Network Working Group F. Templin Internet-Draft Nokia Expires:
April 14,July 20, 2004 T. Gleeson Cisco Systems K.K. M. Talwar D. Thaler Microsoft Corporation October 15, 2003January 20, 2004 Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) draft-ietf-ngtrans-isatap-16.txtdraft-ietf-ngtrans-isatap-17.txt 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 14,July 20, 2004. Copyright Notice Copyright (C) The Internet Society (2003).(2004). All Rights Reserved. Abstract This document specifies anThe Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) thatconnects IPv6 hosts and routers withinneighbors/routers over IPv4 sites.networks. ISATAP treatsviews the site'sIPv4 unicast infrastructurenetwork as a Non-Broadcast, Multiple Access (NBMA) link layer for IPv6 with no requirement for IPv4 multicast.and views other nodes on the network as potential IPv6 neighbors/routers. ISATAP enablesinterfaces support automatic IPv6-in-IPv4tunneling whether globally assigned or private IPv4 addresses are used. ISATAP supports an abstraction for tunnel interface management similar to the ATM Permanent/Switched Virtual Circuit (PVC/SVC) model. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Requirements . . . . . . . . . . . . . . . . . . . . . . . . . 3 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 4. Basic IPv6Model of Operation . . . . . . . . . . . . . . . . . . . . . . 4 5. Automatic TunnelingNode Requirements . . . . . . . . . . . . . . . . . . . . . 6. 4 6. Neighbor DiscoveryAddressing Requirements . . . . . . . . . . . . . . . . . . . 4 7. Configuration and Management Requirements . . . . . . . . . . 6 8. Automatic Tunneling . . . 8 7. Address Autoconfiguration. . . . . . . . . . . . . . . . . . 12 8.9. Neighbor Discovery . . . . . . . . . . . . . . . . . . . . . . 17 10. Other Requirements for Control Plane Signalling . . . . . . . 19 11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12 9.20 12. Security considerations . . . . . . . . . . . . . . . . . . . 12 10.20 13. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 1220 Normative References . . . . . . . . . . . . . . . . . . . . . 1321 Informative References . . . . . . . . . . . . . . . . . . . . 1422 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 1524 A. Major Changes . . . . . . . . . . . . . . . . . . . . . . . . 1525 B. Interface Identifier Construction . . . . . . . . . . . . . . 1626 Intellectual Property and Copyright Statements . . . . . . . . 1828 1. Introduction This document specifies a simple mechanism called the Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) that enables incremental deployment ofconnects IPv6 [RFC2460] withinneighbors/routers over IPv4 [RFC0791] sites.networks. ISATAP allows dual-stack (IPv6/IPv4) nodes that do not share a link with an IPv6 routerto automatically tunnel packets to the IPv6 next-hop address through IPv4, i.e., ISATAP sees the site'sIPv4 infrastructure is treatednetwork as a link layer for IPv6.IPv6 and views other nodes on the network as potential IPv6 neighbors/routers. ISATAP supports an abstraction for tunnel interface management similar to the Non-Broadcast, Multiple Access (NBMA) [RFC2491] and ATM Permanent/Switched Virtual Circuit (PVC/SVC) [RFC2492] paradigms. The main objectives of this document are to: 1) specifydescribe the operational detailsmodel for ISATAP, 2) specify addressing requirements, 3) discuss configuration and management requirements, 4) specify automatic tunneling of IPv6 over IPv4using ISATAP, 2)5) specify the formatoperational aspects of IPv6 interface identifiers using an embedded IPv4 address, 3) specify the operation ofNeighbor Discovery and Address Autoconfiguration,Discovery, and 4)6) discuss securityIANA; Security considerations. The specification in this document is very similar to [RFC2529], with the primary distinction that ISATAP does not require IPv4 multicast support within the site.2. Requirements The keywords MUST, MUST NOT, REQUIRED, SHALL, SHALL NOT, SHOULD, SHOULD NOT, RECOMMENDED, MAY, and OPTIONAL, when they appear in this document, are to be interpreted as described in [RFC2119]. This document also makes use3. Terminology The terminology of internal conceptual variables to describe protocol behavior and external variables that an implementation must allow system administrators to change. The specific variable names, how their values change, and how their settings influence protocol behavior are provided to demonstrate protocol behavior. An implementation is not required to have them in the exact form described here, so long as its external behavior is consistent with that described in this document. 3. Terminology The terminology of [RFC2460][RFC2461][RFC2462] applies[RFC1122][RFC2460][RFC2461][RFC3582] applies to this document. The following additional terms are defined: site: same as defined in [RFC3582], which is intended to be equivalent to "enterprise" as defined in [RFC1918].ISATAP interface:node: a dual-stack (IPv6/IPv4) node that implements this specification. ISATAP driver: an interface usedISATAP node's network driver module that provides an engine for automatic IPv6-in-IPv4 tunnelingencapsulation, decapsulation and configured over one or more IPv4 addresses assigned to one or moreforwarding of packets between tunnel interfaces and the node'sIPv4 interfacesstack; it also implements an API for tunnel interface management. ISATAP server daemon: an ISATAP node's process that belong tosends/receives tunnel configuration control plane messages, and configures/manages tunnel interfaces via the ISATAP driver API; often will be the same site. advertisingserver daemon used for IPv6 neighbor/router discovery. ISATAP interface: same meaning as advertisingan ISATAP node's point-to-multipoint IPv6 interface used for IPv6-in-IPv4 tunneling of control plane traffic; may also be used to carry data plane traffic in some deployments scenarios, e.g., certain enterprise networks. ISATAP interface identifier: an IPv6 interface identifier with an embedded IPv4 address constructed as specified in ([RFC2461], section 6.2.2).Section 6.1. ISATAP address: an IPv6 unicast address with an on-link prefixassigned onto an ISATAP interface andwith an on-link prefix and an ISATAP interface identifier constructed as specified in Section 4.1.identifier. 4. Basic IPv6Model of Operation ISATAP interfaces automatically tunnel IPv6 packets in IPv4 using the site's IPv4 infrastructure asprovide a link layer, i.e., IPv6 treats the site's IPv4 infrastructurepoint-to-multipoint abstraction for IPv6-in-IPv4 tunneling. They are commonly used as a Non-Broadcast, Multiple Access (NBMA) link layer with properties similar to [RFC2491]. The following sections specify detailsnexus for basicautomatic configuration of point-to-point tunnels via tunnel configuration control plane messages (e.g., IPv6 operation on ISATAP interfaces: 4.1 Interface IdentifiersNeighbor Discovery and Unicast Addresses Interface identifiers forother ICMPv6 messages). For each tunneled packet received, the node's ISATAP are constructed in Modified EUI-64 format as specified in ([ADDR-ARCH], section 2.5.1). They are formed by appendingdriver examines a local forwarding table to determine the correct interface to receive the packet after decapsulation. It forwards tunnel configuration control plane messages via an ISATAP interface to the node's ISATAP server daemon, and forwards data messages to applications via configured tunnel interfaces based on a specific match for the encapsulating IPv4 source address. The ISATAP server daemon sends and receives control plane messages, and configures/manages tunnels via the ISATAP driver API. Each such configured tunnel provides a nexus for multiplexing one or more applications between the remote and local tunnel endpoints using IPv6 addresses as application identifiers. Each such application identifier provides a nexus for multiplexing one or more sessions. In summary, each configured tunnel provides a single point-to-point connection between peers that can be used to carry multiple applications and multiple instances of each application. 