--- 1/draft-ietf-ngtrans-isatap-07.txt 2006-02-05 00:50:52.000000000 +0100 +++ 2/draft-ietf-ngtrans-isatap-08.txt 2006-02-05 00:50:52.000000000 +0100 @@ -1,22 +1,22 @@ NGTRANS Working Group F. Templin Internet-Draft Nokia -Expires: June 13, 2003 T. Gleeson +Expires: June 19, 2003 T. Gleeson Cisco Systems K.K. M. Talwar D. Thaler Microsoft Corporation - December 13, 2002 + December 19, 2002 Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) - draft-ietf-ngtrans-isatap-07.txt + draft-ietf-ngtrans-isatap-08.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. @@ -25,42 +25,75 @@ 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 June 13, 2003. + This Internet-Draft will expire on June 19, 2003. Copyright Notice Copyright (C) The Internet Society (2002). All Rights Reserved. Abstract This document specifies an Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) that connects IPv6 hosts and routers within IPv4 sites. ISATAP is a transition mechanism that treats the site's IPv4 infrastructure as a Non-Broadcast Multiple Access (NBMA) link layer for IPv6 with no requirement for IPv4 multicast. ISATAP enables intra-site automatic IPv6-in-IPv4 tunneling whether globally assigned or private IPv4 addresses are used. +Table of Contents + + 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 + 2. Applicability Statement . . . . . . . . . . . . . . . . . . 3 + 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 + 4. Transmission of IPv6 Packets on ISATAP Links . . . . . . . . 4 + 4.1 ISATAP Interface Identifier Construction . . . . . . . . . . 4 + 4.2 Stateless Autoconfiguration and Link-Local Addresses . . . . 5 + 4.3 ISATAP Link/Interface Configuration . . . . . . . . . . . . 5 + 4.4 Sending Rules and Address Mapping . . . . . . . . . . . . . 6 + 4.5 Validity Checks for Received Packets . . . . . . . . . . . . 6 + 4.6 Tunnel MTU and Fragmentation . . . . . . . . . . . . . . . . 6 + 5. Neighbor Discovery for ISATAP Links . . . . . . . . . . . . 9 + 5.1 Address Resolution . . . . . . . . . . . . . . . . . . . . . 9 + 5.2 Router and Prefix Discovery . . . . . . . . . . . . . . . . 10 + 5.2.1 Conceptual Data Structures . . . . . . . . . . . . . . . . . 10 + 5.2.2 Validity Checks for Router Advertisements . . . . . . . . . 11 + 5.2.3 Router Specification . . . . . . . . . . . . . . . . . . . . 12 + 5.2.4 Host Specification . . . . . . . . . . . . . . . . . . . . . 12 + 6. ISATAP Deployment Considerations . . . . . . . . . . . . . . 13 + 6.1 Host And Router Deployment Considerations . . . . . . . . . 13 + 6.2 Site Administration Considerations . . . . . . . . . . . . . 13 + 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . 14 + 8. Security considerations . . . . . . . . . . . . . . . . . . 14 + 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 15 + Normative References . . . . . . . . . . . . . . . . . . . . 16 + Informative References . . . . . . . . . . . . . . . . . . . 17 + Authors' Addresses . . . . . . . . . . . . . . . . . . . . . 17 + A. Major Changes . . . . . . . . . . . . . . . . . . . . . . . 18 + B. Rationale for Interface Identifier Construction Rules . . . 21 + C. INTELLECTUAL PROPERTY . . . . . . . . . . . . . . . . . . . 22 + Intellectual Property and Copyright Statements . . . . . . . 23 + 1. Introduction This document presents a simple approach that enables incremental deployment of IPv6 [1] within IPv4-based [2] sites in a manner that is compatible with inter-domain transition mechanisms, e.g., RFC 3056 - (6to4) [17]. We refer to this approach as the Intra-Site Automatic + (6to4) [19]. We refer to this approach as the Intra-Site Automatic Tunnel Addressing Protocol, or ISATAP (pronounced: "ice-a-tap"). ISATAP allows dual-stack nodes that do not share a common link with an IPv6 router to automatically tunnel packets to the IPv6 next-hop address through IPv4, i.e., the site's IPv4 infrastructure is treated as an NBMA link layer. This document specifies details for the transmission of IPv6 packets over ISATAP links (i.e., automatic IPv6-in-IPv4 tunneling), including a new EUI-64 based interface identifier format [3][4][5] that embeds an IPv4 address. This format supports configuration of global, @@ -86,21 +119,21 @@ multicast) o supports both stateless address autoconfiguration and manual configuration o supports networks that use non-globally unique IPv4 addresses (e.g., when private address allocations [8] are used), but does not allow the virtual ISATAP link to span a Network Address Translator [9] - o compatible with other NGTRANS mechanisms (e.g., 6to4 [17]) + o compatible with other NGTRANS mechanisms (e.g., 6to4 [19]) 3. Terminology The terminology of RFC 2460 [1] applies to this document. The following additional terms are defined: link: same definition as [6][7]. underlying link: @@ -220,138 +253,130 @@ List (see Section 5.2.1), i.e., previous hop is an on-link ISATAP router Packets that do not satisfy at least one of the above checks are silently discarded. 