receiving an EAP-Request/Identity
message, then the peer MAY use an EAP-AKA identity in the EAP-
Response/Identity packet. In this case, the peer performs the
following steps.
If the peer has maintained fast re-authentication state information
and if the peer wants to use fast re-authentication, then the peer
transmits the fast re-authentication identity in
EAP-Response/Identity.
Else, if the peer has a pseudonym username available, then the peer
transmits the pseudonym identity in EAP-Response/Identity.
In other cases, the peer transmits the permanent identity in
EAP-Response/Identity.
4.1.4. Server Operation in the Beginning of EAP-AKA Exchange
As discussed in Section 4.1.2.2, the server SHOULD NOT rely on an
identity string received in EAP-Response/Identity. Therefore, the
RECOMMENDED way to start an EAP-AKA exchange is to ignore any
received identity strings. The server SHOULD begin the EAP-AKA
exchange by issuing the EAP-Request/AKA-Identity packet with an
identity-requesting attribute to indicate that the server wants the
peer to include an identity in the AT_IDENTITY attribute of the EAP-
Response/AKA-Identity message. Three methods to request an identity
from the peer are discussed below.
If the server chooses to not ignore the contents of
EAP-Response/Identity, then the server may already receive an EAP-AKA
identity in this packet. However, if the EAP server has not received
any EAP-AKA peer identity (permanent identity, pseudonym identity, or
fast re-authentication identity) from the peer when sending the first
EAP-AKA request, or if the EAP server has received an
EAP-Response/Identity packet but the contents do not appear to be a
valid permanent identity, pseudonym identity, or a re-authentication
identity, then the server MUST request an identity from the peer
using one of the methods below.
The server sends the EAP-Request/AKA-Identity message with the
AT_PERMANENT_ID_REQ attribute to indicate that the server wants the
peer to include the permanent identity in the AT_IDENTITY attribute
of the EAP-Response/AKA-Identity message. This is done in the
following cases:
Arkko & Haverinen Informational [Page 23]
RFC 4187 EAP-AKA Authentication January 2006
o The server does not support fast re-authentication or identity
privacy.
o The server decided to process a received identity, and the server
recognizes the received identity as a pseudonym identity, but the
server is not able to map the pseudonym identity to a permanent
identity.
The server issues the EAP-Request/AKA-Identity packet with the
AT_FULLAUTH_ID_REQ attribute to indicate that the server wants the
peer to include a full authentication identity (pseudonym identity or
permanent identity) in the AT_IDENTITY attribute of the
EAP-Response/AKA-Identity message. This is done in the following
cases:
o The server does not support fast re-authentication and the server
supports identity privacy
o The server decided to process a received identity, and the server
recognizes the received identity as a re-authentication identity
but the server is not able to map the re-authentication identity
to a permanent identity
The server issues the EAP-Request/AKA-Identity packet with the
AT_ANY_ID_REQ attribute to indicate that the server wants the peer to
include an identity in the AT_IDENTITY attribute of the
EAP-Response/AKA-Identity message, and the server does not indicate
any preferred type for the identity. This is done in other cases,
such as when the server ignores a received EAP-Response/Identity,
when the server does not have any identity, or when the server does
not recognize the format of a received identity.
4.1.5. Processing of EAP-Request/AKA-Identity by the Peer
Upon receipt of an EAP-Request/AKA-Identity message, the peer MUST
perform the following steps.
If the EAP-Request/AKA-Identity includes AT_PERMANENT_ID_REQ, and if
the peer does not have a pseudonym available, then the peer MUST
respond with EAP-Response/AKA-Identity and include the permanent
identity in AT_IDENTITY. If the peer has a pseudonym available, then
the peer MAY refuse to send the permanent identity; hence, in this
case the peer MUST either respond with EAP-Response/AKA-Identity and
include the permanent identity in AT_IDENTITY or respond with
EAP-Response/AKA-Client-Error packet with code "unable to process
packet".
If the EAP-Request/AKA-Identity includes AT_FULL_AUTH_ID_REQ, and if
the peer has a pseudonym available, then the peer SHOULD respond with
EAP-Response/AKA-Identity and include the pseudonym identity in
Arkko & Haverinen Informational [Page 24]
RFC 4187 EAP-AKA Authentication January 2006
AT_IDENTITY. If the peer does not have a pseudonym when it receives
this message, then the peer MUST respond with EAP-Response/
AKA-Identity and include the permanent identity in AT_IDENTITY. The
Peer MUST NOT use a fast re-authentication identity in the
AT_IDENTITY attribute.
If the EAP-Request/AKA-Identity includes AT_ANY_ID_REQ, and if the
peer has maintained fast re-authentication state information and
wants to use fast re-authentication, then the peer responds with
EAP-Response/AKA-Identity and includes the fast re-authentication
identity in AT_IDENTITY. Else, if the peer has a pseudonym identity
available, then the peer responds with EAP-Response/AKA-Identity and
includes the pseudonym identity in AT_IDENTITY. Else, the peer
responds with EAP-Response/AKA-Identity and includes the permanent
identity in AT_IDENTITY.
