idnits 2.17.1 draft-ietf-multimob-pmipv6-source-09.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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 31, 2014) is 3679 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Experimental ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 4601 (Obsoleted by RFC 7761) == Outdated reference: A later version (-10) exists of draft-ietf-multimob-fmipv6-pfmipv6-multicast-05 Summary: 1 error (**), 0 flaws (~~), 2 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 MULTIMOB Group T. Schmidt, Ed. 3 Internet-Draft HAW Hamburg 4 Intended status: Experimental S. Gao 5 Expires: October 2, 2014 H. Zhang 6 Beijing Jiaotong University 7 M. Waehlisch 8 link-lab & FU Berlin 9 March 31, 2014 11 Mobile Multicast Sender Support in Proxy Mobile IPv6 (PMIPv6) Domains 12 draft-ietf-multimob-pmipv6-source-09 14 Abstract 16 Multicast communication can be enabled in Proxy Mobile IPv6 domains 17 via the Local Mobility Anchors by deploying MLD proxy functions at 18 Mobile Access Gateways, via a direct traffic distribution within an 19 ISP's access network, or by selective route optimization schemes. 20 This document describes a base solution and an experimental protocol 21 to support mobile multicast senders in Proxy Mobile IPv6 domains for 22 all three scenarios. Protocol optimizations for synchronizing PMIPv6 23 with PIM, as well as a peering function for MLD Proxies are defined. 24 Mobile sources always remain agnostic of multicast mobility 25 operations. 27 Requirements Language 29 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 30 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 31 document are to be interpreted as described in RFC 2119 [RFC2119]. 33 Status of This Memo 35 This Internet-Draft is submitted in full conformance with the 36 provisions of BCP 78 and BCP 79. 38 Internet-Drafts are working documents of the Internet Engineering 39 Task Force (IETF). Note that other groups may also distribute 40 working documents as Internet-Drafts. The list of current Internet- 41 Drafts is at http://datatracker.ietf.org/drafts/current/. 43 Internet-Drafts are draft documents valid for a maximum of six months 44 and may be updated, replaced, or obsoleted by other documents at any 45 time. It is inappropriate to use Internet-Drafts as reference 46 material or to cite them other than as "work in progress." 48 This Internet-Draft will expire on October 2, 2014. 50 Copyright Notice 52 Copyright (c) 2014 IETF Trust and the persons identified as the 53 document authors. All rights reserved. 55 This document is subject to BCP 78 and the IETF Trust's Legal 56 Provisions Relating to IETF Documents 57 (http://trustee.ietf.org/license-info) in effect on the date of 58 publication of this document. Please review these documents 59 carefully, as they describe your rights and restrictions with respect 60 to this document. Code Components extracted from this document must 61 include Simplified BSD License text as described in Section 4.e of 62 the Trust Legal Provisions and are provided without warranty as 63 described in the Simplified BSD License. 65 Table of Contents 67 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 68 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 69 3. Base Solution for Source Mobility and PMIPv6 Routing . . . . 4 70 3.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . 4 71 3.2. Base Solution for Source Mobility: Details . . . . . . . 8 72 3.2.1. Operations of the Mobile Node . . . . . . . . . . . . 8 73 3.2.2. Operations of the Mobile Access Gateway . . . . . . . 8 74 3.2.3. Operations of the Local Mobility Anchor . . . . . . . 8 75 3.2.4. IPv4 Support . . . . . . . . . . . . . . . . . . . . 9 76 3.2.5. Efficiency of the Distribution System . . . . . . . . 10 77 4. Direct Multicast Routing . . . . . . . . . . . . . . . . . . 11 78 4.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . 11 79 4.2. MLD Proxies at MAGs . . . . . . . . . . . . . . . . . . . 12 80 4.2.1. Considerations for PIM-SM on the Upstream . . . . . . 13 81 4.2.2. SSM Considerations . . . . . . . . . . . . . . . . . 13 82 4.3. PIM-SM at MAGs . . . . . . . . . . . . . . . . . . . . . 13 83 4.3.1. Routing Information Base for PIM-SM . . . . . . . . . 