Secure Neighbor Discovery Working J. Arkko Group Ericsson Internet-Draft J. Kempf Expires: April 16, 2004 DoCoMo Communications Labs USA B. Sommerfeld Sun Microsystems B. Zill Microsoft P. Nikander Ericsson October 17, 2003 SEcure Neighbor Discovery (SEND) draft-ietf-send-ndopt-00 Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http:// www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. This Internet-Draft will expire on April 16, 2004. Copyright Notice Copyright (C) The Internet Society (2003). All Rights Reserved. Abstract IPv6 nodes use the Neighbor Discovery Protocol (NDP) to discover other nodes on the link, to determine each the link-layer addresses of the nodes on the link, to find routers, and to maintain reachability information about the paths to active neighbors. If not secured, NDP is vulnerable to various attacks. This document Arkko, et al. Expires April 16, 2004 [Page 1] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 specifies security mechanisms for NDP. Unlike to the original NDP specifications, these mechanisms do not make use of IPsec. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . 4 2. Terms . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3. Neighbor and Router Discovery Overview . . . . . . . . . . 7 4. Secure Neighbor Discovery Overview . . . . . . . . . . . . 11 5. Neighbor Discovery Options . . . . . . . . . . . . . . . . 12 5.1 Ordering of the new options . . . . . . . . . . . . . . . 12 5.2 CGA Option . . . . . . . . . . . . . . . . . . . . . . . . 12 5.2.1 Processing Rules for Senders . . . . . . . . . . . . . . . 14 5.2.2 Processing Rules for Receivers . . . . . . . . . . . . . . 15 5.2.3 Configuration . . . . . . . . . . . . . . . . . . . . . . 15 5.3 Signature Option . . . . . . . . . . . . . . . . . . . . . 15 5.3.1 Processing Rules for Senders . . . . . . . . . . . . . . . 18 5.3.2 Processing Rules for Receivers . . . . . . . . . . . . . . 18 5.3.3 Configuration . . . . . . . . . . . . . . . . . . . . . . 19 5.4 Timestamp and Nonce options . . . . . . . . . . . . . . . 20 5.4.1 Timestamp Option . . . . . . . . . . . . . . . . . . . . . 20 5.4.2 Nonce Option . . . . . . . . . . . . . . . . . . . . . . . 21 5.4.3 Processing rules for senders . . . . . . . . . . . . . . . 22 5.4.4 Processing rules for receivers . . . . . . . . . . . . . . 22 5.5 Proxy Neighbor Discovery . . . . . . . . . . . . . . . . . 24 6. Authorization Delegation Discovery . . . . . . . . . . . . 25 6.1 Delegation Chain Solicitation Message Format . . . . . . . 25 6.2 Delegation Chain Advertisement Message Format . . . . . . 27 6.3 Trust Anchor Option . . . . . . . . . . . . . . . . . . . 29 6.4 Certificate Option . . . . . . . . . . . . . . . . . . . . 30 6.5 Router Authorization Certificate Format . . . . . . . . . 31 6.5.1 Router Authorization Certificate Profile . . . . . . . . . 31 6.6 Processing Rules for Routers . . . . . . . . . . . . . . . 32 6.7 Processing Rules for Hosts . . . . . . . . . . . . . . . . 34 7. Securing Neighbor Discovery with SEND . . . . . . . . . . 37 7.1 Neighbor Solicitation Messages . . . . . . . . . . . . . . 37 7.1.1 Sending Secure Neighbor Solicitations . . . . . . . . . . 37 7.1.2 Receiving Secure Neighbor Solicitations . . . . . . . . . 37 7.2 Neighbor Advertisement Messages . . . . . . . . . . . . . 37 7.2.1 Sending Secure Neighbor Advertisements . . . . . . . . . . 37 7.2.2 Receiving Secure Neighbor Advertisements . . . . . . . . . 38 7.3 Other Requirements . . . . . . . . . . . . . . . . . . . . 38 8. Securing Router Discovery with SEND . . . . . . . . . . . 40 8.1 Router Solicitation Messages . . . . . . . . . . . . . . . 40 8.1.1 Sending Secure Router Solicitations . . . . . . . . . . . 40 8.1.2 Receiving Secure Router Solicitations . . . . . . . . . . 40 8.2 Router Advertisement Messages . . . . . . . . . . . . . . 41 8.2.1 Sending Secure Router Advertisements . . . . . . . . . . . 41 Arkko, et al. Expires April 16, 2004 [Page 2] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 8.2.2 Receiving Secure Router Advertisements . . . . . . . . . . 41 8.3 Redirect Messages . . . . . . . . . . . . . . . . . . . . 41 8.3.1 Sending Redirects . . . . . . . . . . . . . . . . . . . . 41 8.3.2 Receiving Redirects . . . . . . . . . . . . . . . . . . . 42 8.4 Other Requirements . . . . . . . . . . . . . . . . . . . . 42 9. Co-Existence of SEND and non-SEND nodes . . . . . . . . . 43 10. Performance Considerations . . . . . . . . . . . . . . . . 45 11. Security Considerations . . . . . . . . . . . . . . . . . 46 11.1 Threats to the Local Link Not Covered by SEND . . . . . . 46 11.2 How SEND Counters Threats to Neighbor Discovery . . . . . 47 11.2.1 Neighbor Solicitation/Advertisement Spoofing . . . . . . . 47 11.2.2 Neighbor Unreachability Detection Failure . . . . . . . . 48 11.2.3 Duplicate Address Detection DoS Attack . . . . . . . . . . 48 11.2.4 Router Solicitation and Advertisement Attacks . . . . . . 49 11.2.5 Replay Attacks . . . . . . . . . . . . . . . . . . . . . . 49 11.2.6 Neighbor Discovery DoS Attack . . . . . . . . . . . . . . 49 11.3 Attacks against SEND Itself . . . . . . . . . . . . . . . 50 12. IANA Considerations . . . . . . . . . . . . . . . . . . . 51 Normative References . . . . . . . . . . . . . . . . . . . 52 Informative References . . . . . . . . . . . . . . . . . . 54 Authors' Addresses . . . . . . . . . . . . . . . . . . . . 55 A. Contributors . . . . . . . . . . . . . . . . . . . . . . . 57 B. IPR Considerations . . . . . . . . . . . . . . . . . . . . 58 C. Cache Management . . . . . . . . . . . . . . . . . . . . . 59 D. Comparison to AH-Based Approach . . . . . . . . . . . . . 60 Intellectual Property and Copyright Statements . . . . . . 63 Arkko, et al. Expires April 16, 2004 [Page 3] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 1. Introduction IPv6 defines the Neighbor Discovery Protocol (NDP) in RFC 2461 [6]. Nodes on the same link use NDP to discover each other's presence, to determine each other's link-layer addresses, to find routers, and to maintain reachability information about the paths to active neighbors. NDP is used both by hosts and routers. Its functions include Neighbor Discovery (ND), Router Discovery (RD), Address Autoconfiguration, Address Resolution, Neighbor Unreachability Detection (NUD), Duplicate Address Detection (DAD), and Redirection. RFC 2461 called for the use of IPsec for protecting the NDP messages. However, it does not specify detailed instructions for using IPsec to secure NDP. It turns out that in this particular application, IPsec can only be used with a manual configuration of security associations, due to chicken-and-egg problems in using IKE [22] [19]. Furthermore, the number of such manually configured security associations needed for protecting NDP can be very large [23], making that approach impractical for most purposes. This document is organized as follows. Section 4 describes the overall approach to securing NDP. This approach involves the use of new NDP options to carry public-key based signatures. A zero-configuration mechanism is used for showing address ownership on individual nodes; routers are certified by a trust anchor [11]. The formats, procedures, and cryptographic mechanisms for the zero-configuration mechanism are described in a related specification [26]. Section 6 describes the mechanism for distributing certificate chains to establish an authorization delegation chain to a common trust anchor. The required new NDP options are discussed in Section 5. Section 7 and Section 8 show how to apply these components to securing Neighbor and Router Discovery. Finally, Section 9 discusses the co-existence of secure and non-secure Neighbor Discovery on the same link, Section 10 discusses performance considerations, and Section 11 discusses security considerations for Secure Neighbor Discovery. Arkko, et al. Expires April 16, 2004 [Page 4] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 2. Terms Authorization Delegation Discovery (ADD) A process through which SEND nodes can acquire a certificate chain from a peer node to a trust anchor. Cryptographically Generated Addresses (CGAs) A technique [26] [30] where the IPv6 address of a node is cryptographically generated using a one-way hash function from the node's public key and some other parameters. Duplicate Address Detection (DAD) A mechanism defined in RFC 2462 [7] that assures that two IPv6 nodes on the same link are not using the same addresses. Internet Control Message Protocol version 6 (ICMPv6) The IPv6 control signaling protocol. Neighbor Discovery is a part of ICMPv6. Neighbor Discovery Protocol (NDP) The IPv6 Neighbor Discovery Protocol [6]. Neighbor Discovery (ND) The Neighbor Discovery function of the Neighbor Discovery Protocol (NDP). NDP contains also other functions but ND. Neighbor Unreachability Detection (NUD) This mechanism defined in RFC 2461 [6] is used for tracking the reachability of neighbors. Nonce A random number generated by a node and used exactly once, and never again. In SEND, nonces are used to ensure that a particular advertisement is linked to the solicitation that triggered it. Router Authorization Certificate An X.509v3 [11] PKC certificate using the profile specified in Section 6.5.1. Arkko, et al. Expires April 16, 2004 [Page 5] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 SEND node An IPv6 node that implements this specification. non-SEND node An IPv6 node that does not implement this specification but uses the legacy RFC 2461 and RFC 2462 mechanisms. Router Discovery (RD) The Router Discovery function of the Neighbor Discovery Protocol (NDP). Arkko, et al. Expires April 16, 2004 [Page 6] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 3. Neighbor and Router Discovery Overview IPv6 Neighbor and Router Discovery have several functions. Many of these functions are overloaded on a few central message types, such as the ICMPv6 Neighbor Discovery message. In this section we review some of these tasks and their effects in order to understand better how the messages should be treated. This section is not normative, and if this section and the original Neighbor Discovery RFCs are in conflict, the original RFCs take precedence. In IPv6, many of the tasks traditionally preformed at lower the layers, such as ARP, have been moved to the IP layer. As a consequence, a set of unified mechanisms can be applied across link layers, any introduced security mechanisms or other extensions can be adopted more easily, and a clear separation of the roles between the IP and link layer has been achieved. The main functions of IPv6 Neighbor Discovery are the following. o Neighbor Unreachability Detection (NUD) is used for tracking the reachability of neighboring nodes, both hosts and routers. NUD is defined in Section 7.3 of RFC 2461 [6]. NUD is security-sensitive, because an attacker could falsely claim that reachability exists when it in fact does not. o Duplicate Address Detection (DAD) is used for preventing address collisions [7]. A node that intends to assign a new address to one of its interfaces first runs the DAD procedure to verify that there is no other node using the same address. Since the rules forbid the use of an address until it has been found unique, no higher layer traffic is possible until this procedure has been completed. Thus, preventing attacks against DAD can help ensure the availability of communications for the node in question. o Address Resolution is similar to IPv4 ARP [18]. The Address Resolution function resolves a node's IPv6 address to the corresponding link-layer address for nodes on the link. Address Resolution is defined in Section 7.2 of RFC 2461 [6], and it is used for hosts and routers alike. Again, no higher level traffic can proceed until the sender knows the hardware address of the destination node or the next hop router. Note that like its predecessor in ARP, IPv6 Neighbor Discovery does not check the source link layer address against the information learned through Address Resolution. This allows for an easier addition of network elements such as bridges and proxies, and eases the stack implementation requirements as less information needs to be passed from layer to layer. Arkko, et al. Expires April 16, 2004 [Page 7] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 o Address Autoconfiguration is used for automatically assigning addresses to a host [7]. This allows hosts to operate without explicit configuration related to IP connectivity. The Address Autoconfiguration mechanism defined in [7] is stateless. To create IP addresses, the hosts use any prefix information delivered to them during Router Discovery, and then test the newly formed addresses for uniqueness using the DAD procedure. A stateful mechanism, DHCPv6 [24], provides additional Autoconfiguration features. Router and Prefix Discovery and Duplicate Address Detection have an