An EAP-AKA exchange may include several EAP/AKA-Identity rounds. The
server may issue a second EAP-Request/AKA-Identity, if it was not
able to recognize the identity the peer used in the previous
AT_IDENTITY attribute. At most three EAP/AKA-Identity rounds can be
used, so the peer MUST NOT respond to more than three
EAP-Request/AKA-Identity messages within an EAP exchange. The peer
MUST verify that the sequence of EAP-Request/AKA-Identity packets the
peer receives comply with the sequencing rules defined in this
document. That is, AT_ANY_ID_REQ can only be used in the first
EAP-Request/AKA-Identity; in other words, AT_ANY_ID_REQ MUST NOT be
used in the second or third EAP-Request/AKA-Identity.
AT_FULLAUTH_ID_REQ MUST NOT be used if the previous
EAP-Request/AKA-Identity included AT_PERMANENT_ID_REQ. The peer
operation, in cases when it receives an unexpected attribute or an
unexpected message, is specified in Section 6.3.1.
4.1.6. Attacks against Identity Privacy
The section above specifies two possible ways the peer can operate
upon receipt of AT_PERMANENT_ID_REQ because a received
AT_PERMANENT_ID_REQ does not necessarily originate from the valid
network. However, an active attacker may transmit an
EAP-Request/AKA-Identity packet with an AT_PERMANENT_ID_REQ attribute
to the peer, in an effort to find out the true identity of the user.
If the peer does not want to reveal its permanent identity, then the
peer sends the EAP-Response/AKA-Client-Error packet with the error
code "unable to process packet", and the authentication exchange
terminates.
Basically, there are two different policies that the peer can employ
with regard to AT_PERMANENT_ID_REQ. A "conservative" peer assumes
that the network is able to maintain pseudonyms robustly. Therefore,
Arkko & Haverinen Informational [Page 25]
RFC 4187 EAP-AKA Authentication January 2006
if a conservative peer has a pseudonym username, the peer responds
with EAP-Response/AKA-Client-Error to the EAP packet with
AT_PERMANENT_ID_REQ, because the peer believes that the valid network
is able to map the pseudonym identity to the peer's permanent
identity. (Alternatively, the conservative peer may accept
AT_PERMANENT_ID_REQ in certain circumstances, for example if the
pseudonym was received a long time ago.) The benefit of this policy
is that it protects the peer against active attacks on anonymity. On
the other hand, a "liberal" peer always accepts the
AT_PERMANENT_ID_REQ and responds with the permanent identity. The
benefit of this policy is that it works even if the valid network
sometimes loses pseudonyms and is not able to map them to the
permanent identity.
4.1.7. Processing of AT_IDENTITY by the Server
When the server receives an EAP-Response/AKA-Identity message with
the AT_IDENTITY (in response to the server's identity requesting
attribute), the server MUST operate as follows.
If the server used AT_PERMANENT_ID_REQ, and if the AT_IDENTITY does
not contain a valid permanent identity, then the server sends an
EAP-Request/AKA-Notification packet with AT_NOTIFICATION code
"General failure" (16384) to terminate the EAP exchange. If the
server recognizes the permanent identity and is able to continue,
then the server proceeds with full authentication by sending
EAP-Request/AKA-Challenge.
If the server used AT_FULLAUTH_ID_REQ, and if AT_IDENTITY contains a
valid permanent identity or a pseudonym identity that the server can
map to a valid permanent identity, then the server proceeds with full
authentication by sending EAP-Request/AKA-Challenge. If AT_IDENTITY
contains a pseudonym identity that the server is not able to map to a
valid permanent identity, or an identity that the server is not able
to recognize or classify, then the server sends EAP-Request/
AKA-Identity with AT_PERMANENT_ID_REQ.
If the server used AT_ANY_ID_REQ, and if the AT_IDENTITY contains a
valid permanent identity or a pseudonym identity that the server can
map to a valid permanent identity, then the server proceeds with full
authentication by sending EAP-Request/ AKA-Challenge.
If the server used AT_ANY_ID_REQ, and if AT_IDENTITY contains a valid
fast re-authentication identity and the server agrees on using
re-authentication, then the server proceeds with fast
re-authentication by sending EAP-Request/AKA-Reauthentication
(Section 5).
Arkko & Haverinen Informational [Page 26]
RFC 4187 EAP-AKA Authentication January 2006
If the server used AT_ANY_ID_REQ, and if the peer sent an EAP-
Response/AKA-Identity with AT_IDENTITY that contains an identity that
the server recognizes as a fast re-authentication identity, but the
server is not able to map the identity to a permanent identity, then
the server sends EAP-Request/AKA-Identity with AT_FULLAUTH_ID_REQ.
If the server used AT_ANY_ID_REQ, and if AT_IDENTITY contains a valid
fast re-authentication identity, which the server is able to map to a
permanent identity, and if the server does not want to use fast
re-authentication, then the server proceeds with full authentication
by sending EAP-Request/AKA-Challenge.
If the server used AT_ANY_ID_REQ, and AT_IDENTITY contains an
identity that the server recognizes as a pseudonym identity but the
server is not able to map the pseudonym identity to a permanent
identity, then the server sends EAP-Request/AKA-Identity with
AT_PERMANENT_ID_REQ.