4.6 Tunnel MTU and Fragmentation ISATAP interfaces implement automatic tunnels that may be configured over multiple underlying links with diverse MTUs. The ISATAP - interface MTU (ISATAP_MTU) SHOULD be no larger than the largest MTU - of all underlying links (LINK_MTU), minus 20 bytes for IPv4 - encapsulation. - - The minimum value (ISATAP_MINMTU) MUST be at least 1280 bytes [1], - but SHOULD be set to 1380 bytes (see note 1). The maximum value used - for ISATAP_MTU SHOULD be 4140 bytes (see note 2). The maximum - receive unit (ISATAP_MRU) MUST be at least 4400 bytes. + interface MTU (ISATAP_MTU) SHOULD be set to the largest MTU of all + underlying links (LINK_MTU) minus 120 bytes for possible link-layer + encapsulations (see note 1). The minimum value (ISATAP_MINMTU) MUST + be at least 1280 bytes [1], but SHOULD be set to 1380 bytes (see note + 2). - IPv6 path MTU discovery [12] is required for IPv6 interfaces that + IPv6 path MTU discovery [14] is required for IPv6 interfaces that send packets larger than 1280 bytes. The following considerations for ISATAP interfaces are noted: o ISATAP encapsulators and decapsulators are IPv6 neighbors since they share a common link layer, i.e., the ISATAP link o ISATAP neighbors may be separated by multiple IPv4 hops requiring - IPv4 path MTU discovery [13] to establish per-neighbor MTUs + IPv4 path MTU discovery [15] to establish per-neighbor MTUs (NBR_MTU) o NBR_MTU information is stored as link-layer (IPv4) information (e.g., in the IPv4 path MTU discovery cache), thus it may not be visible to upper layers in all implementations o NBR_MTU information may not always be available for each neighbor due to finite storage limitations o IPv4 path MTU discovery delivers ICMPv4 "fragmentation needed" - messages, but these cannot be translated into ICMPv6 "packet too - big" messages. Thus, encapsulated packets MUST be sent with the - DF flag in the IPv4 header NOT set unless additional state is - maintained in the encapsulator (see note 3) - - Traditional packetization and network (IPv6) layer implementations - view ISATAP interfaces as ordinary IPv6 interfaces with a single MTU - (ISATAP_MTU). Such implementations forward only those IPv6 packets - of size ISATAP_MTU or smaller to the ISATAP interface. All other - packets are dropped, and an IPv6 ICMP "packet too big" message with - MTU = ISATAP_MTU is returned. - - Modified packetization and network (IPv6) layer implementations MAY - look into the ISATAP link layer for per-neighbor MTU information. - When available, this information supersedes ISATAP_MTU in determining - whether to forward the packet or return an ICMPv6 "packet too big" - (see above). + messages, but these do not provide enough state for translation + into ICMPv6 "packet too big" messages (see: RFC 792 [12] and RFC + 1812 [13], section 4.3.2.3). - For IPv6 packets forwarded to the ISATAP interface, all - implementations employ the following algorithm at the link layer to - determine when to perform IPv6-in-IPv4 encapsulation and when to - return an IPv6 ICMP "packet too big" message: + IPv6 sees the ISATAP interface as an ordinary IPv6 interface with a + fixed MTU (ISATAP_MTU), i.e., only those IPv6 packets of size + ISATAP_MTU or smaller are accepted. All other packets are dropped, + and an IPv6 ICMP "packet too big" message with MTU = ISATAP_MTU is + returned. ISATAP interfaces SHOULD implement the following + link-layer algorithm to determine when to perform IPv6-in-IPv4 + encapsulation and when to return an ICMPv6 "packet too big" message: Determine per-neighbor LINK_MTU; NBR_MTU, e.g., by consulting IPv4 forwarding table and/or IPv4 path MTU discovery cache, then: if NBR_MTU information exists - if packet is larger than NBR_MTU - 20 and packet + if packet is larger than NBR_MTU - 120 and packet is larger than ISATAP_MINMTU Send IPv6 ICMP "packet too big" with - MTU = MAX(NBR_MTU - 20, ISATAP_MINMTU) + MTU = MAX(NBR_MTU - 120, ISATAP_MINMTU) Drop packet else - Encapsulate but do not set the Don't Fragment + if packet is larger than ISATAP_MINMTU + Encapsulate and set the Don't Fragment flag in the IPv4 header + else + Encapsulate but do not set the Don't + Fragment flag in the IPv4 header + endif + endif else - if packet is larger than LINK_MTU - 20 and packet is + if packet is larger than LINK_MTU - 120 and packet is larger than ISATAP_MINMTU Send IPv6 ICMP "packet too big" with MTU = ISATAP_MINMTU Drop packet else - if IPv6 neighbor is an IPv4 neighbor on the - underlying link, or packet is less than - or == ISATAP_MINMTU + if IPv6 neighbor is also an IPv4 neighbor on + the underlying link, or packet is less than + or == ISATAP_MINMTU Encapsulate but do not set the Don't - Fragment flag in the IPv4 hdr + Fragment flag in the IPv4 header else send ICMPv6 "packet too big" with MTU = ISATAP_MINMTU Drop packet endif endif endif Figure 2 - NOTES: - - 1. Nearly all IPv4 routers can forward 1500 byte packets without - fragmentation. However, sub-IPv4 layer encapsulation (e.g., for - VPNs) may occur on some paths. Commonly-deployed VPNs use an MTU - of 1400 bytes, thus 1380 bytes SHOULD be used as ISATAP_MINMTU. - - 2. TCP adapts to an overestimated MSS by reducing the segment size - based on IPv6 "packet too big" messages ([12], section 5.4), thus - setting ISATAP_MTU to the largest MTU of all underlying links - would optimize performance for asymmetric paths. - - SCTP ([14], section 7.3) and other packetization layers ([12], - section 5.5), perform upper-layer fragmentation based on IPv6 - "packet too big" messages, which may result in unacceptable loss - when the initial MTU estimate is too large. + Encapsulators MAY maintain per-neighbor MTU (NBR_MTU) values by + periodically probing the IPv4 path, e.g., by sending packets larger + than ISATAP_MINMTU with the DF bit set in the IPv4 header. Large + data packets and/or Neighbor Solicitation (NS) packets with padding + bytes added (up to a total length of ISATAP_MTU) may be used for this + purpose. (NS packets are preferred, since successful delivery + results in a positive acknowledgement from the decapsulator.) + When probing, implementations SHOULD maintain state for translating + ICMPv4 "fragmentation needed" messages into ICMPv6 "packet too big" + messages for at least the round-trip time (RTT) between the + encapsulator and decapsulator (see note 3). Implementations SHOULD + repeat the polling process within REACHABLE_TIME ([7], section 10) to + detect link MTU restrictions. - 4140 is the RECOMMENDED maximum value for ISATAP_MTU, since: + NOTES: - * 4140 bytes makes efficient use of common larger-than- ethernet - MTUs in the internet (e.g., FDDI) + 1. ISATAP requires 20 bytes for link-layer (IPv4) encapsulation. + However, sub-IPv4 layer encapsulation (e.g., for VPNs) may occur + on some paths. Commonly-deployed VPNs on Ethernet use an MTU of + 1400 bytes, thus 100 bytes (1500 minums 1400) are reserverd for + sub-IPv4 layer encapsulation. - * Locally-generated ICMPv6 "packet too big" messages are likely - to advertise an MTU of 1380, resulting in at most three - fragments and limiting loss probability + 2. Nearly all IPv4 routers can forward 1500 byte packets without + fragmentation. Thus, 1380 bytes (1500 minus 100 minus 20) is + RECOMMENDED as ISATAP_MINMTU. - 3. Implementations MAY cache recently-sent IPv6 packets to provide - state for translating ICMPv4 "fragmentation needed" messages into - ICMPv6 "packet too big" messages. Such implementations MAY set - the DF flag in the IPv4 header in the above algorithm for packets - that will be retained in the cache at least as long as the - round-trip time (RTT) between the encapsulator and decapsulator. + 3. ICMPv4 "fragmentation needed" messages can be injected by + malicious nodes, but this same problem exists in IPv4. Using + Neighbor Solicitation messages for probing and receiving a + positive acknowledgement from a trusted decapsulator MAY help + encapsulators recognize spoofed ICMPv4 "fragmentation needed" + messages. 