effect on the Address Autoconfiguration tasks. o The Redirect function is used for automatically redirecting hosts to an alternate router. Redirect is specified in Section 8 of RFC 2461 [6]. It is similar to the ICMPv4 Redirect function [17]. o The Router Discovery function allows IPv6 hosts to discover the local routers on an attached link. Router Discovery is described in Section 6 of RFC 2461 [6]. The main purpose of Router Discovery is to find neighboring routers that are willing to forward packets on behalf of hosts. Prefix discovery involves determining which destinations are directly on a link; this information is necessary in order to know whether a packet should be sent to a router or to the destination node directly. Typically, address autoconfiguration and other tasks can not proceed until suitable routers and prefixes have been found. The Neighbor Discovery messages follow the ICMPv6 message format. They have ICMPv6 types from 133 to 137. The IPv6 Next Header value for ICMPv6 is 58. The actual Neighbor Discovery message includes an NDP message header, consisting of an ICMPv6 header and ND message-specific data, and zero or more NDP options. <------------NDP Message----------------> *-------------------------------------------------------------* | IPv6 Header | ICMPv6 | ND message- | ND Message | | Next Header = 58 | Header | specific | Options | | (ICMPv6) | | data | | *-------------------------------------------------------------* <--NDP Message header--> The NDP message options are formatted in the Type-Length-Value format. All IPv6 NDP functions are realized using the following ICMPv6 messages: ICMPv6 Type Message ------------------------------------ Arkko, et al. Expires April 16, 2004 [Page 8] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 133 Router Solicitation (RS) 134 Router Advertisement (RA) 135 Neighbor Solicitation (NS) 136 Neighbor Advertisement (NA) 137 Redirect The various functions are realized using these messages as follows: o Router Discovery uses the RS and RA messages. o Duplicate Address Detection uses the NS and NA messages. o Address Autoconfiguration uses the NS, NA, RS, and RA messages. o Address Resolution uses the NS and NA messages. o Neighbor Unreachability Detection uses the NS and NA messages. o Redirect uses the Redirect message. The NDP messages are always meant to be used within a link, and never intended to leak outside of it. The destination and source addresses used in these messages are as follows: o Neighbor Solicitation: The destination address is either the Solicited-Node multicast address, a unicast address, or an anycast address. The source address is either the unspecified address (in DAD) or a unicast address assigned to the sending interface. In a typical case, the source address is equal to the source address of the outgoing packet, locally triggering the need to send the solicitation. o Neighbor Advertisement: The destination address is either a unicast address or the link-scoped All-Nodes multicast address [12]. The source address is a unicast address assigned to the sending interface. o Router Solicitation: The destination address is typically the All-Routers multicast address [12]. The source address is either the unspecified address or a unicast address assigned to the sending interface. An unspecified source address does not have any special semantics; it is just an optimization for startup. o Router Advertisement: The destination address can be either a unicast or the link-scoped All-Nodes multicast address [12]. The source address is a link-local address assigned to the sending interface. Arkko, et al. Expires April 16, 2004 [Page 9] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 o Redirect: This message is always sent to the source address of the packet that triggered the Redirect. Hosts verify that the IP source address of the Redirect is the same as the current first-hop router for the specified ICMP Destination Address. Rules in [12] dictate that anycast, or multicast addresses may not be used as source addresses. If the source address is an unspecified address, it is impossible to send a Redirect, since the unspecified address is forbidden as the destination address. Therefore, the destination address must always be a unicast address. The source address is a link-local address assigned to the sending interface. Arkko, et al. Expires April 16, 2004 [Page 10] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 4. Secure Neighbor Discovery Overview To secure the various functions, a set of new Neighbor Discovery options introduced. They are used in to protect Neighbor and Router Discovery messages. This specification introduces these options, an authorization delegation discovery process, an address ownership proof mechanism, and requirements for the use of these components for Neighbor Discovery. The components of the solution specified in this document are as follows: o Certificate chains, anchored on trusted parties, are expected to certify the authority of routers. A host and a router must have at least one common trust anchor before the host can adopt the router as its default router. Delegation Chain Solicitation and Advertisement messages are used to discover a certificate chain to the trust anchor without requiring the actual Router Discovery messages to carry lengthy certificate chains. o Cryptographically Generated Addresses are used to assure that the sender of a Neighbor or Router Advertisement is the "owner" of the claimed address. A public-private key pair needs to be generated by all nodes before they can claim an address. A new Neighbor Discovery option, the CGA option, is used to carry the public key and associated parameters. This specification also allows one to use non-CGA addresses and to use certificates to authorized their use. However, the details of such use have been left for future work. o A new Neighbor Discovery option, the Signature option, is used to protect all messages relating to Neighbor and Router discovery. Public key signatures are used to protect the integrity of the messages and to authenticate the identity of their sender. The authority of a public key is established either with the authorization delegation process, using certificates, or through the address ownership proof mechanism, using CGAs, or both, depending on configuration and the type of the message protected. o In order to prevent replay attacks, two new Neighbor Discovery options, Timestamp and Nonce, are used. Given that Neighbor and Router Discovery messages are in some cases sent to multicast addresses, the Timestamp option offers replay protection without any previously established state or sequence numbers. When the messages are used in solicitation - advertisement pairs, they protected using the Nonce option. Arkko, et al. Expires April 16, 2004 [Page 11] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 5. Neighbor Discovery Options The following new NDP options and mechanisms are REQUIRED to be implemented by all SEND nodes: o The CGA option MAY be present in all Neighbor Discovery messages, and SHOULD be present in most cases. o The Signature option is REQUIRED in all Neighbor Discovery messages. o The Nonce option is REQUIRED in all Neighbor Discovery solicitations, and in all solicited advertisements. o The Timestamp option is REQUIRED in all Neighbor Discovery advertisements and Redirects. o Proxy Neighbor Discovery is not supported by this specification; it is planned to be specified in a future document. 5.1 Ordering of the new options The ordering of the new options MUST obey the following rules: The CGA option MUST appear before the Signature option. The Nonce option SHOULD appear before the Timestamp option. The Signature option MUST NOT be be followed CGA, Nonce, or Timestamp options. It is RECOMMENDED that the options appear in the following order: CGA, Nonce, Timestamp, Signature. 5.2 CGA Option The CGA option allows the verification of the sender's CGA. The format of the CGA option is described as follows. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Modifier | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Collision Cnt | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Arkko, et al. Expires April 16, 2004 [Page 12] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 | | . . . Key Information . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . Padding . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The meaning of the fields is described as follows. Type TBD for CGA. Length The length of the option, in units of 8 octets. Modifier A random number used in CGA generation. Its semantics are defined in [26]. Collision Cnt An 8-bit collision count, which can get values 0, 1 and 2. Its semantics are defined in [26]. Reserved A 24-bit field reserved for future use. The value MUST be initialized to zero by the sender, and MUST be ignored by the receiver. Key Information A variable length field containing the public key of the sender, represented as an ASN.1 type SubjectPublicKeyInfo [11], encoded as described in Section 4 of [26]. This specification requires that if both the CGA option and the Signature option are present, then the publicKey field in the former option MUST be the public key referred by the Key Hash Arkko, et al. Expires April 16, 2004 [Page 13] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 field in the latter option. Packets received with two different keys MUST be silently discarded. Note that a future extension may provide a mechanism which allows the owner of an address and the signer to be different parties. The length of the Key Information field is determined by the ASN.1 encoding. Padding A variable length field making the option length a multiple of 8. It begins after the ASN.1 encoding of the previous field has ends, and continues to the end of the option, as specified by the Length field. 5.2.1 Processing Rules for Senders A node sending a message using the CGA option MUST construct the message as follows. The Modifier, Collision Cnt, and Key Information fields in the CGA option are filled in according to the rules presented above and in [26]. The used public key is taken from configuration; typically from a data structure associated with the source address. An address MUST be constructed as specified in Section 4 of [26]. In the typical case, the address is constructed long before it is used. Depending on the type of the message, this address appears in different places: Redirect The address MUST be the source address of the message. Neighbor Solicitation The address MUST be the Target Address for solicitations sent for the purpose of Duplicate Address Detection, and the source address of the message otherwise. Neighbor Advertisement The address MUST be the source address of the message. Arkko, et al. Expires April 16, 2004 [Page 14] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Router Solicitation The address MUST be the source address of the message, unless the source address is the unspecified address. Router Advertisement The address MUST be the source address of the message. 5.2.2 Processing Rules for Receivers A message containing a CGA option MUST be checked as follows: If the interface has been configued to use CGA, it is REQUIRED that the receiving node verifies the source address of the packet using the algorithm described in Section 5 of [26]. The inputs for the algorithm are the contents of the Modifier, Collision Cnt, and the Key Information fields, the claimed address in the packet (as discussed in the previous section), and the minimum acceptable Sec value. If the CGA verification is successful, the recipient proceeds with the cryptographically more time consuming check of the signature. Note that a receiver which does not support CGA or has not specified its use for a given interface can still verify packets using trust anchors, even if CGA had been used on a packet. In such a case, the CGA property of the address is simply left unverified. 5.2.3 Configuration All nodes that support the verification of the CGA option MUST record the following configuration information: minbits The minimum acceptable key length for the public keys used in the generation of the CGA address. The default SHOULD be 1024 bits. Implementations MAY also set an upper limit in order to limit the amount of computation they need to perform when verifying packets that use these security associations. Any implementation should follow prudent cryptographic practise in determining the appropriate key lengths. 5.3 Signature Option The Signature option allows public-key based signatures to be Arkko, et al. Expires April 16, 2004 [Page 15] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 attached to NDP messages. Both trust anchor authentication and CGAs can be used. The format of the Signature option is described in the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Pad Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | Key Hash | | | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . Digital Signature . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | . . . Padding . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ The meaning of the fields is described below: Type TBD for Signature. Length The length of the option, in units of 8 octets. Pad Length An 8-bit integer field, giving the length of the Pad field in units of an octet. Reserved An an 8-bit field reserved for future use. The value MUST be initialized to zero by the sender, and MUST be ignored by the receiver. Arkko, et al. Expires April 16, 2004 [Page 16] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Key Hash A 128-bit field contains the most significant (leftmost) 128-bits of a SHA1 hash of the public key used for the constructing the signature. The SHA1 is taken over the presentation used in the Key Information field in the CGA option. Its purpose is to associate the signature to a particular key known by the receiver. Such a key can be either stored in the certificate cache of the receiver, or be received in the CGA option in the same message. Digital Signature A variable length field contains the signature constructed using the sender's private key, over the the following sequence of octets: 1. The 128-bit CGA Type Tag [26] value for SEND, 0xXXXX XXXX XXXX XXXX XXXX XXXX XXXX XXXX (To be generated randomly). 2. The 128-bit Source Address field from the IP header. 3. The 128-bit Destination Address field from the IP header. 4. The 32-bit ICMP header, i.e., the Type, Code, and Checksum fields. 5. The Neighbor Discovery message header, i.e., the Reserved field in the Router Solicitation message, the Cur Hop Limit, M, O, Reserved, Router Lifetime, Reachable Time, and Retrans Timer fields in the Router Advertisement message, Reserved and Target Address fields in the Neighbor Solicitation message, R, S, O, Reserved, and Target Address fields in the Neighbor Advertisement message, and Reserved, Target Address, and Destination Address fields in the Redirect message. 6. All NDP options preceding the Signature option. The signature is constructed using the RSA algorithm and MUST be encoded as private key encryption in PKCS#1 format [15]. The signature value is computed with the RSASSA-PKCS1-v2_1 algorithm and SHA-1 hash as defined in [15]. This field starts after the Key Hash field. The length of the Digital Signature field is determined by the length of the Signature option minus the length of the other fields (including the variable length Pad field). Arkko, et al. Expires April 16, 2004 [Page 17] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 This variable length field contains padding, as many bytes as is given by the Pad Length Field. 5.3.1 Processing Rules for Senders A node sending a message using the Signature option MUST construct the message as follows: o The message is constructed in its entirety. o The Signature option is added as the last option in the message. o For the purpose of constructing a signature, the following data items are concatenated: * The 128-bit CGA Type Tag. * The source address of the message. * The destination address of the message. * The contents of the message, starting from the ICMPv6 header, up to but excluding the Signature option. o The message, in the form defined above, is signed using the configured private key, and the resulting PKCS#1 signature is put to the Digital Signature field. 5.3.2 Processing Rules for Receivers A message containing a Signature option MUST be checked as follows: o The Signature option MUST appear as the last option. o The Key Hash field MUST indicate the use of a known public key, either one learned from a preceeding CGA option, or one known by other means. o TheDigital Signature field MUST have correct encoding, and do not exceed the length of the Signature option. o The Digital Signature verification MUST show that the signature has been calculated as specified in the previous section. o If the use of a trust anchor has been configured, a valid authorization delegation chain MUST be known between the Arkko, et al. Expires April 16, 2004 [Page 18] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 receiver's trust anchor and the sender's public key. Note that the receiver may verify just the CGA property of a packet, even if, in addition to CGA, the sender has used a trust anchor. Messages that do not pass all the above tests MUST be silently discarded. The receiver MAY silently drop packets also otherwise, e.g., as a response to an apparent CPU exhausting DoS attack. 5.3.3 Configuration All nodes that support the reception of the Signature options MUST record the following configuration information for each separate Neighbor Discovery Protocol message type: authorization method This parameter determines the method through which the authority of the sender is determined. It can have four values: trust anchor The authority of the sender is verified as described in Section 6.5. The sender may claim additional authorization through the use of CGAs, but that is neither required nor verified. CGA The CGA property of the sender's address is verified as described in [26]. The sender may claim additional authority through a trust anchor, but that is neither required nor verified. trust anchor and CGA Both the trust anchor and the CGA verification is required. trust anchor or CGA Either the trust anchor or the CGA verification is required. anchor The public keys of the allowed trust anchor(s), if authorization method is not set to CGA. Arkko, et al. Expires April 16, 2004 [Page 19] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 minSec The minimum acceptable Sec value, if CGA verification is required (see Section 2 in [26]). This parameter is intended to facilitate future extensions and experimental work. Currently, the minSec value SHOULD always be set to zero. All nodes that support the sending of Signature options MUST record the following configuration information: keypair A public-private key pair. If authorization delegation is in use, there must exist a delegation chain from a trust anchor to this key pair. CGA flag A flag that indicates whether CGA is used or is not used. This flag may be per interface or per node. CGA parameters Optionally any information required to construct CGAs, including the used Sec and Modifier values, and the CGA address itself. 5.4 Timestamp and Nonce options 5.4.1 Timestamp Option The purpose of the Timestamp option is to ensure that unsolicited advertisements and redirects have not been replayed. The format of the Timestamp option is described in the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Timestamp + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Where the fields are as follows: Arkko, et al. Expires April 16, 2004 [Page 20] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Type TBD for Timestamp. Length The length of the option, in units of 8 octets, i.e., 2. Reserved A 48-bit field reserved for future use. The value MUST be initialized to zero by the sender, and MUST be ignored by the receiver. Timestamp A 64-bit unsigned integer field containing a timestamp. The value indicates the number of seconds since January 1,, 1970 00:00 UTC, using a fixed point format. In this format the integer number of seconds is contained in the first 48 bits of the field, and the remaining 16 bits indicate the number of 1/64K fractions of a second. 5.4.2 Nonce Option The purpose of the Nonce option is to ensure that an advertisement is a fresh response to a solicitation sent earlier by the receiving same node. The format of the Nonce option is as described in the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Nonce ... | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | . . . . | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Where the fields are as follows: Type TBD for Nonce. Arkko, et al. Expires April 16, 2004 [Page 21] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Length The length of the option, in units of 8 octets. Nonce A field containing a random number selected by the sender of the solicitation message. The length of the random number MUST be at least 6 bytes. 5.4.3 Processing rules for senders All solicitation messages MUST include a Nonce. All solicited-for announcements MUST include a Nonce, copying the nonce value from the received solicitation. When sending a solication, the sender MUST store the nonce internally so that it can recognize any replies containing that particular nonce. All NDP messages MUST include a Timestamp. Senders SHOULD set the Timestamp field to the current time, according to their real time clock. If a message has both Nonce and Timestamp options, the Nonce option SHOULD precede the Timestamp option in order. The receiver MUST be prepared to receive them in any order, as per RFC 2461 [6] Section 9. 5.4.4 Processing rules for receivers The processing of the Nonce and Timestamp options depends on whether a packet is a solicited-for advertisement or not. A system may implement the distinction in various means. Section 5.4.4.1 defines the processing rules for solicited-for advertisements. Section 5.4.4.2 defines the processing rules for all other messages. An implementation may utilize some mechanism such as a timestamp cache to strengthen resistance to replay attacks. When there is a very large number of nodes on the same link, or when a cache filling attack is in progress, it is possible that the cache holding the most recent timestamp per sender becomes full. In this case the node MUST remove some entries from the cache or refuse some new requested entries. The specific policy as to which entries are preferred over the others is left as an implementation decision. However, typical policies may prefer existing entries over new ones, CGAs with a large Sec value over smaller Sec values, and so on. The issue is briefly discussed in Appendix C. Arkko, et al. Expires April 16, 2004 [Page 22] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 5.4.4.1 Processing solicited-for advertisements The receiver MUST verify that it has recently send a matching solicitation, and that the received advertisement does contain a copy of the Nonce sent in the solicitation. If the message does not contain a Nonce option, it MAY be considered as a non-solicited-for announcement, and processed according to Section 5.4.4.2. If the message does contain a Nonce option, but the Nonce value is not recognized, the message MUST be silently dropped. If the message is accepted, the receiver SHOULD store the receive time of the message and the time stamp time in the message, as specified in Section 5.4.4.2 5.4.4.2 Processing all other messages Receivers SHOULD be configured with an allowed timestamp Delta value and an allowed clock drift parameter. The recommended default value for the allowed Delta is 3,600 seconds (1 hour) and for clock dritf 1% (0.01). To facilitate timestamp checking, each node SHOULD store the following information per each peer: The receive time of the last received, acepted SEND message. This is called RDlast. The time stamp in the last received, accepted SEND message. This is called TSlast. Receivers SHOULD then check the Timestamp field as follows: o When a message is received from a new peer, i.e., one that is not stored in the cache, the received timestamp, TSnew, is checked and the packet is accepted if the timestamp is recent enough with respect to the receival time of the packet, RDnew: -Delta < (RDnew - TSnew) < +Delta The RDnew and TSnew values SHOULD be stored into the cache as RDlast and TSlast. o If the timestamp is NOT within the boundaries but the message is a Neighbor Solicitation message that should be responded to by the receiver, the receiver MAY respond to the message. However, if it Arkko, et al. Expires April 16, 2004 [Page 23] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 does respond to the message, it MUST NOT create a neighbor cache entry. This allows nodes that have large difference in their clocks to still communicate with each other, by exchanging NS/NA pairs. o When a message is received from a known peer, i.e., one that already has an entry in the cache, the time stamp is checked against the previously received SEND message: TSnew > TSlast + (RDnew - RDlast) x (1 - drift) o If TSnew < TSlast, which is possible if packets arrive rapidly and out of order, TSlast MUST NOT be updated, i.e., the stored TSlast for a given node MUST NOT ever decrease. Otherwise TSlast SHOULD be updated. Independent on whether TSlast is updated or not, RDlast is updated in any case. 5.5 Proxy Neighbor Discovery The Target Address in Neighbor Advertisement is required to be equal to the source address of the packet, except in the case of proxy Neighbor Discovery. Proxy Neighbor Discovery is not supported