If the server used AT_ANY_ID_REQ, and AT_IDENTITY contains an
identity that the server is not able to recognize or classify, then
the server sends EAP-Request/AKA-Identity with AT_FULLAUTH_ID_REQ.
4.2. Message Sequence Examples (Informative)
This section contains non-normative message sequence examples to
illustrate how the peer identity can be communicated to the server.
4.2.1. Usage of AT_ANY_ID_REQ
Obtaining the peer identity with EAP-AKA attributes is illustrated in
Figure 5 below.
Peer Authenticator
| |
| +------------------------------+
| | Server does not have any |
| | Subscriber identity available|
| | When starting EAP-AKA |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_ANY_ID_REQ) |
|<------------------------------------------------------|
| |
| EAP-Response/AKA-Identity |
| (AT_IDENTITY) |
|------------------------------------------------------>|
| |
Figure 5: Usage of AT_ANY_ID_REQ
Arkko & Haverinen Informational [Page 27]
RFC 4187 EAP-AKA Authentication January 2006
4.2.2. Fall Back on Full Authentication
Figure 6 illustrates the case when the server does not recognize the
fast re-authentication identity the peer used in AT_IDENTITY.
Peer Authenticator
| |
| +------------------------------+
| | Server does not have any |
| | Subscriber identity available|
| | When starting EAP-AKA |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_ANY_ID_REQ) |
|<------------------------------------------------------|
| |
| EAP-Response/AKA-Identity |
| (AT_IDENTITY containing a fast re-auth. identity) |
|------------------------------------------------------>|
| +------------------------------+
| | Server does not recognize |
| | The fast re-auth. |
| | Identity |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_FULLAUTH_ID_REQ) |
|<------------------------------------------------------|
| EAP-Response/AKA-Identity |
| (AT_IDENTITY with a full-auth. Identity) |
|------------------------------------------------------>|
| |
Figure 6: Fall back on full authentication
If the server recognizes the fast re-authentication identity, but
still wants to fall back on full authentication, the server may issue
the EAP-Request/AKA-Challenge packet. In this case, the full
authentication procedure proceeds as usual.
Arkko & Haverinen Informational [Page 28]
RFC 4187 EAP-AKA Authentication January 2006
4.2.3. Requesting the Permanent Identity 1
Figure 7 illustrates the case when the EAP server fails to decode a
pseudonym identity included in the EAP-Response/Identity packet.
Peer Authenticator
| EAP-Request/Identity |
|<------------------------------------------------------|
| EAP-Response/Identity |
| (Includes a pseudonym) |
|------------------------------------------------------>|
| +------------------------------+
| | Server fails to decode the |
| | Pseudonym. |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_PERMANENT_ID_REQ) |
|<------------------------------------------------------|
| |
| EAP-Response/AKA-Identity |
| (AT_IDENTITY with permanent identity) |
|------------------------------------------------------>|
| |
Figure 7: Requesting the permanent identity 1
If the server recognizes the permanent identity, then the
authentication sequence proceeds as usual with the EAP Server issuing
the EAP-Request/AKA-Challenge message.
Arkko & Haverinen Informational [Page 29]
RFC 4187 EAP-AKA Authentication January 2006
4.2.4. Requesting the Permanent Identity 2
Figure 8 illustrates the case when the EAP server fails to decode the
pseudonym included in the AT_IDENTITY attribute.
Peer Authenticator
| |
| +------------------------------+
| | Server does not have any |
| | Subscriber identity available|
| | When starting EAP-AKA |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_ANY_ID_REQ) |
|<------------------------------------------------------|
| |
|EAP-Response/AKA-Identity |
|(AT_IDENTITY with a pseudonym identity) |
|------------------------------------------------------>|
| +------------------------------+
| | Server fails to decode the |
| | Pseudonym in AT_IDENTITY |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_PERMANENT_ID_REQ) |
|<------------------------------------------------------|
| EAP-Response/AKA-Identity |
| (AT_IDENTITY with permanent identity) |
|------------------------------------------------------>|
| |
Figure 8: Requesting the permanent identity 2
4.2.5. Three EAP/AKA-Identity Round Trips
Figure 9 illustrates the case with three EAP/AKA-Identity round
trips.
Arkko & Haverinen Informational [Page 30]
RFC 4187 EAP-AKA Authentication January 2006
Peer Authenticator
| |
| +------------------------------+
| | Server does not have any |
| | Subscriber identity available|
| | When starting EAP-AKA |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_ANY_ID_REQ) |
|<------------------------------------------------------|
| |
| EAP-Response/AKA-Identity |
| (AT_IDENTITY with fast re-auth. identity) |
|------------------------------------------------------>|
| +------------------------------+
| | Server does not accept |
| | The fast re-authentication |
| | Identity |
| +------------------------------+
| |
: :
: :
: :
: :
| EAP-Request/AKA-Identity |
| (AT_FULLAUTH_ID_REQ) |
|<------------------------------------------------------|
|EAP-Response/AKA-Identity |
|(AT_IDENTITY with a pseudonym identity) |
|------------------------------------------------------>|
| +------------------------------+
| | Server fails to decode the |
| | Pseudonym in AT_IDENTITY |
| +------------------------------+
| EAP-Request/AKA-Identity |
| (AT_PERMANENT_ID_REQ) |
|<------------------------------------------------------|
| EAP-Response/AKA-Identity |
| (AT_IDENTITY with permanent identity) |
|------------------------------------------------------>|
| |
Figure 9: Three EAP-AKA Start rounds
After the last EAP-Response/AKA-Identity message, the full
authentication sequence proceeds as usual.
