]> Cisco Systems

Cessna Business Park, Sarjapura Marathalli Outer Ring Road Bangalore KARNATAKA 560 087 India +91 80 4426 0474 shwethab@cisco.com

Cisco Systems

170 West Tasman Drive San Jose CA 95134 United States sgundave@cisco.com

Cisco Systems

11 New Square Park Bedfont Lakes FELTHAM TW14 8HA England mgrayson@cisco.com

Cisco Systems

1414 Massachusetts Ave Boxborough MA 01719 United States volz@cisco.com

Broadcom Limited

3151 Zanker Rd San Jose CA 95134 United States jouni.nospam@gmail.com

This document specifies the format and mechanism that is to be used for encoding Access-Network Identifiers in DHCPv4 and DHCPv6 messages by defining new Access-Network-Identifier options and sub-options.

Access-network identification of a network device has a range of applications. For example, the Local Mobility Anchor (LMA) in a Proxy Mobile IPv6 (PMIPv6) domain is able to provide service treatment for the mobile node's traffic based on the access network to which the mobile node is attached. This document specifies the Dynamic Host Configuration Protocol for IPv4 (DHCPv4) and the Dynamic Host Configuration Protocol for IPv6 (DHCPv6) options for access-network identification that is added by the relay agent in the DHCPv4 or DHCPv6 messages sent towards the server. The scope of applicability for this option is between a DHCP relay agent and a mobile access gateway where the same operator typically operates both these functions A DHCP relay agent that is aware of the access network and access operator adds this information in the DHCP messages. This information can be used to provide differentiated services and policing of traffic based on the access network to which a client is attached. Examples of how this information can be used in mobile networks can be found in .

PMIPv6 can be used for supporting network-based mobility management in various types of network deployments. The network architectures, such as service provider Wi-Fi access aggregation or WLAN integrated mobile packet core, are examples where PMIPv6 is a component of the overall architecture. Some of these architectures require the ability of the LMA to provide differentiated services and policing of traffic to the mobile nodes based on the access network to which they are attached. Policy systems in mobility architectures, such as Policy and Charging Control (PCC) and Access Network Discovery and Selection Function (ANDSF) in the 3GPP system, allow configuration of policy rules with conditions based on the access-network information. For example, the service treatment for the mobile node's traffic may be different when they are attached to an access network owned by the home operator than when owned by a roaming partner. In the case of access networks based on IEEE 802.11, the service treatment can also be different based on the configured Service Set Identifiers (SSIDs). Other examples of services include the operator's ability to apply tariff based on the location. The PMIPv6 extension as specified in defines PMIPv6 options to carry Access-Network Identifiers in PMIPv6 signaling from the Mobile Access Gateway (MAG) to the LMA. The MAG can learn this information from the DHCP options as inserted by the DHCP relay agent in the access network. If the MAG relays the DHCP messages to the LMA as specified in , this information can be inserted by the MAG towards the LMA in the forwarded DHCP messages. Figure 1 illustrates an example of PMIPv6 deployment. In this example, the access network is based on IEEE 802.11 technology, the DHCP relay agent function is located on the Access Point (AP), and the DHCP server function is located on the MAG. The MAG delivers the information elements related to the access network to the LMA over PMIPv6 signaling messages. The MAG obtains these information elements from the DHCP relay agent as per this specification. The information elements related to the access network include the SSID of the used IEEE 802.11 network, the geo-location of the access network to which the mobile node is attached, and the identity of the operator running the IEEE 802.11 access-network infrastructure.

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 . All the DHCP-related terms used in this document are to be interpreted as defined in DHCPv4 and DHCPv6 specifications. "DHCP message" refers to both DHCPv4 and DHCPv6 messages throughout this document. All the mobility-related terms used in this document are to be interpreted as defined in the PMIPv6 specifications and . Additionally, this document uses the following abbreviations: Service Set Identifier (SSID) The Service Set Identifier (SSID) identifies the name of the IEEE 802.11 network. The SSID differentiates from one network to the other. Operator-Identifier The Operator-Identifier is the Structure of Management Information (SMI) Network Management Private Enterprise Code of the IANA-maintained "Private Enterprise Numbers" registry . It identifies the operator running the access network where the client is attached.

The Access-Network Identifier (ANI) carries information related to the identity of the access network to which the client is attached. This information includes access-technology type, network identifier, and access network operator identifiers. Relay agents that include ANI information include one or more sub-options (see Section 4.1) in the Relay Agent Information option .