by this specification; it is planned to be specified in a future document. Arkko, et al. Expires April 16, 2004 [Page 24] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 6. Authorization Delegation Discovery Several protocols, including the IPv6 Neighbor Discovery Protocol, allow a node to automatically configure itself based on information it learns shortly after connecting to a new link. It is particularly easy to configure "rogue" routers on an unsecured link, and it is particularly difficult for a node to distinguish between valid and invalid sources of information, when the node needs this information before being able to communicate with nodes outside of the link. Since the newly-connected node cannot communicate off-link, it can not be responsible for searching information to help validating the router(s); however, given a chain of appropriately signed certificates, it can check someone else's search results and conclude that a particular message comes from an authorized source. In the typical case, a router, which is already connected to beyond the link, can (if necessary) communicate with off-link nodes and construct such a certificate chain. The Secure Neighbor Discovery Protocol introduces two new ICMPv6 messages that are used between hosts and routers to allow the host to learn a certificate chain with the assistance of the router. Where hosts themselves are certified by a trust anchor, these messages MAY also optionally be used between hosts to acquire the peer's certificate chain. However, the details of such usage are left for future specification. The Delegation Chain Solicitation (DCS) message is sent by a host when it wishes to request a certificate chain between a router and the one of the host's trust anchors. The Delegation Chain Advertisement (DCA) message is sent as an answer to the DCS message. It MAY be periodically sent to the link-scoped All-Nodes multicast address. These messages are separate from the rest of Neighbor and Router Discovery, in order to reduce the effect of the potentially voluminous certificate chain information on other messages. The Authorization Delegation Discovery (ADD) process does not exclude other forms of discovering certificate chains. For instance, during fast movements mobile nodes may learn information - including the certificate chains - of the next router from a previous router. 6.1 Delegation Chain Solicitation Message Format Hosts send Delegation Chain Solicitations in order to prompt routers to generate Delegation Chain Advertisements quickly. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 Arkko, et al. Expires April 16, 2004 [Page 25] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Identifier | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Options ... +-+-+-+-+-+-+-+-+-+-+-+- IP Fields: Source Address An IP address assigned to the sending interface, or the unspecified address if no address is assigned to the sending interface. Destination Address Typically the All-Routers multicast address, the Solicited-Node multicast address, or the address of the host's default router. Hop Limit 255 ICMP Fields: Type TBD for Delegation Chain Solicitation. Code 0 Checksum The ICMP checksum [8]. Identifier A 16-bit unsigned integer field, acting as an identifier to help matching advertisements to solicitations. The Identifier field MUST NOT be zero, and its value SHOULD be randomly generated. (This randomness does not need to be cryptographically hard, though. Its purpose is to avoid collisions.) Arkko, et al. Expires April 16, 2004 [Page 26] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Reserved An unused field. It MUST be initialized to zero by the sender and MUST be ignored by the receiver. Valid Options: Trust Anchor One or more trust anchors that the client is willing to accept. The first (or only) Trust Anchor option MUST contain a DER Encoded X.501 Name; see Section 6.3. If there are more than one Trust Anchor options, the options past the first one may contain any types of Trust Anchors. Future versions of this protocol may define new option types. Receivers MUST silently ignore any options they do not recognize and continue processing the message. 6.2 Delegation Chain Advertisement Message Format Routers send out Delegation Chain Advertisement messages periodically, or in response to a Delegation Chain Solicitation. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Code | Checksum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Identifier | Component | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Options ... +-+-+-+-+-+-+-+-+-+-+-+- IP Fields: Source Address MUST be a unicast address assigned to the interface from which this message is sent. Destination Address Either the Solicited-Node multicast address of the receiver or the link-scoped All-Nodes multicast address. Arkko, et al. Expires April 16, 2004 [Page 27] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Hop Limit 255 ICMP Fields: Type TBD for Delegation Chain Advertisement. Code 0 Checksum The ICMP checksum [8]. Identifier A 16-bit unsigned integer field, acting as an identifier to help matching advertisements to solicitations. The Identifier field MUST be zero for unsolicited advertisements and MUST NOT be zero for solicited advertisements. Component A 16-bit unsigned integer field, used for informing the receiver which certificate is being sent, and how many are still left to be sent in the whole chain. A single advertisement MUST be broken into separately sent components if there is more than one Certificate option, in order to avoid excessive fragmentation at the IP layer. Unlike the fragmentation at the IP layer, individual components of an advertisement may be stored and used before all the components have arrived; this makes them slightly more reliable and less prone to Denial-of-Service attacks. The first message in a N-component advertisement has the Component field set to N-1, the second set to N-2, and so on. Zero indicates that there are no more components coming in this advertisement. The components MUST be ordered so that the trust anchor end of the chain is the one sent first. Each certificate sent after it can be verified with the previously sent certificates. The Arkko, et al. Expires April 16, 2004 [Page 28] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 certificate of the sender comes last. Reserved An unused field. It MUST be initialized to zero by the sender and MUST be ignored by the receiver. Valid Options: Certificate One certificate is provided in each Certificate option, to establish a (part of a) certificate chain to a trust anchor. Trust Anchor Zero or more Trust Anchor options may be included to help receivers decide which advertisements are useful for them. If present, these options MUST appear in the first component of a multi-component advertisement. Future versions of this protocol may define new option types. Receivers MUST silently ignore any options they do not recognize and continue processing the message. 6.3 Trust Anchor Option The format of the Trust Anchor option is as described in the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Name Type | Pad Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Name ... +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Where the fields are as follows: Type TBD for Trust Anchor. Length The length of the option, (including the Type, Length, Name Type, Arkko, et al. Expires April 16, 2004 [Page 29] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Name Length, and Name fields,) in units of 8 octets. Name Type The type of the name included in the Name field. This specification defines only one legal value for this field: 1 DER Encoded X.501 Name 2 FQDN Pad Length The number of padding octets beyond the end of the Name field but within the length specified by the Length field. Padding octets MUST be set to zero by senders and ignored by receivers. Name When the Name Type field is set to 1, the Name field contains a DER encoded X.501 certificate Name, represented and encoded exactly as in the matching X.509v3 trust anchor certificate. When the Name Type field is set to 2, the Name field contains a Fully Qualified Domain Name of the trust anchor, for example, "trustanchor.example.com". The name is stored as a string, in the "preferred name syntax" DNS format, as specified in RFC 1034 [1] Section 3.5. Additionally, the restrictions discussed in RFC 3280 [11] Section 4.2.1.7 apply. All systems MUST implement support the DER Encoded X.501 Name. Implementations MAY support the FQDN name type. 6.4 Certificate Option The format of the certificate option is as described in the following: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Cert Type | Pad Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Certificate ... +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Where the fields are as follows: Arkko, et al. Expires April 16, 2004 [Page 30] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Type TBD for Certificate. Length The length of the option, (including the Type, Length, Cert Type, Pad Length, and Certificate fields,) in units of 8 octets. Cert Type The type of the certificate included in the Certificate field. This specification defines only one legal value for this field: 1 X.509v3 Certificate, as specified below Pad Length The number of padding octets beyond the end of the Certificate field but within the length specified by the Length field. Padding octets MUST be set to zero by senders and ignored by receivers. Certificate When the Cert Type field is set to 1, the Certificate field contains an X.509v3 certificate [11], as described in Section 6.5.1. 6.5 Router Authorization Certificate Format The certificate chain of a router terminates in a Router Authorization Certificate that authorizes a specific IPv6 node to act as a router. Because authorization chains are not a common practice in the Internet at the time this specification is being written, the chain MUST consist of standard Public Key Certificates (PKC, in the sense of [21]). The certificates chain MUST start from the identity of a trust anchor that is shared by the host and the router. This allows the host to anchor trust for the router's public key in the trust anchor. Note that there MAY be multiple certificates issued by a single trust anchor. 6.5.1 Router Authorization Certificate Profile Router Authorization Certificates be X.509v3 certificates, as defined in RFC 3280 [11], and MUST contain at least one instance of the X.509 extension for IP addresses, as defined in [13]. The parent certificates in the certificate chain MUST contain one or more X.509 Arkko, et al. Expires April 16, 2004 [Page 31] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 IP address extensions, back up to the delegating authority (the Regional Address Registry or IANA) that delegated the original IP address space block. The certificates for intermediate delegating authorities MUST contain X.509 IP address extension(s) for subdelegations. The router's certificate is signed by the delegating authority for the prefixes the router is authorized to to advertize. The X.509 IP address extension MUST contain at least one addressesOrRanges element that contains an addressPrefix element with an IPv6 address prefix for a prefix the router or the intermediate entity is authorized to advertize. If the entity is allowed to route any prefix, the used IPv6 address prefix is the null prefix, 0/0. The addressFamily element of the containing IPAddrBlocks sequence element MUST contain the IPv6 AFI (0002), as specified in [13] for IPv6 prefixes. Instead of an addressPrefix element, the addressesOrRange element MAY contain an addressRange element for a range of prefixes, if more than one prefix is authorized. The X.509 IP address extension MAY contain additional IPv6 prefixes, expressed either as an addressPrefix or an addressRange. A SEND node receiving a Router Authorization Certificate MUST first check whether the certificate's signature was generated by the delegating authority. Then the client MUST check whether all the addressPrefix or addressRange entries in the router's certificate are contained within the address ranges in the delegating authority's certificate, and whether the addressPrefix entries match any addressPrefix entries in the delegating authority's certificate. If an addressPrefix or addressRange is not contained within the delegating authority's prefixes or ranges, the client MAY attept to take an intersection of the ranges/prefixes, and use that intersection. If the addressPrefix in the certificate is the null prefix, 0/0, such an intersection SHOULD be used. (In that case the intersection is the parent prefix or range.) If the resulting intersection is empty, the client MUST NOT accept the certificate. The above check SHOULD be done for all certificates in the chain received through DCA messages. If any of the checks fail, the client MUST NOT accept the certificate. Since it is possible that some PKC certificates used with SEND do not immediately contain the X.509 IP address extension element, an implementation MAY contain facilities that allow the prefix and range checks to be relaxed. However, any such configuration options SHOULD be off by default. That is, the system SHOULD have a default configuration that requires rigorious prefix and range checks. 