idnits 2.17.1 draft-ietf-ospf-ipv4-embedded-ipv6-routing-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 : ---------------------------------------------------------------------------- == There are 2 instances of lines with non-RFC3849-compliant IPv6 addresses in the document. If these are example addresses, they should be changed. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (April 12, 2013) is 4004 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 6145 (Obsoleted by RFC 7915) Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Cheng 3 Internet-Draft Huawei Technologies 4 Intended status: Informational M. Boucadair 5 Expires: October 14, 2013 France Telecom 6 A. Retana 7 Cisco Systems 8 April 12, 2013 10 Routing for IPv4-embedded IPv6 Packets 11 draft-ietf-ospf-ipv4-embedded-ipv6-routing-10 13 Abstract 15 This document describes routing packets destined to IPv4-embedded 16 IPv6 addresses across an IPv6 core using OSPFv3 with a separate 17 routing table. 19 Status of This Memo 21 This Internet-Draft is submitted in full conformance with the 22 provisions of BCP 78 and BCP 79. 24 Internet-Drafts are working documents of the Internet Engineering 25 Task Force (IETF). Note that other groups may also distribute 26 working documents as Internet-Drafts. The list of current Internet- 27 Drafts is at http://datatracker.ietf.org/drafts/current/. 29 Internet-Drafts are draft documents valid for a maximum of six months 30 and may be updated, replaced, or obsoleted by other documents at any 31 time. It is inappropriate to use Internet-Drafts as reference 32 material or to cite them other than as "work in progress." 34 This Internet-Draft will expire on October 14, 2013. 36 Copyright Notice 38 Copyright (c) 2013 IETF Trust and the persons identified as the 39 document authors. All rights reserved. 41 This document is subject to BCP 78 and the IETF Trust's Legal 42 Provisions Relating to IETF Documents 43 (http://trustee.ietf.org/license-info) in effect on the date of 44 publication of this document. Please review these documents 45 carefully, as they describe your rights and restrictions with respect 46 to this document. Code Components extracted from this document must 47 include Simplified BSD License text as described in Section 4.e of 48 the Trust Legal Provisions and are provided without warranty as 49 described in the Simplified BSD License. 51 Table of Contents 53 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 54 1.1. The Scenario . . . . . . . . . . . . . . . . . . . . . . 3 55 1.2. Routing Solution per RFC5565 . . . . . . . . . . . . . . 4 56 1.3. An Alternative Routing Solution with OSPFv3 . . . . . . . 4 57 1.4. OSPFv3 Routing with a Specific Topology . . . . . . . . . 6 58 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 7 59 3. Provisioning . . . . . . . . . . . . . . . . . . . . . . . . 7 60 3.1. Deciding the IPv4-embedded IPv6 Topology . . . . . . . . 7 61 3.2. Maintaining a Dedicated IPv4-embedded IPv6 Routing Table 7 62 3.3. OSPFv3 Topology with a Separate Instance ID . . . . . . . 7 63 3.4. OSPFv3 Topology with the Default Instance . . . . . . . . 8 64 4. IP Packets Translation . . . . . . . . . . . . . . . . . . . 8 65 4.1. Address Translation . . . . . . . . . . . . . . . . . . . 9 66 5. Advertising IPv4-embedded IPv6 Routes . . . . . . . . . . . . 9 67 5.1. Advertising IPv4-embedded IPv6 Routes through an IPv6 68 Transit Network . . . . . . . . . . . . . . . . . . . . . 10 69 5.1.1. Routing Metrics . . . . . . . . . . . . . . . . . . . 10 70 5.1.2. Forwarding Address . . . . . . . . . . . . . . . . . 10 71 5.2. Advertising IPv4 Addresses into Client Networks . . . . . 10 72 6. Aggregation on IPv4 Addresses and Prefixes . . . . . . . . . 11 73 7. Forwarding . . . . . . . . . . . . . . . . . . . . . . . . . 11 74 8. Backdoor Connections . . . . . . . . . . . . . . . . . . . . 12 75 9. Prevention of Loops . . . . . . . . . . . . . . . . . . . . . 12 76 10. MTU Issues . . . . . . . . . . . . . . . . . . . . . . . . . 12 77 11. Security Considerations . . . . . . . . . . . . . . . . . . . 13 78 12. Operational Considerations . . . . . . . . . . . . . . . . . 