5. Node Requirements Nodes that use this specification implement the common functionality required by [NODEREQ] as well as the additional features specified in this document. 6. Addressing Requirements 6.1 ISATAP Interface Identifiers ISATAP interface identifiers are constructed in Modified EUI-64 format as specified in ([ADDR-ARCH], section 2.5.1). They are formed by appending a 32-bit IPv4 address to the 32-bit leading token '0000:5EFE', then setting the universal/local ("u") bit as follows: When the IPv4 address is known to be globally unique (i.e., provider-assigned),unique, the "u" bit is set to 1 and the leading token becomes '0200:5EFE'. When the IPv4 address is from a private allocation [RFC1918],or not otherwise known to be globally unique, the "u" bit is set to 0 and the leading token remains as '0000:5EFE'. Global and link-localSee: Appendix B for additional non-normative details. 6.2 ISATAP Addresses Any IPv6 unicast addressesaddress ([ADDR-ARCH], sections 2.5.4, 2.5.6) forsection 2.5) that has an ISATAP interface identifier and an on-link prefix on an ISATAP interface is considered an ISATAP address. ISATAP addresses are constructed as follows: | 64 bits | 32 bits | 32 bits | +------------------------------+---------------+----------------+ | global/link-localprefix | 000[0/2]:5EFE | IPv4 Address | +------------------------------+---------------+----------------+ (Appendix B provides additional non-normative details.) 4.2 ISATAP Interface Management The IP Tunnel MIB [MIB] is used, with the following additions for ISATAP interfaces: o For each IPv4 address an6.3 Multicast/Anycast ISATAP interface is configured over,interfaces recognize a tuple consisting of the IPv4 address and ifIndex for the corresponding IPv4 interface ([RFC2863],node's required IPv6 multicast/anycast addresses ([ADDR-ARCH], section 3.1.5) is added to ifRcvAddressTable ([MIB],2.8). Section 8.2 discusses encapsulation for multicast/anycast packets. 6.4 Source/Target Link Layer Address Options Source/Target Link Layer Address Options ([RFC2461], section 3.1.2). o tunnelIfRemoteInetAddress in the tunnelIfEntry object ([MIB], section 4) is set to 0.0.0.0 for ISATAP interfaces. When an IPv4 address over which an ISATAP interface is configured is removed from its IPv4 interface, the corresponding (IPv4 addres, ifIndex)-tuple MUST be removed from the ISATAP interface ifRcvAddressTable. If the IPv4 address is also used as tunnelIfLocalInetAddress ([MIB], section 5) in the ISATAP interface tunnelIfEntry, the interface MUST either set tunnelIfLocalInetAddress to a different IPv4 address or be disabled. When a new IPv4 address is added to an IPv4 interface an ISATAP interface is configured over, a new (IPv4 address, ifIndex)-tuple MAY be added to ifRcvAddressTable and tunnelIfLocalInetAddress MAY be set to the new address. 4.3 Multicast and Anycast ISATAP interfaces recognize an IPv6 node's required addresses ([ADDR-ARCH], section 2.8). The following multicast mappings are defined for packets sent on ISATAP interfaces: o When the IPv6 destination address is the 'All-Routers' ([ADDR-ARCH], section 2.7.1) or 'All_DHCP_Relay_Agents_and_Servers' ([RFC3315], section 1.2) multicast address, it is mapped to V4ADDR(i) for one or more PRL(i)'s (see: Section 6.1). The manner of selecting PRL(i)'s is up to the implementation. Other multicast mappings, and mechanisms for general-purpose multicast/anycast emulation on ISATAP interfaces are beyond the scope of this document. 4.4 Source/Target Link Layer Address Options Source/Target Link Layer Address Options ([RFC2461], section 4.6.1) for ISATAP have4.6.1) for ISATAP have the following format: +-------+-------+-------+-------+-------+-------+-------+--------+ | Type |Length | 0 | 0 | IPv4 Address | +-------+-------+-------+-------+-------+-------+-------+--------+ Type: 1 for Source Link-layer address. 2 for Target Link-layer address. Length: 1 (in units of 8 octets). IPv4 Address: TheA 32 bit IPv4 address, in network byte order. 5. Automatic Tunnelingorder ([RFC2223bis], section 3.4). 7. Configuration and Management Requirements 7.1 Network Management ISATAP interfaces use the basic transition mechanisms specified in [MECH] withnodes MAY support network management; ISATAP nodes that support network management SHOULD support the following exceptions: 5.1 Tunnel MTU and FragmentationMIBs: [FTMIB][IPMIB][TUNNELMIB]. The specification in ([MECH], section 3.2) is not used;configuration objects cited throughout the specification inremainder of this section is used instead. The minimum MTU for IPv6 interfaces is 1280 bytes ([RFC2460], Section 5), butdocument were selected to match the following operational considerations are noted: o Nearlynames of MIB objects. ISATAP nodes that do not support network management MAY choose their own local representation of these objects. 7.2 ISATAP Driver API The ISATAP driver provides an API for tunnel interface configuration and management that may be accessed by processes running on the ISATAP node, e.g., startup scripts, manual command line entry, kernel processes, ISATAP server daemons, etc. Access MUST be restricted to privileged users and applications. The API provides the following primitives; operational details are given in the subsections that follow: 'TUNNEL_CREATE': creates a tunnel interface. Takes as parameters a tunnel encapsulation method, parameters for setting read-write objects for the tunnel, and a list of receive addresses to initialize a forwarding entry in the system's ifRcvAddressTable. Returns an index for the new tunnel interface, or a failure code. 'TUNNEL_DELETE': deletes an existing tunnel interface. Takes as parameter an index of the tunnel interface to be deleted. Returns success or a failure code. 'TUNNEL_MODIFY': adds or deletes attributes for an existing tunnel interface, and its corresponding forwarding entry in the ifRcvAddressTable. Takes the same list of parameters as for TUNNEL_CREATE, plus a flag that denotes the operation (i.e., "add" or "delete"). Returns success or a failure code. 'TUNNEL_DUP': duplicates an existing tunnel interface. Takes as parameter the index of the tunnel interface to be duplicated. Returns an index for the newly-created tunnel interface, or a failure code. 