6.6 Processing Rules for Routers Arkko, et al. Expires April 16, 2004 [Page 32] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Routers SHOULD possess a key pair and a certificate from at least one certificate authority. A router MUST silently discard any received Delegation Chain Solicitation messages that do not satisfy all of the following validity checks: o The IP Hop Limit field MUST have a value of 255, i.e., the packet could not possibly have been forwarded by a router. o If the message includes an IP Authentication Header, the message authenticates correctly. o ICMP Checksum is valid. o ICMP Code is 0. o ICMP length (derived from the IP length) is 8 or more octets. o Identifier field is non-zero. o All included options have a length that is greater than zero. The contents of the Reserved field, and of any unrecognized options, MUST be ignored. Future, backward-compatible changes to the protocol may specify the contents of the Reserved field or add new options; backward-incompatible changes may use different Code values. The contents of any defined options that are not specified to be used with Router Solicitation messages MUST be ignored and the packet processed in the normal manner. The only defined option that may appear is the Trust Anchor option. A solicitation that passes the validity checks is called a "valid solicitation". Routers MAY send unsolicited Delegation Chain Advertisements for their configured trust anchor(s). When such advertisements are sent, their timing MUST follow the rules given for Router Advertisements in RFC 2461 [6]. The only defined options that may appear are the Certificate and Trust Anchor options. At least one Certificate option MUST be present. Router SHOULD also include at least one Trust Anchor option to indicate the trust anchor on which the Certificate is based. In addition to sending periodic, unsolicited advertisements, a router sends advertisements in response to valid solicitations received on an advertising interface. If the source address in the solicitation was the unspecified address, the router MUST send the response to the link-scoped All-Nodes multicast address. If the source address was a unicast address, the router MUST send the response to the Arkko, et al. Expires April 16, 2004 [Page 33] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Solicited-Node multicast address corresponding to the source address. In a solicited-for advertisement, the router SHOULD include suitable Certificate options so that a delegation chain to the solicited trust anchor can be established. The anchor is identified by the Trust Anchor option. If the Trust Anchor option is represented as a DER Encoded X.501 Name, then the Name must be equal to the Subject field in the anchor's certificate. If the Trust Anchor option is represented as an FQDN, the FQDN must be equal to an FQDN in the subjectAltName field of the anchor's certificate. The router SHOULD include the Trust Anchor option(s) in the advertisement for which the delegation chain was found. If the router is unable to find a chain to the requested anchor, it SHOULD send an advertisement without any certificates. In this case the router SHOULD include the Trust Anchor options which were solicited. Rate limiting of Delegation Chain Advertisements is performed as specified for Router Advertisements in RFC 2461 [6]. 6.7 Processing Rules for Hosts Hosts SHOULD possess the public key and trust anchor name of at least one certificate authority, they SHOULD possess their own key pair, and they MAY posses a certificate from the above mentioned certificate authority. A host MUST silently discard any received Delegation Chain Advertisement messages that do not satisfy all of the following validity checks: o IP Source Address MUST be a unicast address. Note that routers may use multiple addresses, and therefore this address not sufficient for the unique identification of routers. o IP Destination Address MUST be either the link-scoped All-Nodes multicast address or the Solicited-Node multicast address corresponding to one of the unicast addresses assigned to the host. o The IP Hop Limit field MUST have a value of 255, i.e., the packet could not possibly have been forwarded by a router. o If the message includes an IP Authentication Header, the message authenticates correctly. o ICMP Checksum is valid. Arkko, et al. Expires April 16, 2004 [Page 34] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 o ICMP Code is 0. o ICMP length (derived from the IP length) is 16 or more octets. o All included options have a length that is greater than zero. The contents of the Reserved field, and of any unrecognized options, MUST be ignored. Future, backward-compatible changes to the protocol may specify the contents of the Reserved field or add new options; backward-incompatible changes may use different Code values. The contents of any defined options that are not specified to be used with Delegation Chain Advertisement messages MUST be ignored and the packet processed in the normal manner. The only defined options that may appear are the Certificate and Trust Anchor options. An advertisement that passes the validity checks is called a "valid advertisement". Hosts SHOULD store certificate chains retrieved in Delegation Chain Discovery messages if they start from an anchor trusted by the host. The certificates chains SHOULD be verified, as defined in Section 6.5, before storing them. Routers are required to send the certificates one by one, starting from the trust anchor end of the chain. Except for temporary purposes to allow for message loss and reordering, hosts SHOULD NOT store certificates received in a Delegation Chain Advertisement unless they contain a certificate which can be immediately verified either to the trust anchor or to a certificate which has been verified earlier. Note that it may be useful to cache this information and implied verification results for use over multiple attachments to the network. When an interface becomes enabled, a host may be unwilling to wait for the next unsolicited Delegation Chain Advertisement. To obtain such advertisements quickly, a host MAY transmit up to MAX_RTR_SOLICITATIONS Delegation Chain Solicitation messages, each separated by at least RTR_SOLICITATION_INTERVAL seconds. Delegation Chain Solicitations MAY be sent after any of the following events: o The interface is initialized at system startup time. o The interface is reinitialized after a temporary interface failure or after being temporarily disabled by system management. o The system changes from being a router to being a host, by having its IP forwarding capability turned off by system management. o The host attaches to a link for the first time. Arkko, et al. Expires April 16, 2004 [Page 35] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 o A movement has been indicated by lower layers or has been inferred from changed information in a Router Advertisement. o The host re-attaches to a link after being detached for some time. o A Router Advertisement has been received with a public key that is not stored in the hosts' cache of certificates, or there is no authorization delegation chain to the host's trust anchor. Delegation Chain Solicitations SHOULD NOT be sent if the host has a currently valid certificate chain from a reachable router to a trust anchor. When soliciting certificates for a router, a host MUST send Delegation Chain Solicitations either to the All-Routers multicast address, if it has not selected a default router yet, or to the default router's IP address, if it has already been selected. If two hosts want to establish trust with the DCS and DCA messages, the DCS message SHOULD be sent to the Solicited-Node multicast address of the receiver. The advertisements SHOULD be sent as specified above for routers. However, the exact details are left for a future specification. Delegation Chain Solicitations SHOULD be rate limited and timed similarly with Router Solicitations, as specified in RFC 2461 [6]. When processing possible advertisements sent as responses to a solicitation, the host MAY prefer to process first those advertisements with the same Identifier field value as in the solicitation. This makes Denial-of-Service attacks against the mechanism harder (see Section 11.3). Arkko, et al. Expires April 16, 2004 [Page 36] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 7. Securing Neighbor Discovery with SEND This section describes how to use the mechanisms from Section 5, Section 6, and the reference [26] in order to provide security for Neighbor Discovery. There is no requirement that nodes use both Secure Neighbor Discovery (as described in this Section) and Secure Router Discovery (as described in Section 8. They MAY be used indepedently. 7.1 Neighbor Solicitation Messages All Neighbor Solicitation messages are protected with SEND. 7.1.1 Sending Secure Neighbor Solicitations Secure Neighbor Solicitation messages are sent as described in RFC 2461 and 2462, with the additional requirements as listed in the following: All Neighbor Solicitation messages sent MUST contain the Nonce, Timestamp, and Signature options, and MAY contain the CGA option. The Signature option MUST be constructed with the sender's key pair, using the configured authorization method(s), and if applicable, using the trust anchor and/or minSec value as configured. 7.1.2 Receiving Secure Neighbor Solicitations Received Neighbor Solicitation messages are processed as described in RFC 2461 and 2462, with the additional SEND-related requirements as listed in the following: Neighbor Solicitation messages received without the Nonce, Timestamp, or Signature option MUST be silently discarded. The Signature option MUST be constructed with the expected authorization method(s), the used key being within the configured minimum (and maximum) allowable key size, and if applicable, using an acceptable trust anchor and/or minSec value. 7.2 Neighbor Advertisement Messages All Neighbor Advertisement messages are protected with SEND. 7.2.1 Sending Secure Neighbor Advertisements Arkko, et al. Expires April 16, 2004 [Page 37] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Secure Neighbor Advertisement messages are sent as described in RFC 2461 and 2462, with the additional requirements as listed in the following: All Neighbor Advertisement messages sent MUST be sent with the Timestamp and Signature options and MAY be sent with the CGA option. The Signature option MUST be constructed with the sender's key pair, setting the authorization method and additional information as configured. Neighbor Advertisements sent in response to a Neighbor Solicitation MUST additionally contain a copy of the Nonce option included in the solicitation. 7.2.2 Receiving Secure Neighbor Advertisements Received Neighbor Advertisement messages are processed as described in RFC 2461 and 2462, with the additional SEND-related requirements as listed in the following: Any eighbor Advertisement messages received without the Timestamp or Signature options MUST be silently discarded. The Signature option MUST be constructed with the expected authorization method(s), the used key being within the configured minimum (and maximum) allowable key size, and if applicable, using an acceptable trust anchor and/or minSec value. Received Neighbor Advertisements sent to a unicast destination address without a Nonce option MUST be silently discarded. 