13 79 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 14 80 14. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 14 81 15. References . . . . . . . . . . . . . . . . . . . . . . . . . 15 82 15.1. Normative References . . . . . . . . . . . . . . . . . . 15 83 15.2. Informative References . . . . . . . . . . . . . . . . . 15 84 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 15 86 1. Introduction 87 This document describes a routing scenario where IPv4 packets are 88 transported over an IPv6 network, based on [RFC6145] and [RFC6052], 89 along with a separate OSPFv3 routing table for IPv4-embedded IPv6 90 routes in the IPv6 network. This document does not introduce any new 91 IPv6 transition mechanism. 93 In this document the following terminology is used: 95 o An IPv4-embedded IPv6 address denotes an IPv6 address which 96 contains an embedded 32-bit IPv4 address constructed according to 97 the rules defined in [RFC6052]. 99 o IPv4-embedded IPv6 packets are packets of which destination 100 addresses are IPv4-embedded IPv6 addresses. 102 o AFBR (Address Family Border Router, [RFC5565]) refers to an edge 103 router, which supports both IPv4 and IPv6 address families, but 104 the backbone network it connects to only supports either the IPv4 105 or IPv6 address family. 107 o AFXLBR (Address Family Translation Border Router) is defined in 108 this document. It refers to a border router that supports both 109 IPv4 and IPv6 address families, located on the boundary of an 110 IPv4-only network and an IPv6-only network, and is capable of 111 performing IP header translation between IPv4 and IPv6 according 112 to [RFC6145]. 114 1.1. The Scenario 116 Due to exhaustion of public IPv4 addresses, there has been a 117 continuing effort within the IETF on IPv6 transitional techniques. 118 In the course of the transition, it is certain that networks based on 119 IPv4 and IPv6 technologies respectively, will co-exist at least for 120 some time. One scenario of this co-existence is the inter-connection 121 of IPv4-only and IPv6-only networks, and in particular, when an 122 IPv6-only network serves as inter-connection between several 123 segregated IPv4-only networks. In this scenario, IPv4 packets are 124 transported over the IPv6 network between IPv4 networks. In order to 125 forward an IPv4 packet from a source IPv4 network to the destination 126 IPv4 network, IPv4 reachability information must be exchanged between 127 the IPv4 networks by some mechanism. 129 In general, running an IPv6-only network would reduce OPEX and 130 optimize the operation compared to IPv4-IPv6 dual-stack environment. 131 Some solutions have been proposed to allow delivery of IPv4 services 132 over an IPv6-only network. This document focuses on an engineering 133 technique which aims to separate the routing table dedicated to 134 IPv4-embedded IPv6 destinations from native IPv6 ones. 136 Maintaining a separate routing table for IPv4-embedded IPv6 routes 137 optimizes IPv4 packets forwarding. It also prevents overload of the 138 native IPv6 routing tables. A separate routing table can be 139 generated from a separate routing instance or a separate routing 140 topology. 142 1.2. Routing Solution per RFC5565 144 The aforementioned scenario is described in [RFC5565], i.e., IPv4 145 -over-IPv6 scenario, where the network core is IPv6-only, and the 146 inter-connected IPv4 networks are called IPv4 client networks. The P 147 Routers in the core only support IPv6 but the AFBRs (Address Family 148 Border Routers) support IPv4 on interfaces facing IPv4 client 149 networks, and IPv6 on interfaces facing the core. The routing 150 solution defined in [RFC5565] for this scenario is to run i-BGP among 151 AFBRs to exchange IPv4 routing information in the core, and the IPv4 152 packets are forwarded from one IPv4 client network to the other 153 through a softwire using tunneling technology such as MPLS LSP, GRE, 154 L2TPv3, etc. 156 1.3. An Alternative Routing Solution with OSPFv3 158 In this document, we propose an alternative routing solution for the 159 scenario described in