Arkko & Haverinen Informational [Page 31]
RFC 4187 EAP-AKA Authentication January 2006
5. Fast Re-Authentication
5.1. General
In some environments, EAP authentication may be performed frequently.
Because the EAP-AKA full authentication procedure uses the AKA
algorithms, and therefore requires fresh authentication vectors from
the Authentication Centre, the full authentication procedure may
result in many network operations when used very frequently.
Therefore, EAP-AKA includes a more inexpensive fast re-authentication
procedure that does not make use of the AKA algorithms and does not
need new vectors from the Authentication Centre.
Fast re-authentication is optional to implement for both the EAP-AKA
server and peer. On each EAP authentication, either one of the
entities may fall back on full authentication if is does not want to
use fast re-authentication.
Fast re-authentication is based on the keys derived on the preceding
full authentication. The same K_aut and K_encr keys used in full
authentication are used to protect EAP-AKA packets and attributes,
and the original Master Key from full authentication is used to
generate a fresh Master Session Key, as specified in Section 7.
The fast re-authentication exchange makes use of an unsigned 16-bit
counter, included in the AT_COUNTER attribute. The counter has three
goals: 1) it can be used to limit the number of successive
reauthentication exchanges without full-authentication 2) it
contributes to the keying material, and 3) it protects the peer and
the server from replays. On full authentication, both the server and
the peer initialize the counter to one. The counter value of at
least one is used on the first fast re-authentication. On subsequent
fast re-authentications, the counter MUST be greater than on any of
the previous fast re-authentications. For example, on the second
fast re-authentication, counter value is two or greater, etc. The
AT_COUNTER attribute is encrypted.
Both the peer and the EAP server maintain a copy of the counter. The
EAP server sends its counter value to the peer in the fast
re-authentication request. The peer MUST verify that its counter
value is less than or equal to the value sent by the EAP server.
The server includes an encrypted server random nonce (AT_NONCE_S) in
the fast re-authentication request. The AT_MAC attribute in the
peer's response is calculated over NONCE_S to provide a
challenge/response authentication scheme. The NONCE_S also
contributes to the new Master Session Key.
Arkko & Haverinen Informational [Page 32]
RFC 4187 EAP-AKA Authentication January 2006
Both the peer and the server SHOULD have an upper limit for the
number of subsequent fast re-authentications allowed before a full
authentication needs to be performed. Because a 16-bit counter is
used in fast re-authentication, the theoretical maximum number of
re-authentications is reached when the counter value reaches FFFF
hexadecimal. In order to use fast re-authentication, the peer and
the EAP server need to store the following values: Master Key, latest
counter value and the next fast re-authentication identity. K_aut
and K_encr may either be stored or derived again from MK. The server
may also need to store the permanent identity of the user.
5.2. Comparison to AKA
When analyzing the fast re-authentication exchange, it may be helpful
to compare it with the 3rd generation Authentication and Key
Agreement (AKA) exchange used on full authentication. The counter
corresponds to the AKA sequence number, NONCE_S corresponds to RAND,
the AT_MAC in EAP-Request/AKA-Reauthentication corresponds to AUTN,
the AT_MAC in EAP-Response/AKA-Reauthentication corresponds to RES,
AT_COUNTER_TOO_SMALL corresponds to AUTS, and encrypting the counter
corresponds to the usage of the Anonymity Key. Also, the key
generation on fast re-authentication, with regard to random or fresh
material, is similar to AKA -- the server generates the NONCE_S and
counter values, and the peer only verifies that the counter value is
fresh.
It should also be noted that encrypting the AT_NONCE_S, AT_COUNTER,
or AT_COUNTER_TOO_SMALL attributes is not important to the security
of the fast re-authentication exchange.
5.3. Fast Re-Authentication Identity
The fast re-authentication procedure makes use of separate
re-authentication user identities. Pseudonyms and the permanent
identity are reserved for full authentication only. If a fast
re-authentication identity is lost and the network does not recognize
it, the EAP server can fall back on full authentication. If the EAP
server supports fast re-authentication, it MAY include the skippable
AT_NEXT_REAUTH_ID attribute in the encrypted data of EAP- Request/-
AKA-Challenge message. This attribute contains a new
re-authentication identity for the next fast re-authentication. The
attribute also works as a capability flag that indicates that the
server supports fast re-authentication and that the server wants to
continue using fast re-authentication within the current context.
The peer MAY ignore this attribute, in which case it will use full
authentication next time. If the peer wants to use fast
re-authentication, it uses this fast re-authentication identity on
next authentication. Even if the peer has a fast re-authentication
Arkko & Haverinen Informational [Page 33]
RFC 4187 EAP-AKA Authentication January 2006
identity, the peer MAY discard the re-authentication identity and use
a pseudonym or the permanent identity instead, in which case full
authentication MUST be performed. If the EAP server does not include
the AT_NEXT_REAUTH_ID in the encrypted data of
EAP-Request/AKA-Challenge or EAP-Request/AKA-Reauthentication, then
the peer MUST discard its current fast re-authentication state
information and perform a full authentication next time.