7.3 Other Requirements Upon receiving a message for which the receiver has no certificate chain to a trust anchor, the receiver MAY use Authorization Delegation Discovery to learn the certificate chain of the peer. Nodes that use stateless address autoconfiguration, SHOULD generate a new CGA as specified in Section 4 of [26] for each new autoconfiguration run. The nodes MAY continue to use the same public key and modifier, and start the process from Step 4. This specification does not address the protection of Neighbor Discovery packets for nodes that are configured with a static address (e.g., PREFIX::1). Future certificate chain based authorization specifications are needed for such nodes. Arkko, et al. Expires April 16, 2004 [Page 38] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 It is outside the scope of this specification to describe the use of trust anchor authorization between nodes with dynamically changing addresses. Such dynamically changing addresses may be the result of stateful or stateless address autoconfiguration, or through the use of RFC 3041 [9] addresses. If the CGA method is not used, nodes would be required to exchange certificate chains that terminate in a certificate authorizing a node to use an IP address having a particular interface identifier. This specification does not specify the format of such certificates, since there are currently a few cases where such certificates are required by the link layer and it is up to the link layer to provide certification for the interface identifier. This may be the subject of a future specification. It is also outside the scope of this specification to describe how stateful address autoconfiguration works with the CGA method. Arkko, et al. Expires April 16, 2004 [Page 39] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 8. Securing Router Discovery with SEND This section describes how to use the mechanisms from Section 5, Section 6, and the reference [26] in order to provide security for Router Discovery. 8.1 Router Solicitation Messages All Router Solicitation messages are protected with SEND. 8.1.1 Sending Secure Router Solicitations Secure Router Solicitation messages are sent as described in RFC 2461, with the additional requirements as listed in the following: Router Solicitation messages sent with an unspecified source address MUST have the Nonce and Timestamp options. Other Router Solicitations MUST have the Nonce, Timestamp, and Signature options. The Signature option MUST be configured with the sender's key pair, setting the authorization method and additional information as is configured. 8.1.2 Receiving Secure Router Solicitations Received Router Solicitation messages are processed as described in RFC 2461, with the additional SEND-related requirements as listed in the following: Router Solicitation message sent with an unspecified source address and without the Nonce or Timestamp options MUST be silently discarded. Router Solicitation messages received with another type of source address but without the Nonce, Timestamp, or Signature options MUST be silently discarded. The Signature option MUST be constructed with the configured authorization method(s), the used key being within the configured minimum (and maximum) allowable key size, and if applicable, using an acceptable trust anchor and/or minSec value. The configured authorization methods MUST include the trust anchor authorization method, and MAY be additionally configured to require CGA authorization. Arkko, et al. Expires April 16, 2004 [Page 40] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 8.2 Router Advertisement Messages All Router Advertisement messages are protected with SEND. 8.2.1 Sending Secure Router Advertisements Secure Router Advertisement messages are sent as described in RFC 2461, with the additional requirements as listed in the following: All Router Advertisement messages sent MUST contain a Timestamp and Signature options. The Signature option MUST be configured to protect the advertisement with the trust anchor authorization method and MAY be configured to additionally protect it with the CGA authorization method. Router Advertisements sent in response to a Router Solicitation MUST contain a copy of the Nonce option included in the solicitation. 8.2.2 Receiving Secure Router Advertisements Received Router Advertisement messages are processed as described in RFC 2461, with the additional SEND-related requirements as listed in the following: Router Advertisement messages received without the Timestamp and Signature options MUST be silently discarded. Received Router Advertisements sent to a unicast destination address without a Nonce option MUST be silently discarded. The Signature option MUST be constructed with the configured authorization method(s), the used key being within the configured minimum (and maximum) allowable key size, and if applicable, using an acceptable trust anchor and/or minSec value. The configured authorization methods MUST include the trust anchor authorization method, and MAY be additionally configured to require CGA authorization. 8.3 Redirect Messages All Redirect messages are protected with SEND. 8.3.1 Sending Redirects Arkko, et al. Expires April 16, 2004 [Page 41] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Secure Redirect messages are sent as described in RFC 2461, with the additional requirements as listed in the following: All Redirect messages sent MUST contain the Timestamp and Signature options. The Signature option MUST be configured to use the trust anchor authorization method, and MAY be additionally configured to use the CGA method. 8.3.2 Receiving Redirects Received Redirect messages are processed as described in RFC 2461, with the additional SEND-related requirements as listed in the following: Redirect messages received without the Timestamp or Signature options MUST be silently discarded. The Signature option MUST be constructed with the configured authorization method(s), the used key being within the configured minimum (and maximum) allowable key size, and if applicable, using an acceptable trust anchor and/or minSec value. The configured authorization methods MUST include the trust anchor authorization method, and MAY be additionally configured to require CGA authorization. The receiver MUST verify that the Redirect message comes from an IP address to which the host may have earlier sent the packet that the Redirect message now partially returns. That is, the source address of the Redirect message must be the default router or the on-link destination host for traffic sent to the destination of the returned packet. If this is not the case, the message MUST be silently discarded. This step prevents a bogus router from sending a Redirect message when the host is not using the bogus router as a default router. 8.4 Other Requirements Hosts SHOULD use Authorization Delegation Discovery to learn the certificate chain of their default router (or peer host), as explained in Section 6. The receipt of a protected Router Advertisement message for which no router Authorization Certificate and certificate chain is available triggers Authorization Delegation Discovery. Arkko, et al. Expires April 16, 2004 [Page 42] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 9. Co-Existence of SEND and non-SEND nodes During the transition to secure links or as a policy consideration, network operators may want to run a particular link with a mixture of secure and insecure nodes. Nodes that support SEND SHOULD support the use of SEND and the legacy Neighbor Discovery Protocol at the same time. In a mixed environment, SEND nodes receive both secure and insecure messages but give priority to "secured" ones. Here, the "secured" messages are ones that contain a valid signature option, as specified above, and "insecure" messages are ones that contain no signature option. SEND nodes send only secured messages. Legacy Neighbor Discovery nodes will obviously send only insecure messages. Such nodes will (as per RFC 2461 [6]) ignore the unknown options and will treat secured messages in the same way as they treat insecure ones. Secured and insecure nodes share the same network resources, such as prefixes and address spaces. In a mixed environment SEND routers and hosts follow the protocols defined in RFC 2461 and RFC 2462 with the following exceptions: All solicitations sent by SEND nodes MUST be secured. Unsolicited Neighbor and Router Advertisements sent by a SEND router MUST be secured. Secured solicitations MUST contain the Nonce option. Secured advertisements sent in response to a secured solicitation MUST contain a copy of the Nonce option from the solicitation. Unsolicited advertisements and ones sent in response to an insecure solicitation MUST NOT contain the Nonce option. A SEND node that uses the CGA authorization method for protecting Neighbor Solicitations SHOULD perform Duplicate Address Detection as follows. If Duplicate Address Detection indicates the tentative address is already in use, generate a new tentative CGA address. If after 3 consecutive attempts no non-unique address was generated, log a system error and give up attempting to generate an address for that interface. When performing Duplicate Address Detection for the first tentative address, accept both secured and insecure Neighbor Advertisements and Solicitations received as response to the Neighbor Solicitations. When performing Duplicate Address Detection for the second or third tentative address, ignore Arkko, et al. Expires April 16, 2004 [Page 43] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 insecure Neighbor Advertisements and Solicitations. The node SHOULD have a configuration option that causes it to ignore insecure advertisements even when performing Duplicate Address Detection for the first tentative address. This configuration option SHOULD be disabled by default. (This is recovery mechanism for the unlikely case that attacks against the first address become common.) The Neighbor Cache, Prefix List and Default Router list entries MUST have a secured/insecure flag that indicates whether the message that caused the creation or last update of the entry was secured or insecure. Received insecure messages MUST NOT cause changes to existing secured entries in the Neighbor Cache, Prefix List or Default Router List. Received secured messages cause an update of the matching entries and flagging of them as secured. The conceptual sending algorithm is modified so that an insecure router is selected only if there is no reachable SEND router for the prefix. That is, the algorithm for selecting a default router favors reachable SEND routers over reachable non-SEND ones. A SEND node SHOULD have a configuration option that causes it to ignore all insecure ND, RD and Redirect messages. (This can be used to enforce SEND-only networks.) Arkko, et al. Expires April 16, 2004 [Page 44] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 10. Performance Considerations The computations related to the Signature option are computationally relatively expensive. In the application which Signature option has been designed for, however, the nodes typically have the need to perform only a few signature operations as they enter a link, and a few operations as they find a new on-link peer with which to communicate. Routers are required to perform a larger number of operations, particularly when the frequency of router advertisements is high due to mobility requirements. Still, the number of required signature operations is on the order of a few dozen ones per second, some of which can be precomputed as discussed below. A large number of router solicitations may cause higher demand for performing asymmetric operations, although RFC 2461 limits the rate at which responses to solicitations can be sent. Signatures related to the use of the Signature option be precomputed for Multicast Neighbor and Router Advertisements. Typically, solicited advertisements are sent to the unicast address from which the solicitation was sent. Given that the IPv6 header is covered by the signature, it is typically not possible to precompute solicited-for advertisements. Arkko, et al. Expires April 16, 2004 [Page 45] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 11. Security Considerations 11.1 Threats to the Local Link Not Covered by SEND SEND does not compensate for an insecure link layer. In particular, there is no cryptographic binding in SEND between the link layer frame address and the IPv6 address. On an insecure link layer that allows nodes to spoof the link layer address of other nodes, an attacker could disrupt IP service by sending out a Neighbor Advertisement having the source address on the link layer frame of a victim, a valid CGA address and a valid signature corresponding to itself, and a Target Link-layer Address extension corresponding to the victim. The attacker could then proceed to cause a traffic stream to bombard the victim in a DoS attack. To protect against such attacks, link layer security MUST be used. An example of such for 802 type networks is port-based access control defined in the 802.1X standard [34]. Specifically, the 802.1X standard provides a mechanism by which a nodes can be authenticated to a particular point of attachment to a LAN (called a "port" in the standard). If the MAC on frames sent by a node does not correspond to the MAC of the node originally authenticated to this port, then the point of attachment drops the frames. Authorization to use the port is determined by the MAC address of the node that originally authenticated to the port. The way 802.1X protects against this attack is that, if a node authenticated to a particular port attempts to spoof the MAC address of another node, the port will drop the frames. Naturally, this requires that all ports by which nodes can attach to the LAN use 802.1X authentication, and that all node physically attach through a port, as is the case with 802.3 switched LAN. For shared media, such as multiple nodes authenticated through the same 802.11 AP (which acts as a single port for all nodes), other measures are necessary, since an attacker on the wireless link can spoof the MAC address of a victim on the same wireless link. 802.1X does not provide protection for the layer 2 frame - layer 3 packet address binding in traffic (that is, real time filtering to check this binding), and neither does SEND. 802.1X provides authentication and filtering of MAC address to port; SEND provides protection for the layer 2 - layer 3 binding information in the Neighbor Discovery packet, via the CGA address (authorization to use the address via the public key) and the signature on the packet (authentication of contents as from authorized IP address possessor). Prior to participating in Neighbor Discovery and Duplicate Address Detection, nodes must subscribe to the link-scoped All-Nodes Multicast Group and the Solicited-Node Multicast Group for the Arkko, et al. Expires April 16, 2004 [Page 46] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 address that they are claiming for their addresses; RFC 2461 [6]. Subscribing to a multicast group requires that the nodes use MLD [20]. MLD contains no provision for security. An attacker could send an MLD Done message to unsubscribe a victim from the Solicited-Node Multicast address. However, the victim should be able to detect such an attack because the router sends a Multicast-Address-Specific Query to determine whether any listeners are still on the address, at which point the victim can respond to avoid being dropped from the group. This technique will work if the router on the link has not been compromised. Other attacks using MLD are possible, but they primarily lead to extraneous (but not overwhelming) traffic. 