'TUNNEL_GET': copies configuration attributes from system table entries associated with the specified tunnel interface into a user's buffer. Takes as parameter an index of a tunnel interface. Returns the number of system table entry data bytes written into the application's buffer or a failure code. 7.2.1 TUNNEL_CREATE ISATAP drivers implement a 'TUNNEL_CREATE' primitive that provides a means for configuring the 'tunnelIfEncapsMethod', all read-write objects associated with the 'tunnelIfEntry', and a list of receive addresses for the tunnel which consist of an IPv4 address and an index for the interface to which the address is assigned (i.e,. an IPv4 address-to-interface mapping). When a process on the ISATAP node issues 'TUNNEL_CREATE' primitive, it includes a parameter for configuring the 'tunnelIfEncapsMethod' object, and MAY include parameters for configuring other read-write objects in the 'tunnelIfEntry'. It MAY also include one or more receive address parameters. (Any required configuration parameters not included in the 'TUNNEL_CREATE' primitive are to be issued in a subsequent 'TUNNEL_MODIFY' primitive.) When the ISATAP driver processes a 'TUNNEL_CREATE' primitive, it creates an entry in the 'tunnelInetConfigTable', which results in the simultaneous creation of a 'tunnelIfEntry' in the 'tunnelIfTable' and an 'ifEntry' in the appropriate 'ifTable'. Next, it sets the 'tunnelIfEncapsMetod' object to the 'IANAtunnelType' specified by the primitive, and sets any other "read-write" objects in the 'tunnelIfEntry' based on parameters included. After configuring the 'tunnelIfEntry', the driver uses each receive address parameter included to locate a preferred 'ipAddressEntry' in the system's 'ipAddressTable'. It then creates an entry for the new tunnel interface in the 'ifRcvAddressTable' that includes the list of selected 'ipAddressEntry's, 'tunnelLocalInetAddress', 'tunnelRemoteInetAddress', 'tunnelIfEncapsMethod', and the 'ifIndex' for the tunnel interface. After performing the above actions, the ISATAP driver returns either an interface index for the newly-created tunnel interface or a failure code. 7.2.2 TUNNEL_DELETE ISATAP drivers implement a 'TUNNEL_DELETE' primitive that provides a means for deleting all table entries associated with a tunnel interface. When an ISATAP node's process issues a 'TUNNEL_DELETE' primitive, it includes an index for the tunnel interface returned via a previous 'TUNNEL_CREATE' or 'TUNNEL_DUP' primitive. When the ISATAP driver processes a 'TUNNEL_DELETE' primitive, it locates the 'tunnelInetConfigEntry' for the tunnel interface based on the interface index parameter and deletes the entry from the 'tunnelInetConfigTable'. This has the result of simultaneously deleting the 'tunnelIfEntry' and 'ifEntry' from their respective tables. The driver also removes the corresponding forwarding table entry for the tunnel interface from the 'ifRcvAddressTable'. After performing the above actions, the ISATAP driver returns either success or a failure code. 7.2.3 TUNNEL_MODIFY ISATAP drivers implement a 'TUNNEL_MODIFY' primitive that provides a means for modifying all read-write objects associated with the 'tunnelIfEntry' and for adding or deleting entries from the list of receive addresses for the tunnel. The primitive also provides a flag for specifying whether the desired operation is "add" or "delete". (For vector objects, the "add"/"delete" operations have the meaning intended by their names; for scalar objects, the ISATAP driver interprets an "add" operation as: "change to new value" and a "delete" operation as: "reset to default".) When an ISATAP node's process issues a 'TUNNEL_MODIFY' primitive, it includes an index for the tunnel interface returned via a previous 'TUNNEL_CREATE' or 'TUNNEL_DUP' primitive, and also includes a flag that specifies "add" or delete". It MAY include one or more parameter for configuring read-write objects in the 'tunnelIfEntry' and MAY also include one or more receive address (formatted as for 'TUNNEL_CREATE'). When the ISATAP driver processes a 'TUNNEL_MODIFY' primitive, it locates the correct 'tunnelIfEntry' for the interface index parameter and modifies objects for the entry based on any included parameters. If one or more receive address parameters are included, the driver also adds or deletes receive addresses from the forwarding table entry in the 'ifRcvAddressTable' corresponding to the 'tunnelIfEntry'. If no parameters are included, the result is a NO-OP. After performing the above actions, the ISATAP driver returns either success or a failure code. 7.2.4 TUNNEL_DUP ISATAP drivers implement a 'TUNNEL_DUP' primitive that creates a new tunnel interface by duplicating a set of system table entries from an existing tunnel interface. When a user application or a system process issues a 'TUNNEL_MODIFY' primitive, it includes an index for the tunnel interface to be duplicated from a previous 'TUNNEL_CREATE' or 'TUNNEL_DUP' primitive. When the ISATAP driver processes a 'TUNNEL_DUP' primitive, it creates a new entry in the 'tunnelInetConfigTable' exactly as for 'TUNNEL_CREATE' (see: Section 7.2.1). Next, it locates the 'tunnelIfEntry' and 'ifEntry' for the tunnel interface to be duplicated and copies all attributes from those objects into the newly-created 'tunnelIfEntry' and 'ifEntry'. The driver also creates a duplicate forwarding table entry in the 'ifRcvAddressTable' using the existing entry identified by the interface index parameter as a prototype, then sets the newly-created forwarding entry's index to the 'ifIndex' for the newly-created tunnel interface. After performing the above actions, the ISATAP driver returns either an interface index for the newly-created tunnel interface or a failure code. 7.2.5 TUNNEL_GET To aid network administrators, ISATAP drivers SHOULD implement a 'TUNNEL_GET' primitive that returns the current configuration of all tables in the system associated with the specified tunnel interface. When a user application issues a 'TUNNEL_GET' primitive, it includes an index for a tunnel interface from a previous 'TUNNEL_CREATE' or 'TUNNEL_DUP' primitive, a pointer to a character buffer to receive the configuration information, and an integer indicating the available space in the buffer. When the ISATAP driver processes a 'TUNNEL_GET' primitive, it prepares a character string that includes the concatenation of the 'tunnelIfEntry' and the 'ifRcvAddressTable' entry for the tunnel interface identified by the index parameter. (The 'ifEntry' is not concatenated, since its contents may be examined via primitives from other APIs.) Next, the driver copies the character string to the server daemon's character buffer up to the specified available buffer space. After performing the above actions, the ISATAP driver returns either the number of bytes copied or a failure code (to include a code that indicates "operation not supported"). 7.3 ISATAP Interface Configuration ISATAP interfaces are normally configured by an ISATAP node's system startup scripts or via manual configuration, but may also be created by a dynamic process. When a node creates (or later modifies) an ISATAP interface, it assigns to the interface one or more receive address that consists of an IPv4 address and an index for the interface to which the address is assigned (i.e,. an IPv4 address-to-interface mapping). Each receive address assigned MUST represent a mapping for the same site (or, MUST represent a mapping that is routable on the global Internet), i.e., the receive addresses assigned to a single tunnel interface MUST NOT span multiple sites. ISATAP nodes issue 'TUNNEL_CREATE' and 'TUNNEL_MODIFY' primitives for ISATAP interfaces the same as for any tunnel interface; 'TUNNEL_CREATE' primitives include a parameter to set 'tunnelIfEncapsMethod' to an 'IANATunnelType' code for "isatap". A 'TUNNEL_CREATE' or 'TUNNEL_MODIFY' primitive that includes parameters to set 'tunnelIfLocalInetAddress' to an IPv4 address that will be used as one of the interface's receive addresses, and 'tunnelIfRemoteInetAddress' to 0.0.0.0 to denote wildcard match for remote tunnel endpoints SHOULD be issued before the IPv6 interface associated with the tunnel interface is enabled (see below). When an ISATAP interface is created, the ISATAP driver also creates an 'ipv6InterfaceEntry' as the companion 