5. Neighbor Discovery for ISATAP Links Section 3.2 of RFC 2461 [7] provides the following guidelines for non-broadcast multiple access (NBMA) link support: "Redirect, Neighbor Unreachability Detection and next-hop determination should be implemented as described in this document. Address resolution and the mechanism for delivering Router Solicitations and Advertisements on NBMA links is not specified in @@ -406,25 +430,25 @@ A Potential Router List (PRL) is associated with every ISATAP link. The PRL provides a trust basis for router validation (see security considerations). Each entry in the PRL has an IPv4 address and an associated timer. The IPv4 address represents a router's ISATAP interface (likely to be an "advertising interface"), and is used to construct the ISATAP link-local address for that interface. The following sections specify the process for initializing the PRL: When a node enables an ISATAP link, it first discovers a DNS (RFC - 1035 [20]) fully-qualified domain name for the site's ISATAP service. - The domain name MAY be established by a DHCPv4 [15] option for ISATAP + 1035 [22]) fully-qualified domain name for the site's ISATAP service. + The domain name MAY be established by a DHCPv4 [17] option for ISATAP (option code TBD, see IANA Considerations), by manual configuration, or by an unspecified alternative method. The DHCPv4 option for - ISATAP is implemented exactly as in RFC 3361 [16] with the following + ISATAP is implemented exactly as in RFC 3361 [18] with the following noted exceptions: o the DHCP option code for ISATAP (TBD) is used o the encoding byte MUST be 0, i.e.; only FQDNs are accepted o if multiple domain names occur, only the first is used Next, the node initializes the link's PRL with IPv4 addresses associated with the domain name discovered above. IPv4 addresses are @@ -440,21 +464,21 @@ interfaces (normally as address records in the site's name service). Administrators may also advertise the domain name in a DHCPv4 option for ISATAP. 2. There are no mandatory rules for the selection of a domain name, but administrators are encouraged to use the convention "isatap.domainname" (e.g., isatap.example.com). 3. After initialization, nodes periodically re-initialize the PRL (after ResolveInterval). When DNS is used, nodes MUST follow the - cache invalidation procedures in [20] when the DNS time-to-live + cache invalidation procedures in [22] when the DNS time-to-live expires. 5.2.2 Validity Checks for Router Advertisements A node MUST silently discard any Router Advertisement messages it receives that do not satisfy both the validity checks in ([7], section 6.1.2) and the following additional validity check for ISATAP: o the network-layer (IPv6) source address is an ISATAP address and @@ -574,32 +598,32 @@ o ISATAP nodes periodically refresh the entries on the PRL, typically by querying the DNS. Responsible site administration can reduce the control traffic. At a minimum, administrators SHOULD ensure that the site's address records for ISATAP router interfaces are well maintained. 7. IANA Considerations A DHCPv4 option assignment for ISATAP is requested, as outlined in - the procedures found in RFC 2939 [21], section 3. + the procedures found in RFC 2939 [23], section 3. Appendix B proposes a specification for managing the IEEE OUI assigned to IANA for EUI-64 interface identifier construction. This specification is made freely available to IANA for any purpose they may find useful. 8. Security considerations Site administrators are advised that, in addition to possible attacks against IPv6, security attacks against IPv4 MUST also be considered. - Many security considerations in RFC 2529 [18], section 9 apply also + Many security considerations in RFC 2529 [20], section 9 apply also to ISATAP. Responsible IPv4 site security management is strongly encouraged. In particular, border gateways SHOULD implement filtering to detect spoofed IPv4 source addresses at a minimum; ip-protocol-41 filtering SHOULD also be implemented. If IPv4 source address filtering is not correctly implemented, the ISATAP validity checks will not be effective in preventing IPv6 source address spoofing. @@ -613,21 +637,21 @@ Advertisements they receive from on-link routers, as indicated by a value of 255 in the IPv6 'hop-limit' field. Since this field is not decremented when ip-protocol-41 packets traverse multiple IPv4 hops ([10], section 3), ISATAP links require a different trust model. In particular, ONLY those Router Advertisements received from a member of the Potential Routers List are trusted; all others are silently discarded. This trust model is predicated on IPv4 source address filtering, as described above. The ISATAP address format does not support privacy extensions for - stateless address autoconfiguration [19]. However, since the ISATAP + stateless address autoconfiguration [21]. However, since the ISATAP interface identifier is derived from the node's IPv4 address, ISATAP addresses do not have the same level of privacy concerns as IPv6 addresses that use an interface identifier derived from the MAC address. (This issue is the same for NAT'd addresses.) 