idnits 2.17.1 draft-ietf-behave-v6v4-xlate-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The abstract seems to contain references ([I-D.ietf-behave-v6v4-framework], [RFC0791], [RFC0792], [I-D.ietf-behave-address-format], [RFC2460], [RFC2765], [RFC4443]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. -- The draft header indicates that this document obsoletes RFC2765, but the abstract doesn't seem to directly say this. It does mention RFC2765 though, so this could be OK. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (October 19, 2009) is 5297 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'I-D.xli-behave-v4v6-prefix' is defined on line 975, but no explicit reference was found in the text == Unused Reference: 'RFC2119' is defined on line 992, but no explicit reference was found in the text == Unused Reference: 'RFC4301' is defined on line 1077, but no explicit reference was found in the text == Unused Reference: 'RFC4303' is defined on line 1083, but no explicit reference was found in the text ** Obsolete normative reference: RFC 793 (Obsoleted by RFC 9293) ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 2765 (Obsoleted by RFC 6145) == Outdated reference: A later version (-10) exists of draft-ietf-behave-address-format-00 == Outdated reference: A later version (-10) exists of draft-ietf-behave-v6v4-framework-01 == Outdated reference: A later version (-12) exists of draft-ietf-behave-v6v4-xlate-stateful-02 == Outdated reference: A later version (-03) exists of draft-venaas-behave-v4v6mc-framework-00 -- Obsolete informational reference (is this intentional?): RFC 3171 (Obsoleted by RFC 5771) Summary: 5 errors (**), 0 flaws (~~), 10 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 behave X. Li 3 Internet-Draft C. Bao 4 Obsoletes: 2765 (if approved) CERNET Center/Tsinghua University 5 Intended status: Standards Track F. Baker 6 Expires: April 22, 2010 Cisco Systems 7 October 19, 2009 9 IP/ICMP Translation Algorithm 10 draft-ietf-behave-v6v4-xlate-02 12 Status of this Memo 14 This Internet-Draft is submitted to IETF in full conformance with the 15 provisions of BCP 78 and BCP 79. This document may contain material 16 from IETF Documents or IETF Contributions published or made publicly 17 available before November 10, 2008. The person(s) controlling the 18 copyright in some of this material may not have granted the IETF 19 Trust the right to allow modifications of such material outside the 20 IETF Standards Process. Without obtaining an adequate license from 21 the person(s) controlling the copyright in such materials, this 22 document may not be modified outside the IETF Standards Process, and 23 derivative works of it may not be created outside the IETF Standards 24 Process, except to format it for publication as an RFC or to 25 translate it into languages other than English. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF), its areas, and its working groups. Note that 29 other groups may also distribute working documents as Internet- 30 Drafts. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 The list of current Internet-Drafts can be accessed at 38 http://www.ietf.org/ietf/1id-abstracts.txt. 40 The list of Internet-Draft Shadow Directories can be accessed at 41 http://www.ietf.org/shadow.html. 43 This Internet-Draft will expire on April 22, 2010. 45 Copyright Notice 47 Copyright (c) 2009 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents in effect on the date of 52 publication of this document (http://trustee.ietf.org/license-info). 53 Please review these documents carefully, as they describe your rights 54 and restrictions with respect to this document. 56 Abstract 58 This document specifies an update to the Stateless IP/ICMP 59 Translation Algorithm (SIIT) described in RFC 2765. The algorithm 60 translates between IPv4 and IPv6 packet headers (including ICMP 61 headers). 63 This specification addresses both a stateless and a stateful mode. 64 In the stateless mode, translation information is carried in the 65 address itself, permitting both IPv4->IPv6 and IPv6->IPv4 session 66 establishment without maintaining state in the IP/ICMP translator. 67 In the stateful mode, translation state is maintained between IPv4 68 address/transport port tuples and IPv6 address/transport port tuples, 69 enabling IPv6 systems to open sessions with IPv4 systems. The choice 70 of operational mode is made by the operator deploying the network and 71 is critical to the operation of the applications using it. 73 Significant issues exist in the stateless and stateful modes that are 74 not addressed in this document, related to the address assignment and 75 the maintenance of the translation tables, respectively. This 76 document confines itself to the actual translation. 78 Acknowledgement of previous work 80 This document is a product of the 2008-2009 effort to define a 81 replacement for NAT-PT. It is an update to and directly derivative 82 from Erik Nordmark's [RFC2765], which similarly provides both 83 stateless and stateful translation between IPv4 [RFC0791] and IPv6 84 [RFC2460], and between ICMPv4 [RFC0792] and ICMPv6 [RFC4443]. The 85 original document was a product of the NGTRANS working group. 87 The changes in this document reflect five components: 89 1. Redescribing the network model to map to present and projected 90 usage [I-D.ietf-behave-v6v4-framework]. 92 2. Moving the address format to the address format document 93 [I-D.ietf-behave-address-format], to coordinate with other drafts 94 on the topic. 96 3. Describing both stateful and stateless operation. 98 4. Some changes in ICMP. 100 5. Updating references. 102 Requirements 104 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 105 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 106 document are to be interpreted as described in RFC 2119. 108 Table of Contents 110 1. Introduction and Motivation . . . . . . . . . . . . . . . . . 4 111 1.1. Translation Model . . . . . . . . . . . . . . . . . . . . 4 112 1.2. Applicability and Limitations . . . . . . . . . . . . . . 5 113 1.3. Stateless vs. Stateful Mode . . . . . . . . . . . . . . . 