Section 1.1, where several segregated IPv4 160 networks, called IPv4 client networks, are inter-connected by an IPv6 161 network. The IPv6 network and the inter-connected IPv4 networks may 162 or may not belong to the same Autonomous System. We refer to the 163 border node on the boundary of an IPv4 client network and the IPv6 164 network as an Address Family Translation Border Router (AFXLBR), 165 which supports both the IPv4 and IPv6 address families, and is 166 capable of translating an IPv4 packet to an IPv6 packet, and vice 167 versa, according to [RFC6145]. The described scenario is illustrated 168 in Figure 1. 170 +--------+ +--------+ 171 | IPv4 | | IPv4 | 172 | Client | | Client | 173 | Network| | Network| 174 +--------+ +--------+ 175 | \ / | 176 | \ / | 177 | \ / | 178 | X | 179 | / \ | 180 | / \ | 181 | / \ | 183 +--------+ +--------+ 184 | AFXLBR | | AFXLBR | 185 +--| IPv4/6 |---| IPv4/6 |--+ 186 | +--------+ +--------+ | 187 +--------+ | | +--------+ 188 | IPv6 | | | | IPv6 | 189 | Client |----| |----| Client | 190 | Network| | IPv6 | | Network| 191 +--------+ | only | +--------+ 192 | | 193 | +--------+ +--------+ | 194 +--| AFXLBR |---| AFXLBR |--+ 195 | IPv4/6 | | IPv4/6 | 196 +--------+ +--------+ 197 | \ / | 198 | \ / | 199 | \ / | 200 | X | 201 | / \ | 202 | / \ | 203 | / \ | 204 +--------+ +--------+ 205 | IPv4 | | IPv4 | 206 | Client | | Client | 207 | Network| | Network| 208 +--------+ +--------+ 210 Figure 1: Segregated IPv4 Networks Inter-connected by an IPv6 Network 212 Since the scenario occurs most commonly within an organization, an 213 IPv6 prefix can be locally allocated and used by AFXLBRs to construct 214 IPv4-embedded IPv6 addresses according to [RFC6052]. The embedded 215 IPv4 address or prefix belongs to an IPv4 client network that is 216 connected to the AFXLBR. An AFXLBR injects IPv4-embedded IPv6 217 addresses and prefixes into the IPv6 network using OSPFv3, and it 218 also installs IPv4-embedded IPv6 routes advertised by other AFXLBRs. 220 When an AFXLBR receives an IPv4 packet from a locally connected IPv4 221 client network and destined to a remote IPv4 client network, it 222 translates the IPv4 header to the relevant IPv6 header according to 223 [RFC6145], and in that process, source and destination IPv4 address 224 are translated into IPv4-embedded IPv6 addresses, respectively, 225 according to [RFC6052]. The resulting IPv6 packet is then forwarded 226 to the AFXLBR that connects to the destination IPv4 client network. 227 The remote AFXLBR derives the IPv4 source and destination addresses 228 from the IPv4-embedded IPv6 addresses, respectively, according to 229 [RFC6052], and translates the header of the received IPv6 packet to 230 the relevant IPv4 header according to [RFC6145]. The resulting IPv4 231 packet is then forwarded according to the IPv4 routing table 232 maintained on the AFXLBR. 234 There are use cases where the proposed routing solution is useful. 235 One case is that some border nodes do not participate in i-BGP for 236 routes exchange, or i-BGP is not used at all. Another case is when 237 tunnels are not deployed in the IPv6 network, or native IPv6 238 forwarding is preferred. Note that with this routing solution, the 239 IPv4 and IPv6 header translation performed in both directions by the 240 AFXLBR is stateless. 242 1.4. OSPFv3 Routing with a Specific Topology 244 In general, IPv4-embedded IPv6 packets can be forwarded just like 245 native IPv6 packets with OSPFv3 running in the IPv6 network. 246 However, this would require IPv4-embedded IPv6 routes to be flooded 247 throughout the entire IPv6 network and stored on every router. This 248 is not desirable from the scaling perspective. Moreover, since all 249 IPv6 routes are stored in the same routing table, it would be 250 inconvenient to manage the resource required for routing and 251 forwarding based on traffic category, if so desired. 253 To improve the situation, a separate OSPFv3 routing table can be 254 constructed that is dedicated to the IPv4-embedded IPv6 topology, and 255 that table is solely used for routing IPv4-embedded IPv6 packets in 256 the IPv6 network. The IPv4-embedded IPv6 topology includes all the 257 participating AFXLBR routers and a set of P(rovider) Routers 258 providing redundant connectivity with alternate routing paths. 