'ifEntry' to the 'tunnelIfEntry'. After setting the required objects in the 'tunnelIfEntry' (see above), the ISATAP node configures objects in the 'ipv6InterfaceEntry' for an ISATAP interface the same as for any IPv6 interface. For ISATAP interfaces (and other tunnel interfaces that use IPv4 as a link layer for IPv6 ), the node sets the 'ipv6InterfaceType' object to "tunnel". Next, the node sets the 'ipv6InterfacePhysicalAddress' object to an IPv4 address that will be used as one of the tunnel interface's receive addresses; this object MUST be formatted as a 4-octet entity containing an IPv4 address in network byte order ([RFC2223bis], section 3.4). The node next sets the 'ipv6ScopeZoneIndexLinkLocal' object to a zone index identifier that denotes the site for which the tunnel interface's receive addresses are valid. Finally, the node configures all other required read-write parameters in the 'ipv6InterfaceEntry' as for any IPv6 interface, and sets 'ipv6InterfaceAdminStatus' to "up". After configuring the ISATAP interface, the node sets the interface's 'ipv6InterfaceForwarding' object (and, if necessary, the node's 'ip6Forwarding' object) to "forwarding". The node also creates an 'ipv6RouterAdvertEntry' in the 'ipv6RouterAdvertTable' and sets the 'ipv6RouterAdvertIfIndex' object to the same value as 'ipv6InterfaceIfIndex'. Objects in the 'ipv6RouterAdvertEntry' for an ISATAP interface are configured as for any IPv6 router, however 'ipv6RouterAdvertLinkMTU' SHOULD NOT be set to a value other than 0 unless the ISATAP driver will monitor the IPv4 reassembly cache and report fragmentation of tunneled packets to the source by sending IPv6 Router Advertisements with MTU options (see: Section 8.3). Configuration of objects relating to IPv6 forwarding is normally managed by the ISATAP server daemon. 7.4 Dynamic Creation of Configured Tunnels Configured tunnels are normally created through ISATAP driver API calls issued by an ISATAP server daemon in dynamic response to a tunnel creation request. Configured tunnel interfaces are created as for ISATAP interfaces (see: Section 7.3), except that the 'tunnelIfRemoteInetAddress' for the new entry is normally set to a specific IPv4 address for a remote node at the far end of the tunnel, i.e., configured tunnels are normally configured as point-to-point. As for ISATAP interfaces, configured tunnels MUST NOT select a list of receive address mappings that span multiple sites. Processes that create configured tunnels may find the 'TUNNEL_DUP' primitive useful (and, in some cases essential) for reducing administrative complexity. An ISATAP interface may be used as the prototype for the 'TUNNEL_DUP' primitive; the configured tunnel interface inherits the attributes of the ISATAP interface, including the forwarding table entry in the system's 'ifRecvAddressTable'. After creating a configured tunnel via the 'TUNNEL_DUP' primitive, the process uses the 'TUNNEL_MODIFY' primitive to modify specific attributes. 7.5 Reconfigurations Due to IPv4 Address Changes When an 'ipAddressEntry' becomes deprecated (e.g., when an IPv4 address is removed from an IPv4 interface) the 'ipAddressEntry' MUST be removed from all forwarding entries in the 'ifRcvAddressTable' that referenced it. Also, all 'tunnelIfEntry's that used 'ipAddressAddr' as 'tunnelIfLocalInetAddress' and 'ipv6InterfaceEntry's that used 'ipAddressAddr' as 'ipv6InterfacePhysicalAddress' MUST select a different IPv4 address for those objects from their remaining list of receive addresses. Methods for triggering the mandatory changes are implementor's choice. When a new IPv4 address is added to an IPv4 interface, the node MAY add the new 'ipAddressEntry' to the list of receive addresses for forwarding entries in 'ifRcvAddressTable', and MAY set 'tunnelIfLocalInetAddress' and/or 'ipv6InterfacePhysicalAddress' for interfaces referenced by the updated forwarding entries to the new address. 8. Automatic Tunneling ISATAP nodes use the basic tunneling mechanisms specified in [MECH]. The following additional specifications are used for ISATAP: 8.1 Encapsulation The ISATAP driver is responsible for inserting the outermost IPv4 encapsulating header for all tunneled packets. Tunnel interfaces that use various encapsulation methods (e.g., 6over4 [RFC2529], 6to4 [RFC3056], teredo, IP encapsulation within IP [RFC2003], minimal encapsulation within IP [RFC2004], basic IPv6-in-IPv4 encapsulation [MECH], ISATAP encapsulation itself, etc.) can all be configured as encapsulation clients of the ISATAP driver. The ISATAP driver performs AH [RFC2402] and ESP [RFC2406] processing for tunnels that use IPsec, and may also perform header compression prior to encapsulation. 8.2 Multicast/Anycast ISATAP interfaces tunnel only those packets with IPv6 multicast/ anycast destinations to include a node's required multicast/anycast addresses, the 'All_DHCP_Relay_Agents_and_Servers' and 'All_DHCP_Servers' multicast addresses [RFC3315] and multicast addresses discovered via MLD [RFC2710]. Packets with unrecognized IPv6 multicast/anycast destinations are silently dropped. ISATAP interfaces automatically tunnel IPv6 multicast packets with the 'All_DHCP_Relay_Agents_and_Servers' and 'All_DHCP_Servers' using the IPv4 'all hosts' broadcast address (i.e., 0xffffffff broadcast address) as the destination address in the encapsulating IPv4 header under the assumption that DHCPv6 servers will be co-located with DHCPv4 servers. For other IPv6 multicast destinations, ISATAP interfaces automatically tunnel packets using a mapped Organization-Local Scope IPv4 multicast address ([RFC2529], section 6) as the destination address in the encapsulating IPv4 header. ISATAP nodes join the Organization-Local Scope IPv4 multicast groups required to support IPv6 Neighbor Discovery ([RFC2529], Appendix A) on interfaces that may receive IPv4 packets to be forwarded to an ISATAP interface. NOTE: When the ISATAP node enables one or more 6over4 interfaces [RFC2529], the 6over4 interfaces MAY be used (i.e., instead of ISATAP interfaces) for sending and receiving multicast packets. 8.3 Tunnel MTU and Fragmentation The specification in ([MECH], section 3.2) is not used; the specification in this section is used instead: ISATAP interfaces set a static MTU of 1280 bytes, i.e., the minimum MTU for IPv6 interfaces ([RFC2460], section 5) and do not set the Don't Fragment bit in the encapsulating IPv4 headers of tunneled packets. ISATAP interfaces MAY provide a configuration knob for setting a larger MTU, but larger MTUs MUST NOT be configured other than for certain constrained deployments, e.g., in some enterprise networks). Interfaces that may receive IPv4 nodes connectpackets to physical links with MTUsbe forwarded to an ISATAP interface SHOULD configure an Effective MTU to Receive (EMTU_R) [RFC1122], section 3.3.2) of at least 1500 bytes, i.e., they SHOULD be able to reassemble IPv4 packets of 1500 bytes or larger (e.g., Ethernet) o Sub-IPv4larger. 