6 114 1.4. Path MTU discovery and fragmentation . . . . . . . . . . . 6 115 2. Translating from IPv4 to IPv6 . . . . . . . . . . . . . . . . 7 116 2.1. Translating IPv4 Headers into IPv6 Headers . . . . . . . . 8 117 2.2. Translating UDP over IPv4 . . . . . . . . . . . . . . . . 10 118 2.3. Translating ICMPv4 Headers into ICMPv6 Headers . . . . . . 10 119 2.4. Translating ICMPv4 Error Messages into ICMPv6 . . . . . . 13 120 2.5. Translator sending ICMP error message . . . . . . . . . . 13 121 2.6. Transport-layer Header Translation . . . . . . . . . . . . 14 122 2.7. Knowing when to Translate . . . . . . . . . . . . . . . . 14 123 3. Translating from IPv6 to IPv4 . . . . . . . . . . . . . . . . 14 124 3.1. Translating IPv6 Headers into IPv4 Headers . . . . . . . . 16 125 3.2. Translating ICMPv6 Headers into ICMPv4 Headers . . . . . . 18 126 3.3. Translating ICMPv6 Error Messages into ICMPv4 . . . . . . 19 127 3.4. Translator sending ICMPv6 error message . . . . . . . . . 20 128 3.5. Transport-layer Header Translation . . . . . . . . . . . . 20 129 3.6. Knowing when to Translate . . . . . . . . . . . . . . . . 20 130 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 21 131 5. Security Considerations . . . . . . . . . . . . . . . . . . . 21 132 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 21 133 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 22 134 7.1. Normative References . . . . . . . . . . . . . . . . . . . 22 135 7.2. Informative References . . . . . . . . . . . . . . . . . . 23 136 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 24 138 1. Introduction and Motivation 140 An understanding of the framework presented in 141 [I-D.ietf-behave-v6v4-framework] is presumed in this document. 143 The transition mechanisms specified in [RFC4213] handle the case of 144 dual IPv4/IPv6 hosts interoperating with both dual IPv4/IPv6 hosts 145 and IPv4-only hosts, which is needed early in the transition to IPv6. 146 The dual IPv4/IPv6 hosts are assigned both an IPv4 and one or more 147 IPv6 addresses. The number of available globally unique IPv4 148 addresses is becoming smaller and smaller as the Internet grows; we 149 expect that there will be a desire to take advantage of the large 150 IPv6 address space and not require that every new Internet node have 151 a permanently assigned IPv4 address. 153 SIIT [RFC2765] is designed for the case of small networks (e.g., a 154 single subnet) and for a site that has IPv6-only hosts in a dual 155 IPv4/IPv6 network. This use assumes a mechanism for IPv6 nodes to 156 acquire a temporary address from the pool of IPv4 addresses. 157 However, SIIT is not useful in the case when the IPv6 nodes need to 158 acquire temporary IPv4 addresses from a "distant" SIIT box operated 159 by a different administration, or require that the IPv6 Internet 160 contain routes for IPv6-mapped addresses (The latter is known to be a 161 very bad idea due to the size of the IPv4 routing table that would 162 potentially be injected into IPv6 routing in the form of IPv4-mapped 163 addresses.) 165 In addition, due to the IPv4 address deletion problem, it is 166 desirable that a single IPv4 address needs to be shared via transport 167 port multiplexing for different IPv6 nodes when they communicate with 168 other IPv4 hosts. 170 Furthermore, in SIIT [RFC2765], an IPv6-only node that works through 171 SIIT translators needs some modifications beyond a normal IPv6-only 172 node. These modifications are not strictly implied in this document, 173 since normal IPv6 addresses can be used in the IPv6 end nodes. 175 A detailed discussion of translation scenarios is presented in 176 [I-D.ietf-behave-v6v4-framework], while the technical specification 177 of the translation algorithm itself is covered in this document. 179 1.1. Translation Model 181 This document specifies the translation algorithm that is one of the 182 components described in [I-D.ietf-behave-v6v4-framework] needed to 183 make IPv6-only nodes interoperate with IPv4-only nodes as shown in 184 Figure 1. 186 -------- -------- 187 // IPv4 \\ // IPv6 \\ 188 / Domain \ / Domain \ 189 / +----+ +--+ \ 190 | |XLAT| |S2| | Sn: Servers 191 | +--+ +----+ +--+ | Hn: Clients 192 | |S1| +----+ | 193 | +--+ |DNS | +--+ | XLAT: V4/V6 Translator 194 \ +--+ +----+ |H2| / DNS: DNS Server 195 \ |H1| / \ +--+ / 196 \\ +--+ // \\ // 197 -------- -------- 199 Figure 1: Translation Model 201 The translation model consists of two or more network domains 202 connected by one or more IP/ICMP translators. One of those networks 203 either routes IPv4 but not IPv6, or contains some hosts that only 204 implement IPv4 or have IPv4 only applications (even if the host and 205 the network support IPv6). The other network either routes IPv6 but 206 not IPv4, or contains some hosts that only implement IPv6 or has IPv6 207 only applications. Both networks contain clients, servers, and 208 peers. 210 1.2. Applicability and Limitations 212 The use of this translation algorithm assumes that the IPv6 network 213 is somehow well-connected i.e., when an IPv6 node wants to 214 communicate with another IPv6 node there is an IPv6 path between 215 them. Various tunneling schemes exist that can provide such a path, 216 but those mechanisms and their use is outside the scope of this 217 document and [RFC2765]. 219 The translation algorithm can be used not only in a subnet, but can 220 also be used in service provider's backbone network. 222 The translating function specified in this document does not 223 translate any IPv4 options and it does not translate IPv6 routing 224 headers, hop-by-hop extension headers, destination options headers or 225 source routing headers [RFC2765]. 227 The issues and algorithms in the translation of datagram containing 228 TCP segments are described in [RFC5382]. The considerations of that 229 document are applicable in this case as well. 231 Fragmented IPv4 UDP packets that do not contain a UDP checksum (i.e. 232 the UDP checksum field is zero) are not of significant use over wide- 233 areas in the Internet and will not be translated by the IP/ICMP 234 translator [Miller]. 