260 There are two methods to build a separate OSPFv3 routing table for 261 IPv4-embedded IPv6 routes: 263 o The first one is to run a separate OSPFv3 instance for 264 IPv4-embedded IPv6 topology in the IPv6 network according to 265 [RFC5838]. 267 o The second one is to stay with the existing OSPFv3 instance that 268 already operates in the IPv6 network, but maintain a separate 269 IPv4-embedded IPv6 topology, according to 270 [I-D.ietf-ospf-mt-ospfv3]. 272 With either method, there would be a dedicated IPv4-embedded IPv6 273 topology that is maintained on all participating AFXLBR and P 274 Routers, along with a dedicated IPv4-embedded IPv6 routing table. 275 This routing table is then used solely in the IPv6 network for 276 IPv4-embedded IPv6 packets. 278 It would be an operator's preference as which method is to be used. 279 This document elaborates on how configuration is done for each method 280 and related routing issues that are common to both. 282 This document only focuses on unicast routing for IPv4-embedded IPv6 283 packets using OSPFv3. 285 2. Requirements Language 287 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 288 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 289 document are to be interpreted as described in [RFC2119]. 291 3. Provisioning 293 3.1. Deciding the IPv4-embedded IPv6 Topology 295 Before deploying configurations that use a separate OSPFv3 routing 296 table for IPv4-embedded IPv6 addresses and prefixes, a decision must 297 be made on the set of routers and their interfaces in the IPv6 298 network that should be part of the IPv4-embedded IPv6 topology. 300 For the purpose of this IPv4-embedded IPv6 topology, all AFXLBRs that 301 connect to IPv4 client networks MUST be members of this topology. An 302 AFXLBR MUST have at least one connection with a P Router in the IPv6 303 network or another AFXLBR. 305 The IPv4-embedded IPv6 topology is a sub-topology of the entire IPv6 306 network, and if all routers (including AFXLBRs and P-routers) and all 307 their interfaces are included, the two topologies converge. 308 Generally speaking, when this sub-topology contains more inter- 309 connected P Routers, there would be more routing paths across the 310 IPv6 network from one IPv4 client network to the other; however, this 311 requires more routers in the IPv6 network to participate in 312 IPv4-embedded IPv6 routing. In any case, the IPv4-embedded IPv6 313 topology MUST be continuous with no partitions. 315 3.2. Maintaining a Dedicated IPv4-embedded IPv6 Routing Table 317 In an IPv6 network, in order to maintain a separate IPv6 routing 318 table that contains routes for IPv4-embedded IPv6 destinations only, 319 OSPFv3 needs to use the mechanism defined either in [RFC5838] or in 320 [I-D.ietf-ospf-mt-ospfv3] with the required configuration, as 321 described in Section 3.3 and Section 3.4, respectively. 323 3.3. OSPFv3 Topology with a Separate Instance ID 324 It is assumed that the IPv6 network that is inter-connected with IPv4 325 networks in this document is under one administration and as such, an 326 OSPFv3 instance ID (IID) is allocated locally and used for OSPFv3 327 operation dedicated to unicast IPv4-embedded IPv6 routing in an IPv6 328 network. This IID is configured on OSPFv3 router interfaces that 329 participate in the IPv4-embedded IPv6 topology. 331 The range for a locally configured OSPFv3 IID is from 192 to 255, 332 inclusive, and this IID must be used to encode the "Instance ID" 333 field in the packet header of OSPFv3 packets associated with the 334 OSPFv3 instance. 336 In addition, the "AF" bit in the OSPFv3 Option field MUST be set. 338 During Hello packet processing, an adjacency may only be established 339 when the received Hello packet contains the same Instance ID as 340 configured on the receiving OSPFv3 interface. This insures that only 341 interfaces configured as part of the OSPFv3 unicast IPv4-embedded 342 IPv6 topology are used for IPv4-embedded IPv6 unicast routing. 