In environments where a realm portion is needed in the peer identity,
the fast re-authentication identity received in AT_NEXT_REAUTH_ID
MUST contain both a username portion and a realm portion, as per the
NAI format. The EAP Server can choose an appropriate realm part in
order to have the AAA infrastructure route subsequent fast
re-authentication-related requests to the same AAA server. For
example, the realm part MAY include a portion that is specific to the
AAA server. Hence, it is sufficient to store the context required
for fast re-authentication in the AAA server that performed the full
authentication.
The peer MAY use the fast re-authentication identity in the
EAP-Response/Identity packet or, in response to the server's
AT_ANY_ID_REQ attribute, the peer MAY use the fast re-authentication
identity in the AT_IDENTITY attribute of the EAP-Response/
AKA-Identity packet.
The peer MUST NOT modify the username portion of the fast
re-authentication identity, but the peer MAY modify the realm portion
or replace it with another realm portion. The peer might need to
modify the realm in order to influence the AAA routing, for example,
to make sure that the correct server is reached. It should be noted
that sharing the same fast re-authentication key among several
servers may have security risks, so changing the realm portion of the
NAI in order to change the EAP server is not desirable.
Even if the peer uses a fast re-authentication identity, the server
may want to fall back on full authentication, for example, because
the server does not recognize the fast re-authentication identity or
does not want to use fast re-authentication. If the server was able
to decode the fast re-authentication identity to the permanent
identity, the server issues the EAP-Request/AKA-Challenge packet to
initiate full authentication. If the server was not able to recover
the peer's identity from the fast re-authentication identity, the
server starts the full authentication procedure by issuing an
EAP-Request/AKA-Identity packet. This packet always starts a full
authentication sequence if it does not include the AT_ANY_ID_REQ
attribute.
Arkko & Haverinen Informational [Page 34]
RFC 4187 EAP-AKA Authentication January 2006
5.4. Fast Re-Authentication Procedure
Figure 10 illustrates the fast re-authentication procedure. In this
example, the optional protected success indication is not used.
Encrypted attributes are denoted with '*'. The peer uses its fast
re-authentication identity in the EAP-Response/Identity packet. As
discussed above, an alternative way to communicate the fast
re-authentication identity to the server is for the peer to use the
AT_IDENTITY attribute in the EAP-Response/AKA-Identity message. This
latter case is not illustrated in the figure below, and it is only
possible when the server requests that the peer send its identity by
including the AT_ANY_ID_REQ attribute in the EAP-Request/AKA-Identity
packet.
If the server recognizes the identity as a valid fast
re-authentication identity, and if the server agrees to use fast
re-authentication, then the server sends the EAP- Request/AKA-
Reauthentication packet to the peer. This packet MUST include the
encrypted AT_COUNTER attribute, with a fresh counter value, the
encrypted AT_NONCE_S attribute that contains a random number chosen
by the server, the AT_ENCR_DATA and the AT_IV attributes used for
encryption, and the AT_MAC attribute that contains a message
authentication code over the packet. The packet MAY also include an
encrypted AT_NEXT_REAUTH_ID attribute that contains the next fast
re-authentication identity.
Fast re-authentication identities are one-time identities. If the
peer does not receive a new fast re-authentication identity, it MUST
use either the permanent identity or a pseudonym identity on the next
authentication to initiate full authentication.
The peer verifies that AT_MAC is correct and that the counter value
is fresh (greater than any previously used value). The peer MAY save
the next fast re-authentication identity from the encrypted
AT_NEXT_REAUTH_ID for next time. If all checks are successful, the
peer responds with the EAP-Response/AKA-Reauthentication packet,
including the AT_COUNTER attribute with the same counter value and
the AT_MAC attribute.
The server verifies the AT_MAC attribute and also verifies that the
counter value is the same that it used in the
EAP-Request/AKA-Reauthentication packet. If these checks are
successful, the fast re-authentication has succeeded and the server
sends the EAP-Success packet to the peer.
If protected success indications (Section 6.2) were used, the
EAP-Success packet would be preceded by an EAP-AKA notification
round.