236 IPv4 multicast addresses [RFC3171] cannot be mapped to IPv6 multicast 237 addresses [RFC3307] based on the unicast mapping rule. However, a 238 special rule for address translation can be created for the multicast 239 packet translation algorithm; if that is done, the IP/ICMP header 240 translation aspect of this memo works. 242 1.3. Stateless vs. Stateful Mode 244 The IP/ICMP translator has two possible modes of operation: stateless 245 and stateful. In both cases, we assume that a system that has an 246 IPv4 address but not an IPv6 address is communicating with a system 247 that has an IPv6 address but no IPv4 address, or that the two systems 248 do not have contiguous routing connectivity and hence are forced to 249 have their communications translated. 251 In the stateless mode, a specific IPv6 address range will represent 252 IPv4 systems, and the IPv6 systems have addresses that can be 253 algorithmatically mapped to a subset of the service provider's IPv4 254 addresses. In this mode, there is no need to concern oneself with 255 port translation or translation tables, as the IPv4 and IPv6 256 counterparts are algorithmically related. 258 In the stateful mode, a specific IPv6 address range will represent 259 IPv4 systems, but the IPv6 systems may use any [RFC4291] addresses 260 except in that range. In this case, a translation table is required 261 to bind the IPv6 systems' addresses to the IPv4 addresses maintained 262 in the translator. 264 The address translation mechanisms for the stateless and the stateful 265 translations are defined in [I-D.ietf-behave-address-format]. 267 1.4. Path MTU discovery and fragmentation 269 Due to the different sizes of the IPv4 and IPv6 header, which are 20+ 270 octal and 40+ octal respectively, handling the maximum packet size is 271 critical for the operation of the IPv4/IPv6 translator. There are 272 three mechanisms to handle this issue: the path MTU discovery, 273 fragmentation and transport layer negotiation such as the TCP MSS 274 option. 276 The methods for the Path MTU discovery and fragmentation handling are 277 discussed in "Translating from IPv4 to IPv6" and "Translating from 278 IPv6 to IPv4" sections of this document. The resembling of the 279 fragmented packets in the stateful translator is discussed in 280 [I-D.ietf-behave-v6v4-xlate-stateful], since it requires the state 281 maintenance in the translator. 283 2. Translating from IPv4 to IPv6 285 When an IP/ICMP translator receives an IPv4 datagram addressed to a 286 destination towards the IPv6 domain, it translates the IPv4 header of 287 that packet into an IPv6 header. Since the ICMP [RFC0792][RFC4443], 288 TCP [RFC0793] and UDP [RFC0768] headers contain checksums that cover 289 IP header information, if the address mapping algorithm is not 290 checksum-neutral, the ICMP and transport-layer headers MUST be 291 updated. The data portion of the packet is left unchanged. The IP/ 292 ICMP translator then forwards the packet based on the IPv6 293 destination address. The original IPv4 header on the packet is 294 removed and replaced by an IPv6 header. 296 +-------------+ +-------------+ 297 | IPv4 | | IPv6 | 298 | Header | | Header | 299 +-------------+ +-------------+ 300 | Transport | | Fragment | 301 | Layer | ===> | Header | 302 | Header | |(not always) | 303 +-------------+ +-------------+ 304 | | | Transport | 305 ~ Data ~ | Layer | 306 | | | Header | 307 +-------------+ +-------------+ 308 | | 309 ~ Data ~ 310 | | 311 +-------------+ 313 Figure 2: IPv4-to-IPv6 Translation 315 One of the differences between IPv4 and IPv6, is that in IPv6 path 316 MTU discovery is mandatory but it is optional in IPv4. This implies 317 that IPv6 routers will never fragment a packet - only the sender can 318 do fragmentation. 320 When the IPv4 node performs path MTU discovery (by setting the DF bit 321 in the header) the path MTU discovery can operate end-to-end, i.e., 322 across the translator. In this case either IPv4 or IPv6 routers 323 might send back ICMP "packet too big" messages to the sender. When 324 the IPv6 routers send these ICMP errors they will pass through a 325 translator that will translate the ICMP error to a form that the IPv4 326 sender can understand. In this case, an IPv6 fragment header is only 327 included if the IPv4 packet is already fragmented. 329 However, when the IPv4 sender does not set the DF bit, the translator 330 has to ensure that the packet does not exceed the path MTU on the 331 IPv6 side. This is done by fragmenting the IPv4 packet so that it 332 fits in 1280 byte IPv6 packets, since that is the minimum IPv6 packet 333 size. Also, when the IPv4 sender does not set the DF bit the 334 translator MUST always include an IPv6 fragment header to indicate 335 that the sender allows fragmentation. That is needed should the 336 packet pass through an IP/ICMP translator. 338 The above rules ensure that when packets are fragmented, either by 339 the sender or by IPv4 routers, the low-order 16 bits of the fragment 340 identification are carried end-to-end, ensuring that packets are 341 correctly reassembled. In addition, the rules use the presence of an 342 IPv6 fragment header to indicate that the sender might not be using 343 path MTU discovery, i.e., the packet should not have the DF flag set 344 should it later be translated back to IPv4. 346 Other than the special rules for handling fragments and path MTU 347 discovery, the actual translation of the packet header consists of a 348 simple mapping as defined below. Note that ICMP packets require 349 special handling in order to translate the content of ICMP error 350 message and also to add the ICMP pseudo-header checksum. 352 2.1. Translating IPv4 Headers into IPv6 Headers 354 If the DF flag is not set and the IPv4 packet will result in an IPv6 355 packet larger than 1280 bytes the IPv4 packet MUST be fragmented 356 prior to translating it. Since IPv4 packets with DF not set will 357 always result in a fragment header being added to the packet the IPv4 358 packets must be fragmented so that their length, excluding the IPv4 359 header, is at most 1232 bytes (1280 minus 40 for the IPv6 header and 360 8 for the Fragment header). The resulting fragments are then 361 translated independently using the logic described below. 