The Access-Network-Identifier information will be defined in multiple sub-options allocated from the "DHCP Relay Agent Sub-Option Codes" registry. ANI Sub-options: The ANI sub-options consist of a sequence of Sub-Option Code, Length, and Value tuples for each sub-option, encoded in the following manner:

The 1-octet code for the sub-options defined in the following sections. An unsigned 8-bit integer giving the length of the Sub-option Data field in this sub-option in octets. The data area for the sub-option. The initial assignment of the DHCP Access-Network-Identifier sub-options is as follows:

This sub-option is used for exchanging the type of the access technology of the network to which the client is attached. Its format is as follows:

13 2 An 8-bit field that is unused for now. The value MUST be initialized to 0 by the sender and MUST be ignored by the receiver. An 8-bit field that specifies the access technology through which the client is connected to the access link from the IANA name space "Access Technology Type Option type values" registry defined in .

These sub-options are used for carrying the name of the access network (e.g., an SSID in the case of an IEEE 802.11 access network or a Public Land-based Mobile Network (PLMN) Identifier in the case of a 3GPP access network) and the Access-Point Name to which the client is attached. The format of these sub-options is defined in the following sections. The Network-Identifier sub-options are only for the currently known access-technology types.

14 The length of the Network-Name field. The name of the access network to which the mobile node is attached. The encoding MUST be UTF-8 as described in . The type of the Network-Name is dependent on the access technology to which the mobile node is attached. For networks based on IEEE 802.11, the Network-Name will be the SSID of the network. For 3GPP access-based networks, it is the PLMN Identifier of the access network, and for 3GPP2 access, the Network-Name is the ANI . When encoding the PLMN Identifier, both the Mobile Network Code (MNC) and Mobile Country Code (MCC) MUST be three digits. If the MNC in use only has two digits, then it MUST be preceded with a '0'.

15 The length of the Access-Point-Name field. The name of the access point (physical device name) to which the mobile node is attached. This is the identifier that uniquely identifies the access point. While the Network-Name (e.g., SSID) identifies the operator's access network, the Access-Point-Name identifies a specific network device in the network to which the mobile node is attached. In some deployments, the Access-Point-Name can be set to the string representation of the Media Access Control (MAC) address of the device as specified in (see mac-address typedef) or some unique identifier that can be used by the policy systems in the operator network to unambiguously identify the device. The encoding MUST be UTF-8 as described in .

16 6 The 48-bit Basic SSSID (BSSID) of the access point to which the mobile node is attached.

The Operator-Identifier sub-options can be used for carrying the Operator-Identifiers of the access network to which the client is attached. The format of these sub-options is defined below.

17 4 The Operator-Identifier is a variable-length Private Enterprise Number (PEN) encoded in a network byte order. Please refer to Section 3.1.3 of for additional details.

18 The length of the Operator-Realm field. Realm of the operator (e.g., EXAMPLE.COM). Please refer to Section 3.1.3 of for additional details.

The Access-Network-Identifier options defined here may be added by the DHCPv6 relay agent in Relay-forward messages.

This option is used for exchanging the type of access technology the client uses to attach to the network. Its format is as follows:

OPTION_ANI_ATT (105) 2 An 8-bit field that is unused for now. The value MUST be initialized to 0 by the sender and MUST be ignored by the receiver. The contents of this field are the same as the ATT field described in .

These options can be used for carrying the name of the access network (e.g., an SSID in the case of an IEEE 802.11 access network or a PLMN Identifier in the case of a 3GPP access network) and an Access-Point Name to which the client is attached. The format of these options is defined below.

OPTION_ANI_NETWORK_NAME (106) The length of the Network-Name field. The contents of this field are the same as the Network-Name field described in .

OPTION_ANI_AP_NAME (107) The length of the Access-Point-Name field. The contents of this field are the same as the Access-Point-Name field described in .

OPTION_ANI_AP_BSSID (108) 6 The contents of this field are the same as the Access-Point-BSSID field described in .

The Operator-Identifier options can be used for carrying the Operator-Identifier of the access network to which the client is attached. The format of these options is defined below.

OPTION_ANI_OPERATOR_ID (109) 4 The contents of this field are the same as the DHCPv4 Operator-Identifier Sub-option field described in .

OPTION_ANI_OPERATOR_REALM (110) The length of the Operator-Realm field. The contents of this field are the same as the Operator-Realm field described in .

DHCPv4 relay agents MAY include sub-options as defined in Section 4.2 through 4.4 of in the Relay Agent Information option for providing information about the access network over which DHCP messages from the client are received. The DHCPv4 relay agent MUST include the DHCPv4 Access-Technology-Type Sub-option () when including any of these sub-options in the DHCP message: DHCPv4 Network-Name Sub-option (), DHCPv4 Access-Point-Name Sub-option (), and DHCPv4 Access-Point-BSSID Sub-option (). DHCPv6 Relay Agents MAY include options (defined in ) in the Relay-forward message when forwarding any DHCPv6 message type from clients to the servers to provide information about the access network over which DHCPv6 messages from the client are received. The DHCPv6 relay agent MUST include the DHCPv6 Access-Technology-Type Option () when including any of these options in the DHCP message: DHCPv6 Network-Name Option (), DHCPv6 Access-Point-Name Option (), and DHCPv6 Access-Point-BSSID Option ().