Arkko & Haverinen Informational [Page 35]
RFC 4187 EAP-AKA Authentication January 2006
Peer Authenticator
| |
| EAP-Request/Identity |
|<------------------------------------------------------|
| |
| EAP-Response/Identity |
| (Includes a fast re-authentication identity) |
|------------------------------------------------------>|
| +--------------------------------+
| | Server recognizes the identity |
| | and agrees on using fast |
| | re-authentication |
| +--------------------------------+
| EAP-Request/AKA-Reauthentication |
| (AT_IV, AT_ENCR_DATA, *AT_COUNTER, |
| *AT_NONCE_S, *AT_NEXT_REAUTH_ID, AT_MAC) |
|<------------------------------------------------------|
| |
: :
: :
: :
: :
| |
+-----------------------------------------------+ |
| Peer verifies AT_MAC and the freshness of | |
| the counter. Peer MAY store the new re- | |
| authentication identity for next re-auth. | |
+-----------------------------------------------+ |
| |
| EAP-Response/AKA-Reauthentication |
| (AT_IV, AT_ENCR_DATA, *AT_COUNTER with same value, |
| AT_MAC) |
|------------------------------------------------------>|
| +--------------------------------+
| | Server verifies AT_MAC and |
| | the counter |
| +--------------------------------+
| EAP-Success |
|<------------------------------------------------------|
| |
Figure 10: Reauthentication
Arkko & Haverinen Informational [Page 36]
RFC 4187 EAP-AKA Authentication January 2006
5.5. Fast Re-Authentication Procedure when Counter is Too Small
If the peer does not accept the counter value of EAP-Request/
AKA-Reauthentication, it indicates the counter synchronization
problem by including the encrypted AT_COUNTER_TOO_SMALL in
EAP-Response/AKA-Reauthentication. The server responds with
EAP-Request/AKA-Challenge to initiate a normal full authentication
procedure. This is illustrated in Figure 11. Encrypted attributes
are denoted with '*'.
Peer Authenticator
| EAP-Request/AKA-Identity |
| (AT_ANY_ID_REQ) |
|<------------------------------------------------------|
| |
| EAP-Response/AKA-Identity |
| (AT_IDENTITY) |
| (Includes a fast re-authentication identity) |
|------------------------------------------------------>|
| |
| EAP-Request/AKA-Reauthentication |
| (AT_IV, AT_ENCR_DATA, *AT_COUNTER, |
| *AT_NONCE_S, *AT_NEXT_REAUTH_ID, AT_MAC) |
|<------------------------------------------------------|
+-----------------------------------------------+ |
| AT_MAC is valid but the counter is not fresh. | |
+-----------------------------------------------+ |
| EAP-Response/AKA-Reauthentication |
| (AT_IV, AT_ENCR_DATA, *AT_COUNTER_TOO_SMALL, |
| *AT_COUNTER, AT_MAC) |
|------------------------------------------------------>|
| +----------------------------------------------+
| | Server verifies AT_MAC but detects |
| | That peer has included AT_COUNTER_TOO_SMALL|
| +----------------------------------------------+
| EAP-Request/AKA-Challenge |
|<------------------------------------------------------|
+---------------------------------------------------------------+
| Normal full authentication follows. |
+---------------------------------------------------------------+
| |
Figure 11: Fast re-authentication counter too small
In the figure above, the first three messages are similar to the
basic fast re-authentication case. When the peer detects that the
counter value is not fresh, it includes the AT_COUNTER_TOO_SMALL
attribute in EAP-Response/AKA-Reauthentication. This attribute
Arkko & Haverinen Informational [Page 37]
RFC 4187 EAP-AKA Authentication January 2006
doesn't contain any data but it is a request for the server to
initiate full authentication. In this case, the peer MUST ignore the
contents of the server's AT_NEXT_REAUTH_ID attribute.
On receipt of AT_COUNTER_TOO_SMALL, the server verifies AT_MAC and
verifies that AT_COUNTER contains the same counter value as in the
EAP-Request/AKA-Reauthentication packet. If not, the server
terminates the authentication exchange by sending the
EAP-Request/AKA-Notification packet with AT_NOTIFICATION code
"General failure" (16384). If all checks on the packet are
successful, the server transmits an EAP-Request/AKA-Challenge packet
and the full authentication procedure is performed as usual. Because
the server already knows the subscriber identity, it MUST NOT use the
EAP-Request/AKA-Identity packet to request the identity.
It should be noted that in this case, peer identity is only
transmitted in the AT_IDENTITY attribute at the beginning of the
whole EAP exchange. The fast re-authentication identity used in this
AT_IDENTITY attribute will be used in key derivation (see Section 7).
6. EAP-AKA Notifications
6.1. General
EAP-AKA does not prohibit the use of the EAP Notifications as
specified in [RFC3748]. EAP Notifications can be used at any time in
the EAP-AKA exchange. It should be noted that EAP-AKA does not
protect EAP Notifications. EAP-AKA also specifies method-specific
EAP-AKA notifications, which are protected in some cases.
The EAP server can use EAP-AKA notifications to convey notifications
and result indications (Section 6.2) to the peer.
The server MUST use notifications in cases discussed in
Section 6.3.2. When the EAP server issues an
EAP-Request/AKA-Notification packet to the peer, the peer MUST
process the notification packet. The peer MAY show a notification
message to the user and the peer MUST respond to the EAP server with
an EAP-Response/AKA-Notification packet, even if the peer did not
recognize the notification code.
An EAP-AKA full authentication exchange or a fast re-authentication
exchange MUST NOT include more than one EAP-AKA notification round.
The notification code is a 16-bit number. The most significant bit
is called the Success bit (S bit). The S bit specifies whether the
notification implies failure. The code values with the S bit set to
zero (code values 0...32767) are used on unsuccessful cases. The
Arkko & Haverinen Informational [Page 38]
RFC 4187 EAP-AKA Authentication January 2006
receipt of a notification code from this range implies failed EAP
exchange, so the peer can use the notification as a failure
indication. After receiving the EAP-Response/AKA-Notification for
these notification codes, the server MUST send the EAP-Failure
packet.