363 If the DF bit is set and the packet is not a fragment (i.e., the MF 364 flag is not set and the Fragment Offset is zero) then the translator 365 SHOULD NOT add a fragment header to the packet. The IPv6 header 366 fields are set as follows: 368 Version: 6 370 Traffic Class: By default, copied from IP Type Of Service octet. 371 According to [RFC2474] the semantics of the bits are identical in 372 IPv4 and IPv6. However, in some IPv4 environments these fields 373 might be used with the old semantics of "Type Of Service and 374 Precedence". An implementation of a translator SHOULD provide the 375 ability to ignore the IPv4 "TOS" and always set the IPv6 traffic 376 class to zero. In addition, if the translator is at an 377 administrative boundary, the filtering and update considerations 378 of [RFC2475] may be applicable. 380 Flow Label: 0 (all zero bits) 382 Payload Length: Total length value from IPv4 header, minus the size 383 of the IPv4 header and IPv4 options, if present. 385 Next Header: Protocol field copied from IPv4 header 387 Hop Limit: TTL value copied from IPv4 header. Since the translator 388 is a router, as part of forwarding the packet it needs to 389 decrement either the IPv4 TTL (before the translation) or the IPv6 390 Hop Limit (after the translation). As part of decrementing the 391 TTL or Hop Limit the translator (as any router) needs to check for 392 zero and send the ICMPv4 "ttl exceeded" or ICMPv6 "hop limit 393 exceeded" error. 395 Source Address: The IPv6 source address is derived from the IPv4 396 source address. Note that the IPv6 source address is the IPv4- 397 mapped address. 399 Destination Address: In stateless mode, which is to say that if the 400 IPv4 destination address is within the range of the IPv4 stateless 401 translation prefix, the IPv6 destination address is derived from 402 the IPv4 destination address. Note that the IPv6 destination 403 address is the IPv4-translatable address. 405 In stateful mode, which is to say that if the IPv4 destination 406 address is not within the range of the IPv4 stateless translation 407 prefix, the IPv4-related IPv6 address and corresponding transport- 408 layer destination port are derived from the database reflecting 409 current session state in the translator. Database maintenance is 410 as described in [I-D.ietf-behave-v6v4-xlate-stateful]. 412 If the IPv4 destination address is in the multicast range, the 413 multicast address mapping method should be applied. 415 If IPv4 options are present in the IPv4 packet, they are ignored 416 i.e., there is no attempt to translate them. However, if an 417 unexpired source route option is present then the packet MUST instead 418 be discarded, and an ICMPv4 "destination unreachable/source route 419 failed" (Type 3/Code 5) error message SHOULD be returned to the 420 sender. 422 If there is a need to add a fragment header (the DF bit is not set or 423 the packet is a fragment) the header fields are set as above with the 424 following exceptions: 426 IPv6 fields: 428 Payload Length: Total length value from IPv4 header, plus 8 for 429 the fragment header, minus the size of the IPv4 header and IPv4 430 options, if present. 432 Next Header: Fragment Header (44). 434 Fragment header fields: 436 Next Header: Protocol field copied from IPv4 header. 438 Fragment Offset: Fragment Offset copied from the IPv4 header. 440 M flag More Fragments bit copied from the IPv4 header. 442 Identification The low-order 16 bits copied from the 443 Identification field in the IPv4 header. The high-order 16 444 bits set to zero. 446 2.2. Translating UDP over IPv4 448 When a translator receives an unfragmented UDP IPv4 packet and the 449 checksum field is zero, the translator SHOULD compute the missing UDP 450 checksum as part of translating the packet. Also, the translator 451 SHOULD maintain a counter of how many UDP checksums are generated in 452 this manner. 454 When a stateless translator receives the first fragment of a 455 fragmented UDP IPv4 packet and the checksum field is zero, the 456 translator SHOULD drop the packet and generate a system management 457 event specifying at least the IP addresses and port numbers in the 458 packet. When it receives fragments other than the first it SHOULD 459 silently drop the packet, since there is no port information to log. 461 When a stateful translator receives fragmented UDP IPv4 packets and 462 the checksum field is zero, if the translator has enough resource to 463 reassemble the packets, the stateful translator SHOULD reassemble the 464 packets and SHOULD calculate the checksum. Otherwise, the stateful 465 translator MAY drop the packets. 467 2.3. Translating ICMPv4 Headers into ICMPv6 Headers 469 All ICMP messages that are to be translated require that the ICMP 470 checksum field be updated as part of the translation since ICMPv6, 471 unlike ICMPv4, has a pseudo-header checksum just like UDP and TCP. 473 In addition, all ICMP packets need to have the Type value translated 474 and, for ICMP error messages, the included IP header also needs 475 translation. 477 The actions needed to translate various ICMPv4 messages are: 479 ICMPv4 query messages: 481 Echo and Echo Reply (Type 8 and Type 0) Adjust the type to 128 482 and 129, respectively, and adjust the ICMP checksum both to 483 take the type change into account and to include the ICMPv6 484 pseudo-header. 486 Information Request/Reply (Type 15 and Type 16) Obsoleted in 487 ICMPv6. Silently drop. 489 Timestamp and Timestamp Reply (Type 13 and Type 14) Obsoleted in 490 ICMPv6. Silently drop. 492 Address Mask Request/Reply (Type 17 and Type 18) Obsoleted in 493 ICMPv6. Silently drop. 495 ICMP Router Advertisement (Type 9) Single hop message. Silently 496 drop. 498 ICMP Router Solicitation (Type 10) Single hop message. Silently 499 drop. 501 Unknown ICMPv4 types Silently drop. 503 IGMP messages: While the MLD messages [RFC2710][RFC3590][RFC3810] 504 are the logical IPv6 counterparts for the IPv4 IGMP messages 505 all the "normal" IGMP messages are single-hop messages and 506 should be silently dropped by the translator 507 [I-D.venaas-behave-v4v6mc-framework]. Other IGMP messages 508 might be used by multicast routing protocols and, since it 509 would be a configuration error to try to have router 510 adjacencies across IP/ICMP translators those packets should 511 also be silently dropped. 