The DHCPv4 base specification requires that the DHCPv4 server ignore the DHCPv4 Access-Network-Identifier Option if it does not understand the option. If the DHCPv4 server does not understand the received sub-option defined in Sections 4.1 through 4.4 of , the DHCPv4 Relay-Agent-Information Option, it MUST ignore those sub-options only. If the DHCPv4 server is able to process the DHCPv4 Access-Network-Identifier sub-options defined in Sections 4.1 through 4.4 of , the DHCPv4 Relay-Agent-Information Option, it MAY use this information obtained from the sub-option for address pool selection or for policy decisions as per its configured policy. This information obtained from the sub-option SHOULD NOT be stored unless it is absolutely needed. However, if it is stored, the information MUST be deleted as quickly as possible to eliminate any possibility of the information getting exposed to an intruder. The DHCPv4 server MUST ignore the received DHCPv4 Access-Network-Identifier Option and process the rest of the message as per the base DHCPv4 specifications if the received DHCPv4 message does not include the DHCPv4 Access-Technology-Type Sub-option () but does include any one of these other options: DHCPv4 Network Name Sub-option (), DHCPv4 Access-Point-Name Sub-option (), or DHCPv4 Access-Point-BSSID Sub-option (). DHCPv6 base specification requires that the DHCPv6 server ignore the DHCPv6 Access-Network-Identifier Option if it does not understand the option. If the DHCPv6 server receives the options defined in Section 5 and is configured to use the options defined in Section 5, it SHOULD look for the DHCPv6 Access-Network-Identifier options in the Relay-forward message of the DHCPv6 relay agent(s) based on its configured policy. The server MAY use received ANI options for its address pool selection policy decisions as per its configured policy. This information obtained from the options SHOULD NOT be stored unless it is absolutely needed. However, if it is stored, the information MUST be deleted as quickly as possible to eliminate any possibility of the information getting exposed to an intruder. The DHCPv6 server MUST ignore the received DHCPv6 Access-Network-Identifier Option and process the rest of the message as per the base DHCPv6 specifications if the received DHCPv6 message does not include the DHCPv6 Access-Technology-Type Option () but it does includes any one of these other options: DHCPv6 Network-Name Option (), DHCPv6 Access-Point-Name Option (), or DHCPv6 Access-Point-BSSID Option ().

IANA has assigned sub-option codes for the following DHCPv4 sub-options from the "DHCP Relay Agent Sub-Option Codes" registry, <http://www.iana.org/assignments/bootp-dhcp-parameters>:

IANA has assigned option codes for the following DHCPv6 options from the "Option Codes" registry for DHCPv6, <http://www.iana.org/assignments/dhcpv6-parameters>, as specified in :

Since there is no privacy protection for DHCP messages, an eavesdropper who can monitor the link between the DHCP server and relay agent can discover access-network information. and describe many of the threats in using DHCP. and each provide a solution; the Authentication Option for DHCPv4 and DHCPv6 (respectively). However, neither of these options are in active use and therefore are not a viable mitigation option. DHCP itself is inherently insecure and thus link-layer confidentiality and integrity protection SHOULD be employed to reduce the risk of disclosure and tampering. It is possible for a rogue DHCP relay agent to insert or overwrite with incorrect Access-Network-Identifier options for malicious purposes. A DHCP client can also pose as a rogue DHCP relay agent by sending incorrect Access-Network-Identifier options. While the introduction of fraudulent DHCP relay agent information options can be prevented by a perimeter defense that blocks these options unless the DHCP relay agent is trusted, a deeper defense using the authentication sub-option for the DHCPv4 Relay-Agent-Information Option SHOULD be deployed as well. Administrators SHOULD configure DHCP servers that use this option to communicate with their relay agents using IPsec, as described in Section 21.1 of . The information elements that this document is exposing are the client’s access-network information. These pertain to the access network to which the client is attached, such as Access-Technology Type (e.g., WLAN, Ethernet, etc.), Access-Point Identity (Name, BSSID), and Operator-Identifier and Operator-Realm. In deployments where this information cannot be secured using IPsec or other security protocols, administrators SHOULD disable the capability specified in this document on the DHCP entities.

&RFC2119; &RFC2131; &RFC3315; &RFC3046; &RFC3118; &RFC4030; &RFC5213; &RFC5844; &RFC6757; &RFC3629; &RFC6991; &RFC4301; IANA 3GPP 3GPP 3GPP

The authors would like to thank Kim Kinnear, Ted Lemon, Gaurav Halwasia, Hidetoshi Yokota, Sheng Jiang, and Francis Dupont for their valuable input. Also, thank you to Tomek Mrugalski for a thorough review of the document.