344 For more details, the reader is referred to [RFC5838]. 346 3.4. OSPFv3 Topology with the Default Instance 348 Similar to that as described in the previous section, an OSPFv3 349 multi-topology ID (MT-ID) is locally allocated and used for an OSPFv3 350 operation including unicast IPv4-embedded IPv6 routing in an IPv6 351 network. This MT-ID is configured on each OSPFv3 router interface 352 that participates in this routing topology. 354 The range for a locally configured OSPFv3 MT-ID is from 32 to 255, 355 inclusive, and this MT-ID must be used to encode the "MT-ID" field 356 included in extended Link State Advertisements (LSAs) for the 357 IPv4-embedded IPv6 unicast topology as documented in 358 [I-D.ietf-ospf-mt-ospfv3]. 360 In addition, the MT bit in the OSPFv3 Option field MUST be set. 362 For more details, the reader is referred to 363 [I-D.ietf-ospf-mt-ospfv3]. 365 4. IP Packets Translation 367 When transporting IPv4 packets across an IPv6 network with the 368 mechanism described above, an IPv4 packet is translated to an IPv6 369 packet at the ingress AFXLBR, and the IPv6 packet is translated back 370 to an IPv4 packet at the egress AFXLBR. The IP packet header 371 translation is accomplished in stateless manner according to rules 372 specified in [RFC6145], with the address translation details 373 explained in the next sub-section. 375 4.1. Address Translation 377 Prior to address translation, an IPv6 prefix is allocated by the 378 operator and it is used to form IPv4-embedded IPv6 addresses. 380 The IPv6 prefix can either be the well-known IPv6 prefix (WKP) 381 64:ff9b::/96, or a network-specific prefix that is unique to the 382 organzation; and for the latter case, the IPv6 prefix length may be 383 32, 40, 48, 56 or 64. In either case, this IPv6 prefix is used 384 during the address translation between an IPv4 address and an 385 IPv4-embedded IPv6 address, as described in [RFC6052]. 387 During translation from an IPv4 header to an IPv6 header at an 388 ingress AFXLBR, the source IPv4 address and destination IPv4 address 389 are translated into the corresponding IPv6 source address and 390 destination IPv6 address, respectively, and during translation from 391 an IPv6 header to an IPv4 header at an egress AFXLBR, the source IPv6 392 address and destination IPv6 address are translated into the 393 corresponding IPv4 source address and destination IPv4 address, 394 respectively. Note that the address translation is accomplished in a 395 stateless manner. 397 When a well-known IPv6 prefix (WKP) is used, [RFC6052] allows only 398 global IPv4 addresses to be embedded in the IPv6 address. An IPv6 399 address composed with a WKP and a non-global IPv4 address is hence 400 invalid, and packets that contain such address received by an AFXLBR 401 are dropped. 403 In the case where both the IPv4 client networks and the IPv6 transit 404 network belong to the same organization, non-global IPv4 addresses 405 may be used with a network-specific prefix [RFC6052]. 407 5. Advertising IPv4-embedded IPv6 Routes 409 In order to forward IPv4 packets to the proper destination across an 410 IPv6 network, IPv4 reachability needs to be disseminated throughout 411 the IPv6 network and this is performed by AFXLBRs that connect to 412 IPv4 client networks using OSPFv3. 414 With the scenario described in this document, i.e., a set of AFXLBRs 415 that inter-connect a bunch of IPv4 client networks with an IPv6 416 network, the IPv4 networks and IPv6 networks belong to the same or 417 separate Autonomous Systems, and as such, these AFXLBRs behave as AS 418 Boundary Routers (ASBRs). 420 5.1. Advertising IPv4-embedded IPv6 Routes through an IPv6 Transit 421 Network 423 IPv4 addresses and prefixes in an IPv4 client network are translated 424 into IPv4-embedded IPv6 addresses and prefixes, respectively, using 425 the IPv6 prefix allocated by the operator and the method specified in 426 [RFC6052]. These routes are then advertised by one or more attached 427 ASBRs into the IPv6 transit network using AS-External-LSAs [RFC5340], 428 i.e., with advertising scope comprising the entire Autonomous System. 