11.2 How SEND Counters Threats to Neighbor Discovery The SEND protocol is designed to counter the threats to IPv6 Neighbor Discovery, as outlined in [27]. The following subsections contain a regression of the SEND protocol against the threats, to illustrate what aspects of the protocol counter each threat. 11.2.1 Neighbor Solicitation/Advertisement Spoofing This threat is defined in Section 4.1.1 of [27]. The threat is that a spoofed Neighbor Solicitation or Neighbor Advertisement causes a false entry in a node's Neighbor Cache. There are two cases: 1. Entries made as a side effect of a Neighbor Solicitation or Router Solicitation. There are two cases: 1. A router receiving a Router Solicitation with a firm IPv6 source address and a Target Link-Layer Address extension inserts an entry for the IPv6 address into its Neighbor Cache. 2. A node doing Duplicate Address Detection (DAD) that receives a Neighbor Solicitation for the same address regards the situation as a collision and ceases to solicit for the address. 2. Entries made as a result of a Neighbor Advertisement sent as a response to a Neighbor Solicitation for purposes of on-link address resolution. 11.2.1.1 Solicitations with Effect SEND counters the threat of solicitations with effect in the following ways: Arkko, et al. Expires April 16, 2004 [Page 47] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 1. As discussed in Section 5, SEND nodes preferably send Router Solicitations with a CGA address and a Signature option, which the router can verify, so the Neighbor Cache binding is correct. If a SEND node must send a Router Solicitation with the unspecified address, the router will not update its Neighbor Cache, as per RFC 2461. See Section 11.2.5, below, for discussion about replay protection and timestamps. 11.2.1.2 Address Resolution SEND counters attacks on address resolution by requiring that the responding node include a signature option on the packet, and that the node's interface identifier either be a CGA, or that the node be able to produce a certificate authorizing that node to use the public key. The Neighbor Solicitation and Advertisement pairs implement a challenge-response protocol, as explained in Section 7 and discussed in Section 11.2.5 below. 11.2.2 Neighbor Unreachability Detection Failure This attack is described in Section 4.1.2 of [27]. SEND counters this attack by requiring a node responding to Neighbor Solicitations sent as NUD probes to include a Signature option and proof of authorization to use the interface identifier in the address being probed. If these prerequisites are not met, the node performing NUD discards the responses. 11.2.3 Duplicate Address Detection DoS Attack This attack is described in Section 4.1.3 of [27]. SEND counters this attack by requiring the Neighbor Advertisements sent as responses to DAD to include a Signature option and proof of authorization to use the interface identifier in the address being tested. If these prerequisites are not met, the node performing DAD discards the responses. When a SEND node is used on a link that also connects to non-SEND nodes, the SEND node ignores any insecure Neighbor Solicitations or Advertisements that may be send by the non-SEND nodes. This protects the SEND node from DAD DoS attacks by non-SEND nodes or attackers simulating to non-SEND nodes, at the cost of a potential address collision between a SEND node and non-SEND node. The probability and effects of such an address collision are discussed in [26]. Arkko, et al. Expires April 16, 2004 [Page 48] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 11.2.4 Router Solicitation and Advertisement Attacks These attacks are described in Sections 4.2.1, 4.2.4, 4.2.5, 4.2.6, and 4.2.7 of [27]. SEND counters these attacks by requiring Router Advertisements to contain a Signature option, and that the signature is calculated using the public key of a node that can prove its authorization to route the subnet prefixes contained in any Prefix Information Options. The router proves its authorization by showing a certificate containing the specific prefix or the indication that the router is allowed to route any prefix. A Router Advertisement without these protections is dropped. SEND does not protect against brute force attacks on the router, such as DoS attacks, or compromise of the router, as described in Sections 4.4.2 and 4.4.3 of [27]. 11.2.5 Replay Attacks This attack is described in Section 4.3.1 of [27]. SEND protects against attacks in Router Solicitation/Router Advertisement and Neighbor Solicitation/Neighbor Advertisement transactions by including a Nonce option in the solicitation and requiring the advertisement to include a matching option. Together with the signatures this forms a challenge-response protocol. SEND protects against attacks from unsolicited messages such as Neighbor Advertisements, Router Advertisements, and Redirects by including a Timestamp option. A window of vulnerability for replay attacks exists until the timestamp expires. When timestamps are used, SEND nodes are protected against replay attacks as long as they cache the state created by the message containing the timestamp. The cached state allows the node to protect itself against replayed messages. However, once the node flushes the state for whatever reason, an attacker can re-create the state by replaying an old message while the timestamp is still valid. Since most SEND nodes are likely to use fairly coarse grained timestamps, as explained in Section 5.4.1, this may affect some nodes. 11.2.6 Neighbor Discovery DoS Attack This attack is described in Section 4.3.2 of [27]. In this attack, the attacker bombards the router with packets for fictitious addresses on the link, causing the router to busy itself with performing Neighbor Solicitations for addresses that do not exist. SEND does not address this threat because it can be addressed by techniques such as rate limiting Neighbor Solicitations, restricting the amount of state reserved for unresolved solicitations, and clever Arkko, et al. Expires April 16, 2004 [Page 49] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 cache management. These are all techniques involved in implementing Neighbor Discovery on the router. 11.3 Attacks against SEND Itself The CGAs have a 59-bit hash value. The security of the CGA mechanism has been discussed in [26]. Some Denial-of-Service attacks against NDP and SEND itself remain. For instance, an attacker may try to produce a very high number of packets that a victim host or router has to verify using asymmetric methods. While safeguards are required to prevent an excessive use of resources, this can still render SEND non-operational. When CGA protection is used, SEND deals with the DoS attacks using the verification process described in Section 5.3.2. In this process, a simple hash verification of the CGA property of the address is performed first before performing the more expensive signature verification. When trust anchors and certificates are used for address validation in SEND, the defenses are not quite as effective. Implementations SHOULD track the resources devoted to the processing of packets received with the Signature option, and start selectively dropping packets if too many resources are spent. Implementations MAY also first drop packets that are not protected with CGA. The Authorization Delegation Discovery process may also be vulnerable to Denial-of-Service attacks. An attack may target a router by requesting a large number of delegation chains to be discovered for different trust anchors. Routers SHOULD defend against such attacks by caching discovered information (including negative responses) and by limiting the number of different discovery processes they engage in. Attackers may also target hosts by sending a large number of unnecessary certificate chains, forcing hosts to spend useless memory and verification resources for them. Hosts can defend against such attacks by limiting the amount of resources devoted to the certificate chains and their verification. Hosts SHOULD also prioritize advertisements that sent as a response to their solicitations above unsolicited advertisements. Arkko, et al. Expires April 16, 2004 [Page 50] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 12. IANA Considerations This document defines two new ICMP message types, used in Authorization Delegation Discovery. These messages must be assigned ICMPv6 type numbers from the informational message range: o The Delegation Chain Solicitation message, described in Section 6.1. o The Delegation Chain Advertisement message, described in Section 6.2. This document defines six new Neighbor Discovery Protocol [6] options, which must be assigned Option Type values within the option numbering space for Neighbor Discovery Protocol messages: o The Trust Anchor option, described in Section 6.3. o The Certificate option, described in Section 6.4. o The CGA option, described in Section 5.2. o The Signature option, described in Section 5.3. o The Timestamp option, described in Section 5.4.1. o The Nonce option, described in Section 5.4.2. This document defines a new 128-bit CGA Message Type [26] value, 0xXXXX XXXX XXXX XXXX XXXX XXXX XXXX XXXX (To be generated randomly). XXX: Use existing name spaces for these? This document defines a new name space for the Name Type field in the Trust Anchor option. Future values of this field can be allocated using standards action [5]. Another new name space is allocated for the Cert Type field in the Certificate option. Future values of this field can be allocated using standards action [5]. Arkko, et al. Expires April 16, 2004 [Page 51] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Normative References [1] Mockapetris, P., "Domain names - concepts and facilities", STD 13, RFC 1034, November 1987. [2] Kent, S. and R. Atkinson, "Security Architecture for the Internet Protocol", RFC 2401, November 1998. [3] Kent, S. and R. Atkinson, "IP Authentication Header", RFC 2402, November 1998. [4] Piper, D., "The Internet IP Security Domain of Interpretation for ISAKMP", RFC 2407, November 1998. [5] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA Considerations Section in RFCs", BCP 26, RFC 2434, October 1998. [6] Narten, T., Nordmark, E. and W. Simpson, "Neighbor Discovery for IP Version 6 (IPv6)", RFC 2461, December 1998. [7] Thomson, S. and T. Narten, "IPv6 Stateless Address Autoconfiguration", RFC 2462, December 1998. [8] Conta, A. and S. Deering, "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification", RFC 2463, December 1998. [9] Narten, T. and R. Draves, "Privacy Extensions for Stateless Address Autoconfiguration in IPv6", RFC 3041, January 2001. [10] Bassham, L., Polk, W. and R. Housley, "Algorithms and Identifiers for the Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile", RFC 3279, April 2002. [11] Housley, R., Polk, W., Ford, W. and D. Solo, "Internet X.509 Public Key Infrastructure Certificate and Certificate