513 ICMPv4 error messages: 515 Destination Unreachable (Type 3) For all codes that are not 516 explicitly listed below, set the Type to 1. 518 Translate the code field as follows: 520 Code 0, 1 (net, host unreachable): Set Code to 0 (no route 521 to destination). 523 Code 2 (protocol unreachable): Translate to an ICMPv6 524 Parameter Problem (Type 4, Code 1) and make the Pointer 525 point to the IPv6 Next Header field. 527 Code 3 (port unreachable): Set Code to 4 (port 528 unreachable). 530 Code 4 (fragmentation needed and DF set): Translate to an 531 ICMPv6 Packet Too Big message (Type 2) with code 0. The 532 MTU field needs to be adjusted for the difference between 533 the IPv4 and IPv6 header sizes. Note that if the IPv4 534 router did not set the MTU field, i.e., the router does 535 not implement [RFC1191], then the translator must use the 536 plateau values specified in [RFC1191] to determine a 537 likely path MTU and include that path MTU in the ICMPv6 538 packet. (Use the greatest plateau value that is less 539 than the returned Total Length field.) 541 Code 5 (source route failed): Set Code to 0 (no route to 542 destination). Note that this error is unlikely since 543 source routes are not translated. 545 Code 6,7: Set Code to 0 (no route to destination). 547 Code 8: Set Code to 0 (no route to destination). 549 Code 9, 10 (communication with destination host 550 administratively prohibited): Set Code to 1 (communication 551 with destination administratively prohibited) 553 Code 11, 12: Set Code to 0 (no route to destination). 555 Redirect (Type 5) Single hop message. Silently drop. 557 Source Quench (Type 4) Obsoleted in ICMPv6. Silently drop. 559 Time Exceeded (Type 11) Set the Type field to 3. The Code 560 field is unchanged. 562 Parameter Problem (Type 12) Set the Type field to 4. The 563 Pointer needs to be updated to point to the corresponding 564 field in the translated include IP header. 566 ICMP Error Payload The [RFC4884] length field should be 567 updated to reflect the changed length of the datagram. 568 There are two cases for the length field modifications. 569 That the translated packet is created from scratch and the 570 length field never is filled in. Then an ICMP extension 571 will result in that it will be treated as part of the 572 original datagram field. If the IP payload is copied and 573 then modified then the length field will be unmodified while 574 the original datagram field will become longer by the 575 address translation from v4->v6. Thus cutting off the end 576 of the original datagram field for ICMP extension aware 577 receivers. information. 579 2.4. Translating ICMPv4 Error Messages into ICMPv6 581 There are some differences between the IPv4 and the IPv6 ICMP error 582 message formats as detailed above. In addition, the ICMP error 583 messages contain the IP header for the packet in error, which needs 584 to be translated just like a normal IP header. The translation of 585 this "packet in error" is likely to change the length of the 586 datagram. Thus the Payload Length field in the outer IPv6 header 587 might need to be updated. 589 +-------------+ +-------------+ 590 | IPv4 | | IPv6 | 591 | Header | | Header | 592 +-------------+ +-------------+ 593 | ICMPv4 | | ICMPv6 | 594 | Header | | Header | 595 +-------------+ +-------------+ 596 | IPv4 | ===> | IPv6 | 597 | Header | | Header | 598 +-------------+ +-------------+ 599 | Partial | | Partial | 600 | Transport | | Transport | 601 | Layer | | Layer | 602 | Header | | Header | 603 +-------------+ +-------------+ 605 Figure 3: IPv4-to-IPv6 ICMP Error Translation 607 The translation of the inner IP header can be done by recursively 608 invoking the function that translated the outer IP headers. 610 2.5. Translator sending ICMP error message 612 If the packet is discarded, then the translator SHOULD be able to 613 send back an ICMP message to the original sender of the packet, 614 unless the discarded packet is itself an ICMP message. The ICMP 615 message, if sent, has a type of 3 (Destination Unreachable) and a 616 code of 13 (Communication Administratively Prohibited). The 617 translator device MUST allow to configure whether the ICMP error 618 messages are sent, rate-limited or not sent. 620 2.6. Transport-layer Header Translation 622 If the address translation algorithm is not checksum neutral, the 623 recalculation and updating of the transport-layer headers MUST be 624 performed. UDP/IPv4 datagrams with a checksum of zero MAY be dropped 625 and MAY have their checksum calculated for injection into the IPv6 626 domain. This choice SHOULD be under configuration control. 628 2.7. Knowing when to Translate 630 If the IP/ICMP translator is implemented in a router providing both 631 translation and normal forwarding, and the address is reachable by a 632 more specific route without translation, the router MUST forward it 633 without translating it. Otherwise, when an IP/ICMP translator 634 receives an IPv4 datagram addressed to a destination towards the IPv6 635 domain, the packet will be translated to IPv6. 637 3. Translating from IPv6 to IPv4 639 When an IP/ICMP translator receives an IPv6 datagram addressed to a 640 destination towards the IPv4 domain, it translates the IPv6 header of 641 that packet into an IPv4 header. Since the ICMP [RFC0792][RFC4443], 642 TCP [RFC0793] and UDP [RFC0768] headers contain checksums that cover 643 the IP header, if the address mapping algorithm is not checksum- 644 neutral, the ICMP and transport-layer headers MUST be updated. The 645 data portion of the packet is left unchanged. The IP/ICMP translator 646 then forwards the packet based on the IPv4 destination address. The 647 original IPv6 header on the packet is removed and replaced by an IPv4 648 header. 