430 5.1.1. Routing Metrics 432 By default, the metric in an AS-External-LSA that carries an 433 IPv4-embedded IPv6 address or prefixes is a Type 1 external metric, 434 which is comparable to the link state metric and we assume that in 435 most cases, OSPFv2 is used in client IPv4 networks. This metric is 436 added to the metric of the intra-AS path to the ASBR during the 437 OSPFv3 route calculation. Through ASBR configuration, the metric can 438 be set to a Type 2 external metric, which is considered much larger 439 than the metric for any intra-AS path. Refer to the OSPFv3 440 specification [RFC5340] for more detail. In either case, an external 441 metric may take the same value as in an IPv4 network (using OSPFv2 or 442 another routing protocol), but may also be specified based on some 443 routing policy; the details of which are outside of the scope of this 444 document. 446 5.1.2. Forwarding Address 448 If the "Forwarding Address" field of an OSPFv3 AS-External-LSA is 449 used to carry an IPv6 address, that must also be an IPv4-embedded 450 IPv6 address where the embedded IPv4 address is the destination 451 address in an IPv4 client network. However, since an AFXLBR sits on 452 the border of an IPv4 network and an IPv6 network, it is RECOMMENDED 453 that the "Forwarding Address" field is not used, so that the AFXLBR 454 can make the forwarding decision based on its own IPv4 routing table. 456 5.2. Advertising IPv4 Addresses into Client Networks 458 IPv4-embedded IPv6 routes injected into the IPv6 network from one 459 IPv4 client network MAY be advertised into another IPv4 client 460 network, after the associated destination addresses and prefixes are 461 translated back to IPv4 addresses and prefixes, respectively. This 462 operation is similar to normal OSPFv3 operation, wherein an AS- 463 External-LSA can be advertised in a non-backbone area by default. 465 An IPv4 client network can limit which advertisements it receives 466 through configuration. 468 For the purpose of this document, IPv4-embedded IPv6 routes MUST NOT 469 be advertised into any IPv6 client networks that also connected to 470 the IPv6 transit network. 472 6. Aggregation on IPv4 Addresses and Prefixes 474 In order to reduce the amount of LSAs that are injected to the IPv6 475 network, an implementation should provide mechanisms to aggregate 476 IPv4 addresses and prefixes at AFXLBR prior to advertisement as 477 IPv4-embedded IPv6 addresses and prefixes. In general, the 478 aggregation practice should be based on routing policy, which is 479 outside of the scope of this document. 481 7. Forwarding 483 There are three cases in forwarding IP packets in the scenario 484 described in this document: 486 1. On an AFXLBR, if an IPv4 packet that is received on an interface 487 connecting to an IPv4 client network with a destination IPv4 488 address belonging to another IPv4 client network, the header of 489 the packet is translated to the corresponding IPv6 header as 490 described in Section 4, and the packet is then forwarded to the 491 destination AFXLBR that advertised the IPv4-embedded IPv6 address 492 into the IPv6 network. 494 2. On an AFXLBR, if an IPv4-embedded IPv6 packet is received and the 495 embedded destination IPv4 address is in its IPv4 routing table, 496 the header of the packet is translated to the corresponding IPv4 497 header as described in Section 4, and the packet is then 498 forwarded accordingly. 500 3. On any router that is within the IPv4-embedded IPv6 topology 501 subset of the IPv6 network, if an IPv4-embedded IPv6 packet is 502 received and a route is found in the IPv4-embedded IPv6 routing 503 table, the packet is forwarded to the IPv6 next-hop just like the 504 handling for a normal IPv6 packet, without any translation. 506 The classification of IPv4-embedded IPv6 packet is according to the 507 IPv6 prefix of the destination address, which is either the Well 508 Known Prefix (i.e., 64:ff9b::/96) or locally allocated as defined in 509 [RFC6052]. 