1280 bytes was chosen as the IPv6 interface minimum MTU [DEERING97] to allow extra room for link layer encapsulations (e.g., VPN) may occur on some paths o Commonly-deployed VPN interfaces use anwithout exceeding the Ethernet MTU of 1400 bytes To maximize efficiency and minimize IPv4 fragmentation1500 bytes, i.e., the practical physical cell size of the Internet. The 1280 byte MTU provides a fixed upper bound for the predominant deployment case, LinkMTUsize of IPv6 packets/fragments with a maximum store-and-forward delay budget of ~1 second assuming worst-case link speeds of ~10Kbps [RFC3150], thus allowing a convenient value for ISATAP interfaces SHOULD be setuse in reassembly buffer timer settings. Finally, the 1280 byte MTU allows transport connections (e.g., TCP) to configure a large-enough maximum segment size for improved performance even if the IPv4 interface that will send the tunneled packets uses a smaller MTU. When the size of the IPv6 destination's receive buffer is known, applications MAY send IPv6 packets up to that size using IPv6 fragmentation (or, fragmentation via an alternate form of encapsulation) with a maximum fragment size that is no morelarger than 1380 bytes (1400 minus 20 bytesthe minimum of the MTU of the IPv4 interface used for tunneling and 1280 bytes. Even so, IPv4 encapsulation). LinkMTUfragmentation MAY be setstill occur along some paths; in particular, since the minimum IPv4 fragment size is only 8 bytes ([RFC0791], section 2), middleboxes with unusual implementations of IPv4 fragmentation could shatter the tunneled packets into as many as 187 IPv4 fragments to larger values whenaccommodate a dynamic1500 byte IPv4 packet. Such sustained bursts of small packets could result in poor performance due to increased loss probability on paths with non-negligible packet loss due to, e.g., link layer (IPv4) MTU discovery mechanism is used,errors, congested router queues, etc. Therefore, ISATAP nodes that anticipate or experience poor performance along some paths MAY choose to adaptively vary the maximum size for the packets/fragments they send. For example, implementations may choose to employ a "fragment size slow start" scheme that begins with as little as 8 bytes (i.e., the minimum IPv4 fragment size) and varies the size of the fragments using, e.g., an additive-increase, multiplicative-decrease strategy to determine the size that yields the best performance. The process can be made to converge more quickly when a staticnext-hop IPv6 routers are configured to send Router Advertisements with MTU assignmentoptions when they experience IPv4 fragmentation, since the sender is usedmade aware that fragmentation is occurring, and the anticipated/measured levelMTU option can be used to return the size of fragmentationthe largest IPv4 fragment observed which may help the sender determine the optimal fragment size. Since many nodes are expected to implement this specification, an overall increase in small packets in the Internet may occur as more nodes with tunnel interfaces implement schemes such as the site'sone described above to avoid IPv4 fragmentation-related performance issues. For this reason, network is deemed acceptable. When a dynamic link layer MTU discovery mechanism is not used,equipment manufacturers and network administrators are encouraged to observe the Don't Fragment (DF) bit MUST NOT be setRecommendations on Queue Management and Congestion Avoidance in the encapsulatingInternet [RFC2309]. In particular, byte mode queue averaging for RED is encouraged. With reference to the above, it is RECOMMENDED that ISATAP nodes use adaptive techniques to minimize IPv4 headerfragmentation and use IPv6 fragmentation/reassembly (or, fragmentation/reassembly via an alternate form of encapsulation) to manage the size of packets sent onthe tunneled packets they send. It is also RECOMMENDED that ISATAP interface. In this case, black holes maynodes monitor the IPv4 reassembly cache in rare instances occur along some paths even whenorder to give early indications of IPv4 network fragmentation by sending Router Advertisements with MTU options to the tunnel interface usessource of the IPv6 minimumIPv4 fragments. The MTU options should include a value to indicate the size of 1280 bytes. (This concern is not specificthe largest packet that can be expected to arrive without incurring IPv4 fragmentation. Finally, it is RECOMMENDED that ISATAP interfaces, but applies to all tunnelsnodes set small timeout values, e.g. 1 second, for which nested levelsIPv4 reassembly of sub link-layer encapsulation may occur.) 5.2tunneled packets. 8.4 Handling IPv4 ICMP Errors ISATAP interfaces SHOULD process ARP failures and persistent ICMPv4 errors SHOULD be processedas link-specific information indicating that a path to a neighbor hasmay have failed ([RFC2461], section 7.3.3). 5.38.5 Link-Local Addresses The specification in ([MECH], section 3.7) is not used; the specification in Section 4.16.1 of this document is used instead. 5.48.6 Neighbor Discovery over Tunnels The specification in ([MECH], section 3.8) is not used; the specificationsspecification in Section 6 and Section 79 of this document areis used instead. 5.58.7 Decapsulation/Filtering The specifications in ([MECH], sections 3.6, 3.9 and 4.1) are used. In addition,ISATAP nodes arrange for the decapsulator MUST determineISATAP driver to received all tunneled packets that use an IPv4 header as the outermost layer of encapsulation. Examples include ip-protocol-41 (6to4, 6over4, isatap, etc.), ip-protocol-4 (IP encapsulation within IP, minimal encapsulation within IP, etc.), UDP port 3544 (teredo, etc.) and others. The ISATAP driver determines the correct tunnel interface to receive each IPv4 protocol-41packet via a tablelookup forin the tuple consisting of'ifRcvAdddressTable' for the packet's IPv4 source andaddress, destination address, and ifIndexan index for the receiving IPv4 interface. (Note that ISATAP interfaces match all IPv4 source addresses by default; if a tunnelinterface with a more-specific match on the IPv4 source address exists, it is selected to receiveand the packet as for longest-prefix-match.)type of encapsulation. Packets for which the correct tunnel interface cannot be determined are discarded; in this case, the decapsulator MAY also send an ICMPv4 Destination Unreachable message with code 3 (Port Unreachable) ([RFC1122], section 22.214.171.124) to the IPv4 source address in the packet's outer header.silently discarded. After determining the correct tunnel interface, the decapsulator MUST also verifyISATAP driver verifies that the packet's link-layer (IPv4) source address is correct for the network-layer (IPv6) source address. For configured tunnels, the IPv4 and IPv6 source addresses can be checked directly against the configured tunnel's addresses. For ISATAP interfaces, the packet's link-layer source address is correct if one (or more) of the following are true: o the network-layer source address is an ISATAP address that embeds the link-layer source address in its interface identifier. o the network-layer source address is an IPv6 neighbor withinon an interface that has the same site'ipv6ScopeZoneIndexLinkLocal' as the receiving ISATAP interface,receiving ISATAP interface. o the link-layer source address is a member of the Potential Router List (see: Section 9.1). Packets for which the link-layer source address is incorrect are discarded and, if permitted by the current status of ICMPv6 message rate limiting parameters [ICMPV6], section 2.4, paragraph f), an ICMPv6 Destination Unreachable message SHOULD be generated and sent to the IPv6 source in the inner header of the encapsulated packet. The error message SHOULD include only enough bytes from the invoking packet to convey the IPv6 header information, i.e., it SHOULD NOT include up to the minimum IPv6 MTU. After determining the correct tunnel interface to receive the packet, the ISATAP driver examines the IPv6 and IPv4 source addresses to determine whether a rewrite is required. If the IPv6 source address is an ISATAP address with the 'u/l' and 'g' bits set to 0 (see: Section 6.1), and the IPv4 source address does not match the IPv4 