650 +-------------+ +-------------+ 651 | IPv6 | | IPv4 | 652 | Header | | Header | 653 +-------------+ +-------------+ 654 | Fragment | | Transport | 655 | Header | ===> | Layer | 656 |(if present) | | Header | 657 +-------------+ +-------------+ 658 | Transport | | | 659 | Layer | ~ Data ~ 660 | Header | | | 661 +-------------+ +-------------+ 662 | | 663 ~ Data ~ 664 | | 665 +-------------+ 667 Figure 4: IPv6-to-IPv4 Translation 669 There are some differences between IPv6 and IPv4 in the area of 670 fragmentation and the minimum link MTU that affect the translation. 671 An IPv6 link has to have an MTU of 1280 bytes or greater. The 672 corresponding limit for IPv4 is 68 bytes. Thus, unless there were 673 special measures, it would not be possible to do end-to-end path MTU 674 discovery when the path includes a translator since the IPv6 node 675 might receive ICMP "packet too big" messages originated by an IPv4 676 router that report an MTU less than 1280. However, [RFC2460] section 677 5 requires that IPv6 nodes handle such an ICMP "packet too big" 678 message by reducing the path MTU to 1280 and including an IPv6 679 fragment header with each packet. This allows end-to-end path MTU 680 discovery across the translator as long as the path MTU is 1280 bytes 681 or greater. When the path MTU drops below the 1280 limit the IPv6 682 sender will originate 1280-byte packets that will be fragmented by 683 IPv4 routers along the path after being translated to IPv4. 685 The only drawback with this scheme is that it is not possible to use 686 PMTU to do optimal UDP fragmentation (as opposed to completely 687 avoiding fragmentation) at the sender, since the presence of an IPv6 688 fragment header is interpreted that it is okay to fragment the packet 689 on the IPv4 side. Thus if a UDP application wants to send large 690 packets independent of the PMTU, the sender will only be able to 691 determine the path MTU on the IPv6 side of the translator. If the 692 path MTU on the IPv4 side of the translator is smaller, then the IPv6 693 sender will not receive any ICMP "too big" errors and cannot adjust 694 the size fragments it is sending. 696 Other than the special rules for handling fragments and path MTU 697 discovery the actual translation of the packet header consists of a 698 simple mapping as defined below. Note that ICMP packets require 699 special handling in order to translate the contents of ICMP error 700 message and also to add the ICMP pseudo-header checksum. 702 3.1. Translating IPv6 Headers into IPv4 Headers 704 If there is no IPv6 Fragment header, the IPv4 header fields are set 705 as follows: 707 Version: 4 709 Internet Header Length: 5 (no IPv4 options) 711 Type of Service (TOS) Octet: By default, copied from the IPv6 712 Traffic Class (all 8 bits). According to [RFC2474] the semantics 713 of the bits are identical in IPv4 and IPv6. However, in some IPv4 714 environments, these bits might be used with the old semantics of 715 "Type Of Service and Precedence". An implementation of a 716 translator SHOULD provide the ability to ignore the IPv6 traffic 717 class and always set the IPv4 TOS Octet to a specified value. In 718 addition, if the translator is at an administrative boundary, the 719 filtering and update considerations of [RFC2475] may be 720 applicable. 722 Total Length: Payload length value from IPv6 header, plus the size 723 of the IPv4 header. 725 Identification: All zero. 727 Flags: The More Fragments flag is set to zero. The Don't Fragments 728 flag is set to one. 730 Fragment Offset: All zero. 732 Time to Live: Hop Limit value copied from IPv6 header. Since the 733 translator is a router, as part of forwarding the packet it needs 734 to decrement either the IPv6 Hop Limit (before the translation) or 735 the IPv4 TTL (after the translation). As part of decrementing the 736 TTL or Hop Limit the translator (as any router) needs to check for 737 zero and send the ICMPv4 "ttl exceeded" or ICMPv6 "hop limit 738 exceeded" error. 740 Protocol: Next Header field copied from IPv6 header. 742 Header Checksum: Computed once the IPv4 header has been created. 744 Source Address: In stateless mode, which is to say that if the IPv6 745 source address is within the range of the IPv6 stateless 746 translation prefix, the IPv4 source address is derived from the 747 IPv6 address. Note that the original IPv6 source address is the 748 IPv4-translatable address. 750 In stateful mode, which is to say that if the IPv6 source address 751 is not within the range of the IPv6 stateless translation prefix, 752 the IPv4 source address and transport layer source port 753 corresponding to the IPv4-related IPv6 source address and source 754 port are derived from the database reflecting current session 755 state in the translator. Database maintenance is described in 756 [I-D.ietf-behave-v6v4-xlate-stateful]. 758 Destination Address: The IPv4 destination address is derived from 759 the IPv6 destination address of the datagram being translated. 760 Note that the original IPv6 destination address is the IPv4-mapped 761 address. 763 If the IPv6 destination address is in the multicast range, the 764 multicast address mapping method should be applied. 766 If any of an IPv6 hop-by-hop options header, destination options 767 header, or routing header with the Segments Left field equal to zero 768 are present in the IPv6 packet, they are ignored i.e., there is no 769 attempt to translate them. However, the Total Length field and the 770 Protocol field is adjusted to "skip" these extension headers. 772 If a routing header with a non-zero Segments Left field is present 773 then the packet MUST NOT be translated, and an ICMPv6 "parameter 774 problem/erroneous header field encountered" (Type 4/Code 0) error 775 message, with the Pointer field indicating the first byte of the 776 Segments Left field, SHOULD be returned to the sender. 778 If the IPv6 packet contains a Fragment header the header fields are 779 set as above with the following exceptions: 781 Total Length: Payload length value from IPv6 header, minus 8 for the 782 Fragment header, plus the size of the IPv4 header. 