511 8. Backdoor Connections 513 In some deployments, IPv4 client networks are inter-connected across 514 the IPv6 network, but also directly connected to each other. The 515 direct connections between IPv4 client networks, as sometimes called 516 "backdoor" connections, can certainly be used to transport IPv4 517 packets between IPv4 client networks. In general, backdoor 518 connections are preferred over the IPv6 network since there requires 519 no address family translation. 521 9. Prevention of Loops 523 If an LSA sent from an AFXLBR into a client network could then be 524 received by another AFXLBR, it would be possible for routing loops to 525 occur. To prevent loops, an AFXLBR MUST set the DN-bit [RFC4576] in 526 any LSA that it sends to a client network. The AFXLBR MUST also 527 ignore any LSA received from a client network that already has the 528 DN-bit sent. 530 10. MTU Issues 532 In the IPv6 network, there are no new MTU issues introduced by this 533 document. If a separate OSPFv3 instance (per [RFC5838]) is used for 534 IPv4-embedded IPv6 routing, the MTU handling in the IPv6 network is 535 the same as that of the default OSPFv3 instance. If a separate 536 OSPFv3 topology (according to [I-D.ietf-ospf-mt-ospfv3]) is used for 537 IPv4-embedded IPv6 routing, the MTU handling in the IPv6 network is 538 the same as that of the default OSPFv3 topology. 540 However, the MTU in the IPv6 network may be different than that of 541 IPv4 client networks. Since an IPv6 router will never fragment a 542 packet, the packet size of any IPv4-embedded IPv6 packet entering the 543 IPv6 network must be equal to or less than the MTU of the IPv6 544 network. In order to achieve this requirement, it is recommended 545 that AFXLBRs perform IPv6 path discovery among themselves and the 546 resulting MTU, after taking into account of the difference between 547 the IPv4 header length and the IPv6 header length, must be 548 "propagated" into IPv4 client networks, e.g., included in the OSPFv2 549 Database Description packet. 551 The details of passing the proper MTU into IPv4 client networks are 552 beyond the scope of this document. 554 11. Security Considerations 556 There are several security aspects that require attention in the 557 deployment practice described in this document. 559 In the OSPFv3 transit network, the security considerations for OSPFv3 560 are covered in [RFC5340], and in particular, IPsec can be used for 561 OSPFv3 authentication and confidentiality as suggested in [RFC5838]. 563 When a separate OSPFv3 instance is used to support IPv4-embedded IPv6 564 routing, the same Security Association (SA) (refer to [RFC4552] ) 565 must be used by the embedded IPv4 address instance as other instances 566 utilizing the same link as specified in [RFC5838]. 568 Security considerations as currently documented in [RFC6052] must 569 also be thought through with proper implementation including the 570 following: 572 o The IPv6 prefix that is used to carry an embedded IPv4 address 573 (refer to Section 4.1) must be configured by the authorized 574 operator on all participating AFXLBRs in a secure manner. This is 575 to help prevent an malicious attack resulting in network 576 disruption, denial of service, and possible information 577 disclosure. 579 o Effective mechanisms (such as reverse path checking) must be 580 implemented in the IPv6 transit network (including AFXLIBR nodes) 581 to prevent spoofing on embedded IPv4 addresses, which, otherwise, 582 might be used as source addresses of malicious packets. 584 o If firewalls are used in IPv4 and/or IPv6 networks, the 585 configuration on the routers must be consistent so there are no 586 holes in the IPv4 address filtering. 588 The details of security handling are beyond the scope of this 589 document. 591 12. Operational Considerations 593 This document put together some mechanisms based on existing 594 technologies developed by IETF as an integrated solution to transport 595 IPv4 packets over an IPv6 network using a separate OSPFv3 routing 596 table. There are several aspects that require attention for the 597 deployment and operation. 