Revocation List (CRL) Profile", RFC 3280, April 2002. [12] Hinden, R. and S. Deering, "Internet Protocol Version 6 (IPv6) Addressing Architecture", RFC 3513, April 2003. [13] Lynn, C., "X.509 Extensions for IP Addresses and AS Identifiers", draft-ietf-pkix-x509-ipaddr-as-extn-02 (work in progress), September 2003. [14] Johnson, D., Perkins, C. and J. Arkko, "Mobility Support in Arkko, et al. Expires April 16, 2004 [Page 52] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 IPv6", draft-ietf-mobileip-ipv6-24 (work in progress), July 2003. [15] RSA Laboratories, "RSA Encryption Standard, Version 2.1", PKCS 1, November 2002. [16] National Institute of Standards and Technology, "Secure Hash Standard", FIPS PUB 180-1, April 1995, . Arkko, et al. Expires April 16, 2004 [Page 53] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Informative References [17] Postel, J., "Internet Control Message Protocol", STD 5, RFC 792, September 1981. [18] Plummer, D., "Ethernet Address Resolution Protocol: Or converting network protocol addresses to 48.bit Ethernet address for transmission on Ethernet hardware", STD 37, RFC 826, November 1982. [19] Harkins, D. and D. Carrel, "The Internet Key Exchange (IKE)", RFC 2409, November 1998. [20] Deering, S., Fenner, W. and B. Haberman, "Multicast Listener Discovery (MLD) for IPv6", RFC 2710, October 1999. [21] Farrell, S. and R. Housley, "An Internet Attribute Certificate Profile for Authorization", RFC 3281, April 2002. [22] Arkko, J., "Effects of ICMPv6 on IKE", draft-arkko-icmpv6-ike-effects-02 (work in progress), March 2003. [23] Arkko, J., "Manual Configuration of Security Associations for IPv6 Neighbor Discovery", draft-arkko-manual-icmpv6-sas-02 (work in progress), March 2003. [24] Droms, R., "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", draft-ietf-dhc-dhcpv6-28 (work in progress), November 2002. [25] Kent, S., "IP Encapsulating Security Payload (ESP)", draft-ietf-ipsec-esp-v3-06 (work in progress), July 2003. [26] Aura, T., "Cryptographically Generated Addresses (CGA)", draft-ietf-send-cga-01 (work in progress), August 2003. [27] Nikander, P., "IPv6 Neighbor Discovery trust models and threats", draft-ietf-send-psreq-03 (work in progress), April 2003. [28] Montenegro, G. and C. Castelluccia, "SUCV Identifiers and Addresses", draft-montenegro-sucv-03 (work in progress), July 2002. [29] International Organization for Standardization, "The Directory - Authentication Framework", ISO Standard X.509, 2000. Arkko, et al. Expires April 16, 2004 [Page 54] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 [30] O'Shea, G. and M. Roe, "Child-proof Authentication for MIPv6", Computer Communications Review, April 2001. [31] Nikander, P., "Denial-of-Service, Address Ownership, and Early Authentication in the IPv6 World", Proceedings of the Cambridge Security Protocols Workshop, April 2001. [32] Arkko, J., Aura, T., Kempf, J., Mantyla, V., Nikander, P. and M. Roe, "Securing IPv6 Neighbor Discovery", Wireless Security Workshop, September 2002. [33] Montenegro, G. and C. Castelluccia, "Statistically Unique and Cryptographically Verifiable (SUCV) Identifiers and Addresses", NDSS, February 2002. [34] Institute of Electrical and Electronics Engineers, "Local and Metropolitan Area Networks: Port-Based Network Access Control", IEEE Standard 802.1X, September 2001. Authors' Addresses Jari Arkko Ericsson Jorvas 02420 Finland EMail: jari.arkko@ericsson.com James Kempf DoCoMo Communications Labs USA 181 Metro Drive San Jose, CA 94043 USA EMail: kempf@docomolabs-usa.com Bill Sommerfeld Sun Microsystems 1 Network Drive UBUR02-212 Burlington, MA 01803 USA EMail: sommerfeld@east.sun.com Arkko, et al. Expires April 16, 2004 [Page 55] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Brian Zill Microsoft USA EMail: bzill@microsoft.com Pekka Nikander Ericsson Jorvas 02420 Finland EMail: Pekka.Nikander@nomadiclab.com Arkko, et al. Expires April 16, 2004 [Page 56] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Appendix A. Contributors Tuomas Aura contributed the transition mechanism specification in Section 9. Arkko, et al. Expires April 16, 2004 [Page 57] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Appendix B. IPR Considerations The optional CGA part of SEND uses public keys and hashes to prove address ownership. Several IPR claims have been made about such methods. Arkko, et al. Expires April 16, 2004 [Page 58] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Appendix C. Cache Management In this section we outline a cache management algorithm that allows a node to remain partially functional even under a cache filling DoS attack. This appendix is informational, and real implementations SHOULD use different algorithms in order to avoid he dangers of monocultural code. There are at least two distinct cache related attack scenarios: 1. There are a number of nodes on a link, and someone launches a cache filling attack. The goal here is clearly make sure that the nodes can continue to communicate even if the attack is going on. 2. There is already a cache filling attack going on, and a new node arrives to the link. The goal here is to make it possible for the new node to become attached to the network, inspite of the attack. From this point of view, it is clearly better to be very selective in how to throw out entries. Reducing the timestamp Delta value is very discriminative against those nodess that have a large clock difference, while an attacker can reduce its clock difference into arbitrarily small. Throwing out old entries just because their clock difference is large seems like a bad approach. A reasonable idea seems to be to have a separate cache space for new entries and old entries, and under an attack more eagerly drop new cache entries than old ones. One could track traffic, and only allow those new entries that receive genuine traffic to be converted into old cache entries. While such a scheme will make attacks harder, it will not fully prevent them. For example, an attacker could send a little traffic (i.e. a ping or TCP syn) after each NS to trick the victim into promoting its cache entry to the old cache. Hence, the node may be more intelligent in keeping its cache entries, and not just have a black/white old/new boundary. It also looks like a good idea to consider the sec parameter when forcing cache entries out, and let those entries with a larger sec a higher chance of staying in. Arkko, et al. Expires April 16, 2004 [Page 59] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Appendix D. Comparison to AH-Based Approach This approach has the following benefits compared to the previous Working Group document approach: o The full implementation of the security mechanism, including Nonces and CGAs, exists within one module. There is no need to analyze the security of the mechanism across NDP, IPsec, and CGA layers. o The CGA part of the solution has been separated into its own specification. This is possible because the CGA handling is done in its own option. (The authorization method configuration flag is the only thing common to the CGA and Signature options.) o No extensions or modifications of IPsec processing are required: SPD entries are not required to distinguish ICMP types, AH does not need to support public keys or CGAs, and destination address acgnostic security associations are not needed. o It is not necessary to allocate a new multicast address to represent the Solicited-Node multicast address for SEND nodes. o It is not necessary to change the Neighbor Discovery behavior with regards to the use of the unspecified address. Since all information is available within the Neighbor Discovery messages, unspecified source addresses can be used, still being able to correlate the CGA property with the Target Address in a Neighbor Solicitation during Duplicate Address Detection. o The transition mechanisms for links with both SEND and non-SEND nodes are significantly simpler. In particular, non-SEND nodes will be able to receive DAD probes and other messages sent by the SEND nodes. o Only a single set of Neighbor Discovery messages from the router needs to be transmitted on a link. This helps avoid extra overhead for mobility beacons and other frequently occurring messaging. o Given that the asymmetric computations required in SEND are computationally expensive, it is necessary to control the number of these operations in order to avoid Denial-of-Service attacks. This control is easier to arrange with "application layer" information. For instance, a router need not verify more Router Solicitations with an unspecified source address than it can respond to according to the RFC 2461 rules. Arkko, et al. Expires April 16, 2004 [Page 60] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 o There is no need for an API to communicate certificate chains requests and certificate chains between the IPsec and Neighbor Discovery modules. Also, a good implementation of SEND would not require the user to configure it (beyond perhaps enabling it). In order to achieve this with IPsec, a set of policy entries needs to be automatically created upon system start. o There is no need for the CGA parameters to be stored both in the IPsec and Neighbor Discovery modules, where they are needed for the construction of Authentication Headers and addresses, respectively. o It is not necessary to change existing BITS or BITW IPsec implementations to support SEND and AH_RSA_Sig. There would have been two problems associated with such changes: * A SEND implementation in such environment could not proceed until this modification were completed. * Typical hardware that processes IPsec packets may not be easily changed to process asymmetric transforms. (Of course, such packets can be passed to the main CPU at the node, assuming this can easily be done in the given implementation.) o In addition, many IPsec implementations are highly optimized because they are on the fast path for packet processing. For example, the Linux implementation runs in the kernel interrupt thread. Some of the SEND modifications might have required IPsec processing to wait on a semaphore while, for example, a certificate chain is fetched, an operation that takes place out of band in regular IPsec processing because it is done using IKE. While it might have been possible that the implemenation could have been arranged so that general IPsec processing wasqn't impacted, the resulting code would have been more complex. The use of IPsec to protect NDP would have been possible, but the limits and capabilities of IPsec would have to be stretched. Small changes in the NDP protocol (or our understanding of the issues) might have caused a situation which had no longer been easily handled when the "application" and the security existed at different layers. Although IPsec as defined in RFC 2402 just defines a header format, RFC 2401 and the ensuing years of implementation have evolved a complex interconnected set of components for IPsec which would have required some modification to accommodate SEND. On the other hand, IPsec is the current solution for securing NDP in Arkko, et al. Expires April 16, 2004 [Page 61] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 the original NDP RFCs. Even if the current IPsec can be used only in very limited networks to secure NDP, it could have been argued that it would have been logical to continue its use. Also, the existence of an asymmetric transform in IPsec would have been potentially useful in other contexts as well. Arkko, et al. Expires April 16, 2004 [Page 62] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 Intellectual Property Statement The IETF takes no position regarding the validity or scope of any intellectual property or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; neither does it represent that it has made any effort to identify any such rights. Information on the IETF's procedures with respect to rights in standards-track and standards-related documentation can be found in BCP-11. Copies of claims of rights made available for publication and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementors or users of this specification can be obtained from the IETF Secretariat. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights which may cover technology that may be required to practice this standard. Please address the information to the IETF Executive Director. Full Copyright Statement Copyright (C) The Internet Society (2003). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assignees. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION Arkko, et al. Expires April 16, 2004 [Page 63] Internet-Draft SEcure Neighbor Discovery (SEND) October 2003 HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Acknowledgement Funding for the RFC Editor function is currently provided by the Internet Society. Arkko, et al. Expires April 16, 2004 [Page 64]