784 Identification: Copied from the low-order 16-bits in the 785 Identification field in the Fragment header. 787 Flags: The More Fragments flag is copied from the M flag in the 788 Fragment header. The Don't Fragments flag is set to zero allowing 789 this packet to be fragmented by IPv4 routers. 791 Fragment Offset: Copied from the Fragment Offset field in the 792 Fragment header. 794 Protocol: Next Header value copied from Fragment header. 796 3.2. Translating ICMPv6 Headers into ICMPv4 Headers 798 All ICMP messages that are to be translated require that the ICMP 799 checksum field be updated as part of the translation since ICMPv6 800 (unlike ICMPv4) includes a pseudo-header in the checksum just like 801 UDP and TCP. 803 In addition all ICMP packets need to have the Type value translated 804 and, for ICMP error messages, the included IP header also needs 805 translation. Note that the IPv6 addresses in the IPv6 header may not 806 be the IPv4-translatable addresses and there will be no corresponding 807 IPv4 addresses. In this case, a special block of the IPv4 address 808 can be used to indicate this phenomenon. 810 The actions needed to translate various ICMPv6 messages are: 812 ICMPv6 informational messages: 814 Echo Request and Echo Reply (Type 128 and 129) Adjust the type to 815 0 and 8, respectively, and adjust the ICMP checksum both to 816 take the type change into account and to exclude the ICMPv6 817 pseudo-header. 819 MLD Multicast Listener Query/Report/Done (Type 130, 131, 132) 820 Single hop message. Silently drop. 822 Neighbor Discover messages (Type 133 through 137) Single hop 823 message. Silently drop. 825 Unknown informational messages Silently drop. 827 ICMPv6 error messages: 829 Destination Unreachable (Type 1) Set the Type field to 3. 830 Translate the code field as follows: 832 Code 0 (no route to destination): Set Code to 1 (host 833 unreachable). 835 Code 1 (communication with destination administratively 836 prohibited): Set Code to 10 (communication with destination 837 host administratively prohibited). 839 Code 2 (beyond scope of source address): Set Code to 1 (host 840 unreachable). Note that this error is very unlikely since 841 the IPv4-translatable source address is considered to have 842 global scope. 844 Code 3 (address unreachable): Set Code to 1 (host 845 unreachable). 847 Code 4 (port unreachable): Set Code to 3 (port unreachable). 849 Packet Too Big (Type 2) Translate to an ICMPv4 Destination 850 Unreachable with code 4. The MTU field needs to be adjusted 851 for the difference between the IPv4 and IPv6 header sizes 852 taking into account whether or not the packet in error includes 853 a Fragment header. 855 Time Exceeded (Type 3) Set the Type to 11. The Code field is 856 unchanged. 858 Parameter Problem (Type 4) If the Code is 1, translate this to an 859 ICMPv4 protocol unreachable (Type 3, Code 2). Otherwise set 860 the Type to 12 and the Code to zero. The Pointer needs to be 861 updated to point to the corresponding field in the translated 862 inner IP header. 864 Unknown error messages Silently drop. 866 ICMP Error Payload The [RFC4884] length field should be updated 867 to reflect the changed length of the datagram. 869 3.3. Translating ICMPv6 Error Messages into ICMPv4 871 There are some differences between the IPv4 and the IPv6 ICMP error 872 message formats as detailed above. In addition, the ICMP error 873 messages contain the IP header for the packet in error, which needs 874 to be translated just like a normal IP header. The translation of 875 this "packet in error" is likely to change the length of the datagram 876 thus the Total Length field in the outer IPv4 header might need to be 877 updated. 879 +-------------+ +-------------+ 880 | IPv6 | | IPv4 | 881 | Header | | Header | 882 +-------------+ +-------------+ 883 | ICMPv6 | | ICMPv4 | 884 | Header | | Header | 885 +-------------+ +-------------+ 886 | IPv6 | ===> | IPv4 | 887 | Header | | Header | 888 +-------------+ +-------------+ 889 | Partial | | Partial | 890 | Transport | | Transport | 891 | Layer | | Layer | 892 | Header | | Header | 893 +-------------+ +-------------+ 895 Figure 5: IPv6-to-IPv4 ICMP Error Translation 897 The translation of the inner IP header can be done by recursively 898 invoking the function that translated the outer IP headers. Note 899 that the IPv6 addresses in the IPv6 header may not be the IPv4- 900 translatable addresses and there will be no corresponding IPv4 901 addresses. In this case, a special block of the IPv4 address can be 902 used to indicate this phenomenon. 904 3.4. Translator sending ICMPv6 error message 906 If the packet is discarded, then the translator SHOULD be able to 907 send back an ICMPv6 message to the original sender of the packet, 908 unless the discarded packet is itself an ICMPv6 message. The ICMPv6 909 message, if sent, has a type of 1 (Destination Unreachable) and a 910 code of 1 (Communication with destination administratively 911 prohibited). The translator device MUST allow configuring whether 912 the ICMPv6 error messages are sent, rate-limited or not sent. 914 3.5. Transport-layer Header Translation 916 If the address translation algorithm is not checksum neutral, the 917 recalculation and updating of the transport-layer headers MUST be 918 performed. 920 3.6. Knowing when to Translate 922 If the IP/ICMP translator is implemented in a router providing both 923 translation and normal forwarding, and the address is reachable by a 924 more specific route without translation, the router MUST forward it 925 without translating it. When an IP/ICMP translator receives an IPv6 926 datagram addressed to a destination towards the IPv4 domain, the 927 packet will be translated to IPv4. 929 4. IANA Considerations 931 This memo adds no new IANA considerations. 933 Note to RFC Editor: This section will have served its purpose if it 934 correctly tells IANA that no new assignments or registries are 935 required, or if those assignments or registries are created during 936 the RFC publication process. From the author's perspective, it may 937 therefore be removed upon publication as an RFC at the RFC Editor's 938 discretion. 