The receipt of a notification code with the S bit set to one (values
32768...65536) does not imply failure. Notification code "Success"
(32768) has been reserved as a general notification code to indicate
successful authentication.
The second most significant bit of the notification code is called
the Phase bit (P bit). It specifies at which phase of the EAP-AKA
exchange the notification can be used. If the P bit is set to zero,
the notification can only be used after a successful EAP/AKA-
Challenge round in full authentication or a successful EAP/AKA-
Reauthentication round in re-authentication. A re-authentication
round is considered successful only if the peer has successfully
verified AT_MAC and AT_COUNTER attributes, and does not include the
AT_COUNTER_TOO_SMALL attribute in EAP-Response/AKA-Reauthentication.
If the P bit is set to one, the notification can only by used before
the EAP/AKA-Challenge round in full authentication or before the
EAP/AKA-Reauthentication round in reauthentication. These
notifications can only be used to indicate various failure cases. In
other words, if the P bit is set to one, then the S bit MUST be set
to zero.
Section 9.10 and Section 9.11 specify what other attributes must be
included in the notification packets.
Some of the notification codes are authorization related and hence
not usually considered as part of the responsibility of an EAP
method. However, they are included as part of EAP-AKA because there
are currently no other ways to convey this information to the user in
a localizable way, and the information is potentially useful for the
user. An EAP-AKA server implementation may decide never to send
these EAP-AKA notifications.
6.2. Result Indications
As discussed in Section 6.3, the server and the peer use explicit
error messages in all error cases. If the server detects an error
after successful authentication, the server uses an EAP-AKA
notification to indicate failure to the peer. In this case, the
result indication is integrity and replay protected.
Arkko & Haverinen Informational [Page 39]
RFC 4187 EAP-AKA Authentication January 2006
By sending an EAP-Response/AKA-Challenge packet or an
EAP-Response/AKA-Reauthentication packet (without
AT_COUNTER_TOO_SMALL), the peer indicates that it has successfully
authenticated the server and that the peer's local policy accepts the
EAP exchange. In other words, these packets are implicit success
indications from the peer to the server.
EAP-AKA also supports optional protected success indications from the
server to the peer. If the EAP server wants to use protected success
indications, it includes the AT_RESULT_IND attribute in the
EAP-Request/AKA-Challenge or the EAP-Request/AKA-Reauthentication
packet. This attribute indicates that the EAP server would like to
use result indications in both successful and unsuccessful cases. If
the peer also wants this, the peer includes AT_RESULT_IND in
EAP-Response/AKA-Challenge or EAP-Response/AKA-Reauthentication. The
peer MUST NOT include AT_RESULT_IND if it did not receive
AT_RESULT_IND from the server. If both the peer and the server used
AT_RESULT_IND, then the EAP exchange is not complete yet, but an
EAP-AKA notification round will follow. The following EAP-AKA
notification may indicate either failure or success.
Success indications with the AT_NOTIFICATION code "Success" (32768)
can only be used if both the server and the peer indicate they want
to use them with AT_RESULT_IND. If the server did not include
AT_RESULT_IND in the EAP-Request/AKA-Challenge or
EAP-Request/AKA-Reauthentication packet, or if the peer did not
include AT_RESULT_IND in the corresponding response packet, then the
server MUST NOT use protected success indications.
Because the server uses the AT_NOTIFICATION code "Success" (32768) to
indicate that the EAP exchange has completed successfully, the EAP
exchange cannot fail when the server processes the EAP-AKA response
to this notification. Hence, the server MUST ignore the contents of
the EAP-AKA response it receives to the EAP-Request/AKA-Notification
with this code. Regardless of the contents of the EAP-AKA response,
the server MUST send EAP-Success as the next packet.
6.3. Error Cases
This section specifies the operation of the peer and the server in
error cases. The subsections below require the EAP-AKA peer and
server to send an error packet (EAP-Response/AKA-Client-Error,
EAP-Response/AKA-Authentication-Reject or
EAP-Response/AKA-Synchronization-Failure from the peer and
EAP-Request/AKA-Notification from the server) in error cases.
However, implementations SHOULD NOT rely upon the correct error
reporting behavior of the peer, authenticator, or server. It is
possible for error messages and other messages to be lost in transit,
Arkko & Haverinen Informational [Page 40]
RFC 4187 EAP-AKA Authentication January 2006
or for a malicious participant to attempt to consume resources by not
issuing error messages. Both the peer and the EAP server SHOULD have
a mechanism to clean up state even if an error message or EAP-Success
is not received after a timeout period.
6.3.1. Peer Operation
Two special error messages have been specified for error cases that
are related to the processing of the AKA AUTN parameter, as described
in Section 3: (1) if the peer does not accept AUTN, the peer responds
with EAP-Response/AKA-Authentication-Reject (Section 9.5), and the
server issues EAP-Failure, and (2) if the peer detects that the
sequence number in AUTN is not correct, the peer responds with
EAP-Response/AKA-Synchronization-Failure (Section 9.6), and the
server proceeds with a new EAP-Request/AKA-Challenge.
In other error cases, when an EAP-AKA peer detects an error in a
received EAP-AKA packet, the EAP-AKA peer responds with the
EAP-Response/AKA-Client-Error packet. In response to the
EAP-Response/AKA-Client-Error, the EAP server MUST issue the
EAP-Failure packet, and the authentication exchange terminates.