address encoded in the ISATAP interface identifier, the ISATAP driver copies the IPv4 source address over the IPv4 address embedded in the IPv6 address and sets the 'u/l' bit to 1. Other forms of rewrites (e.g., rewrites for multicast rendezvous points based on the 'u' and 'g' bit) MAY be specified in other documents. Next, the link-layer source address matchesISATAP driver discards the link layer address inencapsulating IPv4 header and locates any existing host-pair information, e.g., via the neighbor cache.IPv6 Flow Label [FLOW]. Then: o the link-layer source addressIf header compression is indicated, the packet's inner header(s) are reconstituted. o If a member ofsecurity association is indicated, AH [RFC2402] or ESP [RFC2406] processing is applied. o If the Potential Router Listpacket is a fragment, it is placed in a buffer for reassembly. The buffer may be, e.g., the site (see: Section 6.1). Packets for whichIPv6 reassembly cache, an application's own data buffer [RFC3542], etc. Finally, when a whole packet has been received, it is delivered to the link-layer source addresscorrect tunnel interface. If there is incorrect are discarded, andclear evidence that reassembly of a fragmented packet has stalled, an ICMPv6 Destination Unreachable"packet too big" message ([ICMPV6], section 3.1) SHOULD be[RFC1981] is sent to the IPv6packet's source in the inner header of the encapsulated packetaddress (subject to rate limiting as in [ICMPV6], section 2.4, paragraph f). 6.ICMPv6 rate-limiting) with an MTU value indicating a size that is likely to incur successful reassembly. 9. Neighbor Discovery ISATAP interfacesnodes use the neighbor discovery mechanisms specified in [RFC2461] along with securing mechanisms such as [SEND] to create/ change neighbor cache entries and to provide control plane signalling for automatic tunnel configuration. ISATAP interfaces also implement the following exceptions: 6.1specifications: 9.1 Conceptual Model Of A Host To the list of Conceptual Data Structures ([RFC2461], section 5.1), ISATAP interfaces add: Potential Router List A set of entries about potential routers for the site;routers; used to support the mechanisms specified in Section 126.96.36.199.2.2.1. Each entry ("PRL(i)") has an associated timer ("TIMER(i)"), and an IPv4 address ("V4ADDR(i)") that represents a router's advertising ISATAP interface. 6.29.2 Router and Prefix Discovery 6.2.1 Message Validation 188.8.131.52 Validation of Router Solicitation Messages To the list of validity checks for Router Soliciation messages ([RFC2461], section 6.1.1), ISATAP interfaces add: o If the message includes a Source Link Layer Address Option, the message also includes an IP authentication Header. 184.108.40.206 Validation of Router Advertisement Messages To the list of validity checks for Router Advertisement messages ([RFC2461], section 6.1.1), ISATAP interfaces add: o IP Source Address is an ISATAP link-local address that embeds V4ADDR(i) for some PRL(i). o If the message includes a Source Link Layer Address Option, the message also includes an IP authentication Header. 220.127.116.11.1 Router Specification As permitted by ([RFC2461], section 6.2.6), advertisingthe ISATAP interfacesserver daemon SHOULD send unicast Router Advertisement messages to the soliciting host'snode's address when the solicitation's source address is not the unspecified address. 18.104.22.168.2 Host Specification 22.214.171.124.2.2.1 Host Variables To the list of host variables ([RFC2461], section 6.3.2), ISATAP interfaces add: PrlRefreshInterval Time in seconds between successive refreshments of the PRL after initialization. It SHOULD be no less than 3600 seconds. The designated value of all 1's (0xffffffff) represents infinity. Default: 3600 seconds MinRouterSolicitInterval Minimum time in seconds between successive solicitations of the same advertising ISATAP interface. It SHOULD be no less than 900 seconds. The designated value of alll 1's (0xffffffff) represents infinity. Default: 900 seconds 126.96.36.199.2.2.2 Interface Initialization The hostISATAP node joins the all-nodes multicast address on ISATAP interfaces, as for multicast-capable interfaces ([RFC2461], section 6.3.3).6.3.3) and MAY also join other multicast groups, e.g., see: Section 8.2 Additionally, the hostnode provisions the ISATAP interface's PRL with IPv4 addresses it discovers via manual configuration, a DNS fully-qualified domain name (FQDN) [RFC1035], a DHCPv4 option for ISATAP [ISDHCP],option, a DHCPv4 vendor-specific option, or an unspecified alternate method. (Support for manual configuration is REQUIRED; other methods are OPTIONAL.) WhenISATAP nodes establish FQDNs are used, the host establishes the FQDNvia manual configuration or an unspecified alternate method. (Support for manual configuration is REQUIRED; other methods are OPTIONAL.) The hostNodes resolves the FQDNFQDNs into IPv4 addresses through lookup in a static host file, a site-specific name service,querying the site'sDNS service, or an unspecified alternate method. When DNS is used, client resolvers use the IPv4 transport. After the hostnode provisions the ISATAP interface's PRL with IPv4 addresses, it sets PrlRefreshIntervalTimer to PrlRefreshInterval seconds. The hostnode re-initializes the PRL (i.e., as specified above) when PrlRefreshIntervalTimer expires, or when an asynchronous re-initialization event occurs. When the hostnode re-initializes the PRL, it resets PrlRefreshIntervalTimer to PrlRefreshInterval seconds. 188.8.131.52.2.2.3 Processing Received Router Advertisements The ISATAP server daemon processes Router Advertisements (RAs) are processedexactly as specified in ([RFC2461], section 6.3.4) except that, if the6.3.4). Router Advertisement messages received on a point-to-point tunnel interface that contain an MTU option is present, the option's value SHOULD be stored inwith a per-neighbor cache entry forvalue less than 1280 bytes cause the source ofinterface to reduce its MTU to the RA; itlesser value, but Router Advertisements received on an ISATAP interface MUST NOT be copied into LinkMTU forcause the ISATAP interface. Additionally, hostsinterface to reduce its MTU to a value less than 1280 bytes. For Router Advertisement messages received on an ISATAP interface that include prefix options and/or non-zero values in the Router Lifetime, the server daemon reset TIMER(i) to schedule the next solicitation event (see: Section 184.108.40.206).220.127.116.11). Let "MIN_LIFETIME" be the minimum value in the Router Lifetime or the lifetime(s) encoded in options included in the RA message. Then, TIMER(i) is reset as follows: TIMER(i) = MAX((0.5 * MIN_LIFETIME), MinRouterSolicitInterval) 18.104.22.168.2.2.4 Sending Router Solicitations To the list of events after which RSs may be sent ([RFC2461], section 6.3.2), ISATAP interfaces add: o TIMER(i) for some PRL(i) expires. Additionally, hoststhe ISATAP server daemon MAY send Router Solicitations to an ISATAP link-local address that embeds V4ADDR(i) for some PRL(i) instead of the All-Routers multicast address. 6.39.3 Address Resolution and Neighbor Unreachability Detection 6.3.1 Message Validation 22.214.171.124 Validation of Neighbor Solicitations To the list of validity checks for Neighbor Solicitation (NS) messages ([RFC2461], section 7.1.1), ISATAP interfaces add: o If the message includes a Source Link Layer Address Option, the message also includes an IP authentication Header. 