940 5. Security Considerations 942 The use of stateless IP/ICMP translators does not introduce any new 943 security issues beyond the security issues that are already present 944 in the IPv4 and IPv6 protocols and in the routing protocols that are 945 used to make the packets reach the translator. 947 As the Authentication Header [RFC4302] is specified to include the 948 IPv4 Identification field and the translating function is not able to 949 always preserve the Identification field, it is not possible for an 950 IPv6 endpoint to verify the AH on received packets that have been 951 translated from IPv4 packets. Thus AH does not work through a 952 translator. 954 Packets with ESP can be translated since ESP does not depend on 955 header fields prior to the ESP header. Note that ESP transport mode 956 is easier to handle than ESP tunnel mode; in order to use ESP tunnel 957 mode, the IPv6 node needs to be able to generate an inner IPv4 header 958 when transmitting packets and remove such an IPv4 header when 959 receiving packets. 961 6. Acknowledgements 963 This is under development by a large group of people. Those who have 964 posted to the list during the discussion include Andrew Sullivan, 965 Andrew Yourtchenko, Brian Carpenter, Dan Wing, Dave Thaler, Ed 966 Jankiewicz, Fred Baker, Hiroshi Miyata, Iljitsch van Beijnum, John 967 Schnizlein, Kevin Yin, Magnus Westerlund, Marcelo Bagnulo Braun, 968 Margaret Wasserman, Masahito Endo, Phil Roberts, Philip Matthews, 969 Remi Denis-Courmont, Remi Despres, and Xing Li. 971 7. References 973 7.1. Normative References 975 [I-D.xli-behave-v4v6-prefix] 976 Bao, C., Baker, F., and X. Li, "IPv4/IPv6 Translation 977 Prefix Recommendation", draft-xli-behave-v4v6-prefix-00 978 (work in progress), April 2009. 980 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 981 August 1980. 983 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 984 September 1981. 986 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 987 RFC 792, September 1981. 989 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 990 RFC 793, September 1981. 992 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 993 Requirement Levels", BCP 14, RFC 2119, March 1997. 995 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 996 (IPv6) Specification", RFC 2460, December 1998. 998 [RFC2765] Nordmark, E., "Stateless IP/ICMP Translation Algorithm 999 (SIIT)", RFC 2765, February 2000. 1001 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 1002 Architecture", RFC 4291, February 2006. 1004 [RFC4443] Conta, A., Deering, S., and M. Gupta, "Internet Control 1005 Message Protocol (ICMPv6) for the Internet Protocol 1006 Version 6 (IPv6) Specification", RFC 4443, March 2006. 1008 [RFC4884] Bonica, R., Gan, D., Tappan, D., and C. Pignataro, 1009 "Extended ICMP to Support Multi-Part Messages", RFC 4884, 1010 April 2007. 1012 [RFC5382] Guha, S., Biswas, K., Ford, B., Sivakumar, S., and P. 1013 Srisuresh, "NAT Behavioral Requirements for TCP", BCP 142, 1014 RFC 5382, October 2008. 1016 7.2. Informative References 1018 [I-D.ietf-behave-address-format] 1019 Huitema, C., Bao, C., Bagnulo, M., Boucadair, M., and X. 1020 Li, "IPv6 Addressing of IPv4/IPv6 Translators", 1021 draft-ietf-behave-address-format-00 (work in progress), 1022 August 2009. 1024 [I-D.ietf-behave-v6v4-framework] 1025 Baker, F., Li, X., Bao, C., and K. Yin, "Framework for 1026 IPv4/IPv6 Translation", 1027 draft-ietf-behave-v6v4-framework-01 (work in progress), 1028 September 2009. 1030 [I-D.ietf-behave-v6v4-xlate-stateful] 1031 Bagnulo, M., Matthews, P., and I. Beijnum, "NAT64: Network 1032 Address and Protocol Translation from IPv6 Clients to IPv4 1033 Servers", draft-ietf-behave-v6v4-xlate-stateful-02 (work 1034 in progress), October 2009. 1036 [I-D.venaas-behave-v4v6mc-framework] 1037 Venaas, S., "Framework for IPv4/IPv6 Multicast 1038 Translation", draft-venaas-behave-v4v6mc-framework-00 1039 (work in progress), July 2009. 1041 [Miller] Miller, G., "Email to the ngtrans mailing list", 1042 March 1999. 1044 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 1045 November 1990. 1047 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 1048 "Definition of the Differentiated Services Field (DS 1049 Field) in the IPv4 and IPv6 Headers", RFC 2474, 1050 December 1998. 1052 [RFC2475] Blake, S., Black, D., Carlson, M., Davies, E., Wang, Z., 1053 and W. Weiss, "An Architecture for Differentiated 1054 Services", RFC 2475, December 1998. 1056 [RFC2710] Deering, S., Fenner, W., and B. Haberman, "Multicast 1057 Listener Discovery (MLD) for IPv6", RFC 2710, 1058 October 1999. 1060 [RFC3171] Albanna, Z., Almeroth, K., Meyer, D., and M. Schipper, 1061 "IANA Guidelines for IPv4 Multicast Address Assignments", 1062 BCP 51, RFC 3171, August 2001. 1064 [RFC3307] Haberman, B., "Allocation Guidelines for IPv6 Multicast 1065 Addresses", RFC 3307, August 2002. 1067 [RFC3590] Haberman, B., "Source Address Selection for the Multicast 1068 Listener Discovery (MLD) Protocol", RFC 3590, 1069 September 2003. 1071 [RFC3810] Vida, R. and L. Costa, "Multicast Listener Discovery 1072 Version 2 (MLDv2) for IPv6", RFC 3810, June 2004. 1074 [RFC4213] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms 1075 for IPv6 Hosts and Routers", RFC 4213, October 2005. 1077 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 1078 Internet Protocol", RFC 4301, December 2005. 1080 [RFC4302] Kent, S., "IP Authentication Header", RFC 4302, 1081 December 2005. 1083 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", 1084 RFC 4303, December 2005. 1086 Authors' Addresses 1088 Xing Li 1089 CERNET Center/Tsinghua University 1090 Room 225, Main Building, Tsinghua University 1091 Beijing, 100084 1092 China 1094 Phone: +86 10-62785983 1095 Email: xing@cernet.edu.cn 1097 Congxiao Bao 1098 CERNET Center/Tsinghua University 1099 Room 225, Main Building, Tsinghua University 1100 Beijing, 100084 1101 China 1103 Phone: +86 10-62785983 1104 Email: congxiao@cernet.edu.cn 1105 Fred Baker 1106 Cisco Systems 1107 Santa Barbara, California 93117 1108 USA 1110 Phone: +1-408-526-4257 1111 Email: fred@cisco.com