599 The tunnel-based solution documented in [RFC5565] and the solution 600 proposed in this document are both used for transporting IPv4 packets 601 over an IPv6 network, with different mechanisms. The two methods are 602 not related to each other, and they can co-exist in the same network 603 if so deployed, without any conflict. 605 If one approach is to be deployed, it is the operator's decision for 606 the choice. Note that each approach has its own characteristics and 607 requirements. E.g., the tunnel-based solution requries a mesh of 608 inter-AFBR softwires (tunnels) spanning the IPv6 network, as well as 609 iBGP to exchange routes between AFBRs ([RFC5565]); the approach in 610 this document requires AFXLBR capable of perfoming IPv4-IPv6 packet 611 header translation per [RFC6145]. 613 To deploy the solution as documented here, there requires some 614 configurations. An IPv6 prefix must first be chosen that is used to 615 form all the IPv4-embedded IPv6 addresses and prefixes advertised by 616 AFXLBR in the IPv6 network; the detail is referred to Section 4.1. 617 If the IPv4-embedded IPv6 routing table is created by using a 618 separate OSPFv3 instance in the IPv6 network, configuration is 619 accomplished according to [RFC5838], as described in Section 3.3, and 620 if the default OSPFv3 instance is used instead, configuration is 621 accomplished according to [I-D.ietf-ospf-mt-ospfv3], as described in 622 Section 3.4. 624 With the solution as described in this document, IPv4-embedded IPv6 625 addresses and prefixes will be injected by AFXLBR into some part of 626 the IPv6 network (see Section 3.1 for details), and a separate 627 routing table will be used for IPv4-embedded IPv6 routing. Care must 628 be taken during the network design, such that 1) aggregation are 629 performed on IPv4 addresses and prefixes before being advertised in 630 the IPv6 network as described in Section 6, and 2) estimates are made 631 as the amount of IPv4-embedded IPv6 routes that would be disseminated 632 in the IPv6 network, and the size of the separate OSPFv3 routing 633 table. Note this document does not change any behavior of OSPFv3, 634 and the existing or common practice should apply. 636 13. IANA Considerations 638 No new IANA assignments are required for this document. 640 14. Acknowledgements 642 Many thanks to Acee Lindem, Dan Wing, Joel Halpern, Mike Shand and 643 Brian Carpenter for their comments. 645 15. References 647 15.1. Normative References 649 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 650 Requirement Levels", BCP 14, RFC 2119, March 1997. 652 [RFC4576] Rosen, E., Psenak, P., and P. Pillay-Esnault, "Using a 653 Link State Advertisement (LSA) Options Bit to Prevent 654 Looping in BGP/MPLS IP Virtual Private Networks (VPNs)", 655 RFC 4576, June 2006. 657 [RFC5565] Wu, J., Cui, Y., Metz, C., and E. Rosen, "Softwire Mesh 658 Framework", RFC 5565, June 2009. 660 [RFC5838] Lindem, A., Mirtorabi, S., Roy, A., Barnes, M., and R. 661 Aggarwal, "Support of Address Families in OSPFv3", RFC 662 5838, April 2010. 664 [RFC6145] Li, X., Bao, C., and F. Baker, "IP/ICMP Translation 665 Algorithm", RFC 6145, April 2011. 667 15.2. Informative References 669 [I-D.ietf-ospf-mt-ospfv3] 670 Mirtorabi, S. and A. Roy, "Multi-topology routing in 671 OSPFv3 (MT-OSPFv3)", draft-ietf-ospf-mt-ospfv3-03 (work in 672 progress), July 2007. 674 [RFC4552] Gupta, M. and N. Melam, "Authentication/Confidentiality 675 for OSPFv3", RFC 4552, June 2006. 677 [RFC5340] Coltun, R., Ferguson, D., Moy, J., and A. Lindem, "OSPF 678 for IPv6", RFC 5340, July 2008. 680 [RFC6052] Bao, C., Huitema, C., Bagnulo, M., Boucadair, M., and X. 681 Li, "IPv6 Addressing of IPv4/IPv6 Translators", RFC 6052, 682 October 2010. 684 Authors' Addresses 686 Dean Cheng 687 Huawei Technologies 688 2330 Central Expressway 689 Santa Clara, California 95050 690 USA 692 Email: dean.cheng@huawei.com 693 Mohamed Boucadair 694 France Telecom 695 Rennes 35000 696 France 698 Email: mohamed.boucadair@orange.com 700 Alvaro Retana 701 Cisco Systems 702 7025 Kit Creek Rd. 703 Research Triangle Park, North Carolina 27709 704 USA 706 Email: aretana@cisco.com