126.96.36.199 Validation of Neighbor Solicitations To the list of validity checks for Neighbor Advertisement (NA) messages ([RFC2461], section 7.1.2), ISATAP interfaces add: o If the message includes a Target Link Layer Address Option, the message also includes an IP authentication Header. 188.8.131.52.1 Address Resolution The specification in ([RFC2461], section 7.2) is used. NS and NA messages MAY omit the source/target link layer address option when the source/target is an ISATAP address.ISATAP addresses for which the neighbor'sneighbor/router's link-layer address cannot otherwise be determined (i.e.,(e.g., from thea neighbor cache or a link layer address option in a received packet)entry) are resolved to link-layer addresses by a static computation, i.e., the last four octets are treated as an IPv4 address. Hosts SHOULD perform an initial reachability confirmation by sending NSNeighbor Solicitation message(s) and receiving a NA message; NSNeighbor Advertisement message (NS messages are sent to the target's unicast address.address). Routers MAY perform anthis initial reachability confirmation, but this might not scale in all environments. As specified in ([RFC2461], section 7.2.4), all nodes MUST send solicited neighbor advertisementsNeighbor Advertisements on ISATAP interfaces. 184.108.40.206.2 Neighbor Unreachability Detection Hosts SHOULD perform Neighbor Unreachability Detection as specified in([RFC2461], section 7.3). Routers MAY perform neighbor unreachability detection, but this might not scale in all environments. 6.4 Redirect Function To the list of validity checks for Redirect messages (([RFC2461], section 8.1), ISATAP interfaces add: o If the message includes a Target Link Layer Address Option, the message also includes an IP authentication Header. 7. Address Autoconfigurationunreachability detection, but this might not scale in all environments. 10. Other Requirements for Control Plane Signalling 10.1 Node Information Queries ISATAP interfaces use the address autoconfiguration mechanismsnodes SHOULD implement Node Information Queries as specified in [RFC2462] with[NIQUERY]. Node Information Queries/Responses provide the following exceptions: 7.1 Address Lifetime Expiry The specification in ([RFC2462], section 5.5.4) is used, exceptadvantages: o the querier receives unambiguous confirmation that an ISATAP address also becomes deprecated whenthe IPv4 address embedded in its interface identifier is removed from an IPv4 interface over whichresponder supports the ISATAP interface is configured. (This deprecation rule applies to all ISATAP addresses, including link-local addresses.) 7.2 Stateful Address Autoconfiguration Whenprotocol. o the site uses DHCPv6 [RFC3315] asquerier receives assurance that responses are coming from the stateful address autoconfiguration mechanism,correct responder. o the server/relay function MUSTquerier discovers some subset of the responder's addresses. 10.2 Linklocal Multicast Name Resolution (LLMNR) ISATAP nodes SHOULD implement Link Local Multicast Name Resolution [LLMNR], since they will commonly be deployed equally on each router that isin environments (e.g., home networks, ad-hoc networks, etc.) with no access to a member of the PRL. 8.Domain Name System (DNS) server. 11. IANA Considerations The IANA is advised to specify construction rules for IEEE EUI-64 addresses formed from the Organizationally Unique Identifier (OUI) "00-00-5E" in the IANA "ethernet-numbers" registry. The non-normative text in Appendix B is offered as an example specification. 9.12. Security considerations The security considerations in [RFC2461][RFC2462][MECH]the normative references apply. Additionally, siteadministrators MUST ensure that lists of IPv4 addresses representing the advertising ISATAP interfaces of PRL members are well maintained. 10.13. Acknowledgments Most of the basic ideas in this document are not original; the authors acknowledge the original architects of those ideas. Portions of this work were sponsored through SRI International internal projects and government contracts. Government sponsors include Monica Farah-Stapleton and Russell Langan (U.S. Army CECOM ASEO), and Dr. Allen Moshfegh (U.S. Office of Naval Research). SRI International sponsors include Dr. Mike Frankel, J. Peter Marcotullio, Lou Rodriguez, and Dr. Ambatipudi Sastry. The following are acknowledged for providing peer review input: Jim Bound, Rich Draves, Cyndi Jung, Ambatipudi Sastry, Aaron Schrader, Ole Troan, Vlad Yasevich. The following are acknowledged for their significant contributions: Alain Durand, Hannu Flinck, Jason Goldschmidt, Nathan Lutchansky, Karen Nielsen, Mohan Parthasarathy, Chirayu Patel, Art Shelest, Pekka Savola, Margaret Wasserman, Brian Zill. The authors acknowledge the work of Quang Nguyen [VET] under the guidance of Dr. Lixia Zhang that proposed very similar ideas to those that appear in this document. This work was first brought to the authors' attention on September 20, 2002. The following individuals are acknowledged for their helpful insights on path MTU discovery: Jari Arkko, Iljitsch van Beijnum, Jim Bound, Ralph Droms, Alain Durand, Jun-ichiro itojun Hagino, Brian Haberman, Bob Hinden, Christian Huitema, Kevin Lahey, Hakgoo Lee, Matt Mathis, Jeff Mogul, Erik Nordmark, Soohong Daniel Park, Chirayu Patel, Michael Richardson, Pekka Savola, Hesham Soliman, Mark Smith, Dave Thaler, Michael Welzl, Lixia Zhang and the members of the Nokia NRC/ COM Mountain View team. "...and I'm one step ahead of the shoe shine, Two steps away from the county line, Just trying to keep my customers satisfied, Satisfi-i-ied!" - Simon and Garfunkel Normative References [ADDR-ARCH] Hinden, R. and S. Deering, "IP Version 6 Addressing Architecture", draft-ietf-ipv6-addr-arch-v4-00 (work in progress), October 2003. [ICMPV6] Conta, A. and S. Deering, "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification", draft-ietf-ipngwg-icmp-v3 (work in progress), November 2001. [LLMNR] Esibov, L., Aboba, B. and D. Thaler, "Linklocal Multicast Name Resolution", draft-ietf-dnsext-mdns (work in progress), January 2004. [MECH] Gilligan, R. and E. Nordmark, "Basic Transition Mechanisms for IPv6 Hosts and Routers", draft-ietf-v6ops-mech-v2-00 (work in progress), February 2003. [MIB] Thaler, D., "IP Tunnel MIB", draft-thaler-inet-tunnel-mib[NIQUERY] Crawford, M., "IPv6 Node Information Queries", draft-ietf-ipngwg-icmp-name-lookups (work in progress), SeptemberJune 2003. [NODEREQ] Loughney, J., "IPv6 Node Requirements", draft-ietf-ipv6-node-requirements (work in progress), October 2003. [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, September 1981. [RFC1122] Braden, R., "Requirements for Internet Hosts - Communication Layers", STD 3, RFC 1122, October 1989. [RFC1981] McCann, J., Deering, S. and J. Mogul, "Path MTU Discovery for IP version 6", RFC 1981, August 1996. [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [RFC2434] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA Considerations Section in RFCs", BCP 26, RFC 2434, October 1998. [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, December 1998. [RFC2461] Narten, T., Nordmark, E. and W. Simpson, "Neighbor Discovery for IP Version 6 (IPv6)", RFC 2461, December 1998. [RFC2462] Thomson, S.[RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 Domains without Explicit Tunnels", RFC 2529, March 1999. [RFC3150] Dawkins, S., Montenegro, G., Kojo, M. and V. Magret, "End-to-end Performance Implications of Slow Links", BCP 48, RFC 3150, July 2001. [RFC3542] Stevens, W., Thomas, M., Nordmark, E. and T. Narten, "IPv6 Stateless Address Autoconfiguration",Jinmei, "Advanced Sockets Application Program Interface (API) for IPv6", RFC 2462, December 1998.3542, May 2003. Informative References [ISDHCP] Templin, F., "Dynamic Host Configuration Protocol (DHCPv4) Option[DEERING97] Deering, S., "http://www.cs-ipv6.lancs.ac.uk/ipv6/ mail-archive/IPng/1997-12/0052.html", November 1997. [FLOW] Rajahalme, J., Conta, A., Carpenter, B. and S. Deering, "IPv6 Flow Label Specification", draft-ietf-ipv6-flow-label (work in progress), December 2003. [FTMIB] Haberman, B. and M. Wasserman, "IP Forwarding Table MIB", draft-ietf-ipv6-rfc2096-update (work in progress), August 2003. [IPMIB] Routhier, S., "Management Information Base for the Intra-Site Automatic Tunnel AddressingInternet Protocol (ISATAP)", draft-templin-isatap-dhcp(IP)", draft-ietf-ipv6-rfc2011-update (work in progress), OctoberSeptember 2003. [RFC1035] Mockapetris, P., "Domain names - implementation and specification", STD 13, RFC 1035, November 1987. [RFC1122] Braden, R., "Requirements for Internet Hosts - Communication Layers", STD 3,[RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 1122,2003, October 1989. [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G.1996. [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, October 1996. [RFC2223bis] Reynolds, J. and E. Lear, "Address AllocationR. Braden, "Instructions to Request for Private Internets", BCP 5,Comments (RFC) Authors", draft-rfc-editor-rfc2223bis (work in progress), August 2003. [RFC2309] Braden, B., Clark, D., Crowcroft, J., Davie, B., Deering, S., Estrin, D., Floyd, S., Jacobson, V., Minshall, G., Partridge, C., Peterson, L., Ramakrishnan, K., Shenker, S., Wroclawski, J. and L. Zhang, "Recommendations on Queue Management and Congestion Avoidance in the Internet", RFC 1918, February 1996.2309, April 1998. [RFC2402] Kent, S. and R. Atkinson, "IP Authentication Header", RFC 2402, November 1998. [RFC2406] Kent, S. and R. Atkinson, "IP Encapsulating Security Payload (ESP)", RFC 2406, November 1998. [RFC2486] Aboba, B. and M. Beadles, "The Network Access Identifier", RFC 2486, January 1999. [RFC2491] Armitage, G., Schulter, P., Jork, M. and G. Harter, "IPv6 over Non-Broadcast Multiple Access (NBMA) networks", RFC 2491, January 1999. [RFC2529] Carpenter, B.[RFC2492] Armitage, G., Schulter, P. and C. Jung, "Transmission of IPv6M. Jork, "IPv6 over IPv4 Domains without Explicit Tunnels",ATM Networks", RFC 2529, March2492, January 1999. [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: Defeating Denial of Service Attacks which employ IP Source Address Spoofing", BCP 38, RFC 2827, May 2000. [RFC2863] McCloghrie, K. and F. Kastenholz, "The Interfaces Group MIB", RFC 2863, June 2000. [RFC3041] Narten, T.[RFC2710] Deering, S., Fenner, W. and R. Draves, "Privacy ExtensionsB. Haberman, "Multicast Listener Discovery (MLD) for Stateless Address Autoconfiguration inIPv6", RFC 3041, January2710, October 1999. [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via IPv4 Clouds", RFC 3056, February 2001. [RFC3315] Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C. and M. Carney, "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", RFC 3315, July 2003. [RFC3582] Abley, J., Black, B. and V. Gill, "Goals for IPv6 Site-Multihoming Architectures", RFC 3582, August 2003. [SEND] Arkko, J., Kempf, J., Sommerfield, B., Zill, B. and P. Nikander, "Secure Neighbor Discovery (SEND)", draft-ietf-send-ndopt (work in progress), October 2003. [TUNNELMIB] Thaler, D., "IP Tunnel MIB", draft-ietf-ipv6-inet-tunnel-mib (work in progress), January 2004. [VET] Nguyen, Q., "http://irl.cs.ucla.edu/vet/report.ps", spring 1998. Authors' Addresses Fred L. Templin Nokia 313 Fairchild Drive Mountain View, CA 94110 US Phone: +1 650 625 2331 EMail: firstname.lastname@example.org Tim Gleeson Cisco Systems K.K. Shinjuku Mitsu Building 2-1-1 Nishishinjuku, Shinjuku-ku Tokyo 163-0409 Japan EMail: email@example.com Mohit Talwar Microsoft Corporation One Microsoft Way Redmond, WA> 98052-6399 US Phone: +1 425 705 3131 EMail: firstname.lastname@example.org Dave Thaler Microsoft Corporation One Microsoft Way Redmond, WA 98052-6399 US Phone: +1 425 703 8835 EMail: email@example.com Appendix A. Major Changes Major changes from earlier versions to version 16:17: o dropped "underlying link" from terminology.added tunnel driver API o expanded section on MTU and fragmentation o expanded sections on encapsulation/decapsulation o specified relation to IPv6 Node Requirements o specified use of additional control plane signalling o specified multicast mappings. o specified link layer address option format. o specified setting of "u" bit in interface id's. o removed obsoleted appendix sections. o re-organized major sections to match normative references. o revised neighbor discovery, address autoconfiguration, security considerations sections. Added new subsections on interface management, decapsulation/filtering, address lifetime expiry. Appendix B. Interface Identifier Construction This section provides an example specification for constructing EUI64 addresses from the Organizationally-Unique Identifier (OUI) owned by the Internet Assigned Numbers Authority (IANA). It can be used to construct both modified EUI-64 format interface identifiers for IPv6 unicast addresses ([ADDR-ARCH], section 2.5.1) and "native" EUI64 addresses for future use: |0 2|2 3|3 3|4 6| |0 3|4 1|2 9|0 3| +------------------------+--------+--------+------------------------+ | OUI ("00-00-5E"+u+g) | TYPE | TSE | TSD | +------------------------+--------+--------+------------------------+ Where the fields are: OUI IANA's OUI: 00-00-5E with "u" and "g" bits (3 octets) TYPE Type field; specifies use of (TSE, TSD) (1 octet) TSE Type-Specific Extension (1 octet) TSD Type-Specific Data (3 octets) And the following interpretations are specified based on TYPE: TYPE (TSE, TSD) Interpretation ---- ------------------------- 0x00-0xFD RESERVED for future IANA use 0xFE (TSE, TSD) together contain an IPv4 address 0xFF TSD is interpreted based on TSE as follows: TSE TSD Interpretation --- ------------------ 0x00-0xFD RESERVED for future IANA use 0xFE TSD contains 24-bit EUI-48 intf id 0xFF RESERVED by IEEE/RAC Using this example specification, if TYPE=0xFE, then TSE is an extension of TSD. If TYPE=0xFF, then TSE is an extension of TYPE. (Other values for TYPE, and other interpretations of TSE, TSD are reserved for future IANA use.) When TYPE='0xFE' the EUI64 address embeds an IPv4 address, encoded in network byte order. For Modified EUI64 format interface identifiers in IPv6 unicast addresses ([ADDR-ARCH], Appendix A) using IANA's OUI, when TYPE=0xFE and the IPv4 address is a globally unique (i.e., provider-assigned) unicast address, the "u" bit is set to 1 to indicate universal scope. When TYPE=0xFE and the IPv4 address is from a private allocation, the "u" bit is set to 0 to indicate local scope. Thus, when the first four octets of the interface identifier in an IPv6 unicast address are either: '02-00-5E-FE' or: '00-00-5E-FE', the next four octets embed an IPv4 address and the interface identifier is said to be in "ISATAP format". 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. The IETF has been notified of intellectual property rights claimed in regard to some or all of the specification contained in this document. For more information consult the online list of claimed rights. Full Copyright Statement Copyright (C) The Internet Society (2003).(2004). 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 HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Acknowledgment Funding for the RFC Editor function is currently provided by the Internet Society.