By default, the peer uses the client error code 0, "unable to process
packet". This error code is used in the following cases:
o EAP exchange is not acceptable according to the peer's local
policy.
o The peer is not able to parse the EAP request, i.e., the EAP
request is malformed.
o The peer encountered a malformed attribute.
o Wrong attribute types or duplicate attributes have been included
in the EAP request.
o A mandatory attribute is missing.
o Unrecognized non-skippable attribute.
o Unrecognized or unexpected EAP-AKA Subtype in the EAP request.
o Invalid AT_MAC. The peer SHOULD log this event.
o Invalid AT_CHECKCODE. The peer SHOULD log this event.
o Invalid pad bytes in AT_PADDING.
o The peer does not want to process AT_PERMANENT_ID_REQ.
6.3.2. Server Operation
If an EAP-AKA server detects an error in a received EAP-AKA response,
the server MUST issue the EAP-Request/AKA-Notification packet with an
AT_NOTIFICATION code that implies failure. By default, the server
uses one of the general failure codes ("General failure after
authentication" (0) or "General failure" (16384)). The choice
Arkko & Haverinen Informational [Page 41]
RFC 4187 EAP-AKA Authentication January 2006
between these two codes depends on the phase of the EAP-AKA exchange,
see Section 6. The error cases when the server issues an
EAP-Request/AKA-Notification that implies failure include the
following:
o The server is not able to parse the peer's EAP response.
o The server encounters a malformed attribute, a non-recognized
non-skippable attribute, or a duplicate attribute.
o A mandatory attribute is missing or an invalid attribute was
included.
o Unrecognized or unexpected EAP-AKA Subtype in the EAP Response.
o Invalid AT_MAC. The server SHOULD log this event.
o Invalid AT_CHECKCODE. The server SHOULD log this event.
o Invalid AT_COUNTER.
6.3.3. EAP-Failure
The EAP-AKA server sends EAP-Failure in three cases:
1. In response to an EAP-Response/AKA-Client-Error packet the server
has received from the peer, or
2. In response to an EAP-Response/AKA-Authentication-Reject packet
the server has received from the peer, or
3. Following an EAP-AKA notification round, when the AT_NOTIFICATION
code implies failure.
The EAP-AKA server MUST NOT send EAP-Failure in other cases than
these three. However, it should be noted that even though the
EAP-AKA server would not send an EAP-Failure, an authorization
decision that happens outside EAP-AKA, such as in the AAA server or
in an intermediate AAA proxy, may result in a failed exchange.
The peer MUST accept the EAP-Failure packet in case 1), case 2), and
case 3) above. The peer SHOULD silently discard the EAP-Failure
packet in other cases.
6.3.4. EAP-Success
On full authentication, the server can only send EAP-Success after
the EAP/AKA-Challenge round. The peer MUST silently discard any
EAP-Success packets if they are received before the peer has
successfully authenticated the server and sent the
EAP-Response/AKA-Challenge packet.
Arkko & Haverinen Informational [Page 42]
RFC 4187 EAP-AKA Authentication January 2006
If the peer did not indicate that it wants to use protected success
indications with AT_RESULT_IND (as discussed in Section 6.2) on full
authentication, then the peer MUST accept EAP-Success after a
successful EAP/AKA-Challenge round.
If the peer indicated that it wants to use protected success
indications with AT_RESULT_IND (as discussed in Section 6.2), then
the peer MUST NOT accept EAP-Success after a successful EAP/
AKA-Challenge round. In this case, the peer MUST only accept
EAP-Success after receiving an EAP-AKA Notification with the
AT_NOTIFICATION code "Success" (32768).
On fast re-authentication, EAP-Success can only be sent after the
EAP/AKA-Reauthentication round. The peer MUST silently discard any
EAP-Success packets if they are received before the peer has
successfully authenticated the server and sent the
EAP-Response/AKA-Reauthentication packet.
If the peer did not indicate that it wants to use protected success
indications with AT_RESULT_IND (as discussed in Section 6.2) on fast
re-authentication, then the peer MUST accept EAP-Success after a
successful EAP/AKA-Reauthentication round.
If the peer indicated that it wants to use protected success
indications with AT_RESULT_IND (as discussed in Section 6.2), then
the peer MUST NOT accept EAP-Success after a successful EAP/AKA-
Reauthentication round. In this case, the peer MUST only accept
EAP-Success after receiving an EAP-AKA Notification with the
AT_NOTIFICATION code "Success" (32768).
If the peer receives an EAP-AKA notification (Section 6) that
indicates failure, then the peer MUST no longer accept the
EAP-Success packet, even if the server authentication was
successfully completed.
7. Key Generation
This section specifies how keying material is generated.
On EAP-AKA full authentication, a Master Key (MK) is derived from the
underlying AKA values (CK and IK keys), and the identity, as follows.
MK = SHA1(Identity|IK|CK)
In the formula above, the "|" character denotes concatenation.
Identity denotes the peer identity string without any terminating
null characters. It is the identity from the last AT_IDENTITY
attribute sent by the peer in this exchange, or, if AT_IDENTITY was