9. Acknowledgements Some of the ideas presented in this draft were derived from work at SRI with internal funds and contractual support. Government sponsors who supported the work include Monica Farah-Stapleton and Russell @@ -639,27 +663,26 @@ The following peer reviewers are acknowledged for taking the time to review a pre-release of this document and provide input: Jim Bound, Rich Draves, Cyndi Jung, Ambatipudi Sastry, Aaron Schrader, Ole Troan, Vlad Yasevich. The authors acknowledge members of the NGTRANS community who have made significant contributions to this effort, including Rich Draves, Alain Durand, Nathan Lutchansky, Karen Nielsen, Art Shelest, Margaret Wasserman, and Brian Zill. - The authors also wish to acknowledge the work of Quang Nguyen [22] + The authors also wish to acknowledge the work of Quang Nguyen [24] 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. Normative References - [1] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, December 1998. [2] Postel, J., "Internet Protocol", STD 5, RFC 791, September 1981. [3] Hinden, R. and S. Deering, "IP Version 6 Addressing Architecture", RFC 2373, July 1998. [4] Hinden, R. and S. Deering, "An IPv6 Aggregatable Global Unicast @@ -681,56 +704,62 @@ [9] Egevang, K. and P. Francis, "The IP Network Address Translator (NAT)", RFC 1631, May 1994. [10] Gilligan, R. and E. Nordmark, "Transition Mechanisms for IPv6 Hosts and Routers", RFC 2893, August 2000. [11] Conta, A. and S. Deering, "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification", RFC 2463, December 1998. - [12] McCann, J., Deering, S. and J. Mogul, "Path MTU Discovery for + [12] Postel, J., "Internet Control Message Protocol", STD 5, RFC + 792, September 1981. + + [13] Baker, F., "Requirements for IP Version 4 Routers", RFC 1812, + June 1995. + + [14] McCann, J., Deering, S. and J. Mogul, "Path MTU Discovery for IP version 6", RFC 1981, August 1996. - [13] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, + [15] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, November 1990. - [14] Stewart, R., Xie, Q., Morneault, K., Sharp, C., Schwarzbauer, + [16] Stewart, R., Xie, Q., Morneault, K., Sharp, C., Schwarzbauer, H., Taylor, T., Rytina, I., Kalla, M., Zhang, L. and V. Paxson, "Stream Control Transmission Protocol", RFC 2960, October 2000. - [15] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131, + [17] Droms, R., "Dynamic Host Configuration Protocol", RFC 2131, March 1997. - [16] Schulzrinne, H., "Dynamic Host Configuration Protocol + [18] Schulzrinne, H., "Dynamic Host Configuration Protocol (DHCP-for-IPv4) Option for Session Initiation Protocol (SIP) Servers", RFC 3361, August 2002. Informative References - [17] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via + [19] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via IPv4 Clouds", RFC 3056, February 2001. - [18] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 + [20] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 Domains without Explicit Tunnels", RFC 2529, March 1999. - [19] Narten, T. and R. Draves, "Privacy Extensions for Stateless + [21] Narten, T. and R. Draves, "Privacy Extensions for Stateless Address Autoconfiguration in IPv6", RFC 3041, January 2001. - [20] Mockapetris, P., "Domain names - implementation and + [22] Mockapetris, P., "Domain names - implementation and specification", STD 13, RFC 1035, November 1987. - [21] Droms, R., "Procedures and IANA Guidelines for Definition of + [23] Droms, R., "Procedures and IANA Guidelines for Definition of New DHCP Options and Message Types", BCP 43, RFC 2939, September 2000. - [22] Nguyen, Q., "http://irl.cs.ucla.edu/vet/report.ps", spring + [24] 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 @@ -758,20 +787,24 @@ Microsoft Corporation One Microsoft Way Redmond, WA 98052-6399 US Phone: +1 425 703 8835 EMail: dthaler@microsoft.com Appendix A. Major Changes + changes from version 07 to version 08: + + o updated MTU section + changes from version 06 to version 07: o clarified address resolution, Neighbor Unreachability Detection o specified MTU/MRU requirements changes from version 05 to version 06: o Addressed operational issues identified in 05 based on discussion between co-authors