13 84 4.3.2. Operations of PIM in Phase One (RP Tree) . . . . . . 14 85 4.3.3. Operations of PIM in Phase Two (Register-Stop) . . . 15 86 4.3.4. Operations of PIM in Phase Three (Shortest-Path Tree) 15 87 4.3.5. PIM-SSM Considerations . . . . . . . . . . . . . . . 16 88 4.3.6. Handover Optimizations for PIM . . . . . . . . . . . 16 89 4.4. BIDIR-PIM . . . . . . . . . . . . . . . . . . . . . . . . 17 90 4.4.1. Routing Information Base for BIDIR-PIM . . . . . . . 17 91 4.4.2. Operations of BIDIR-PIM . . . . . . . . . . . . . . . 17 92 5. MLD Proxy Peering Function for Optimized Source Mobility in 93 PMIPv6 . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 94 5.1. Requirements . . . . . . . . . . . . . . . . . . . . . . 18 95 5.2. Overview . . . . . . . . . . . . . . . . . . . . . . . . 18 96 5.3. Operations in Support of Multicast Senders . . . . . . . 19 97 5.4. Operations in Support of Multicast Listeners . . . . . . 19 99 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 21 100 7. Security Considerations . . . . . . . . . . . . . . . . . . . 21 101 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 22 102 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 22 103 9.1. Normative References . . . . . . . . . . . . . . . . . . 22 104 9.2. Informative References . . . . . . . . . . . . . . . . . 23 105 Appendix A. Multiple Upstream Interface Proxy . . . . . . . . . 24 106 A.1. Operations for Local Multicast Sources . . . . . . . . . 24 107 A.2. Operations for Local Multicast Subscribers . . . . . . . 24 108 Appendix B. Implementation . . . . . . . . . . . . . . . . . . . 25 109 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 25 111 1. Introduction 113 Proxy Mobile IPv6 (PMIPv6) [RFC5213] extends Mobile IPv6 (MIPv6) 114 [RFC6275] by network-based management functions that enable IP 115 mobility for a host without requiring its participation in any 116 mobility-related signaling. Additional network entities called the 117 Local Mobility Anchor (LMA), and Mobile Access Gateways (MAGs), are 118 responsible for managing IP mobility on behalf of the mobile node 119 (MN). An MN connected to a PMIPv6 domain, which only operates 120 according to the base specifications of [RFC5213], cannot participate 121 in multicast communication, as MAGs will discard group packets. 123 Multicast support for mobile listeners can be enabled within a PMIPv6 124 domain by deploying MLD proxy functions at Mobile Access Gateways, 125 and multicast routing functions at Local Mobility Anchors [RFC6224]. 126 This base deployment option is the simplest way to PMIPv6 multicast 127 extensions in the sense that it follows the common PMIPv6 traffic 128 model and neither requires new protocol operations nor additional 129 infrastructure entities. Standard software functions need to be 130 activated on PMIPv6 entities, only, at the price of possibly non- 131 optimal multicast routing. 133 Alternate solutions leverage performance optimization by providing 134 multicast routing at the access gateways directly 135 [I-D.ietf-multimob-fmipv6-pfmipv6-multicast], or by selective route 136 optimization schemes [RFC7028]. Such approaches (partially) follow 137 the model of providing multicast data services in parallel to PMIPv6 138 unicast routing [I-D.ietf-multimob-handover-optimization]. 140 Multicast listener support satisfies the needs of receptive use cases 141 such as IPTV or server-centric gaming on mobiles. However, current 142 trends in the Internet develop towards user-centric, highly 143 interactive group applications like user generated streaming, 144 conferencing, collective mobile sensing, etc. Many of these popular 145 applications create group content at end systems and can largely 146 profit from a direct data transmission to a multicast-enabled 147 network. 149 This document describes the support of mobile multicast senders in 150 Proxy Mobile IPv6 domains subsequently for the base deployment 151 scenario [RFC6224], for direct traffic distribution within an ISP's 152 access network, as well as for selective route optimization schemes. 153 The contribution of this work reflects the source mobility problem as 154 discussed in [RFC5757]. Mobile Nodes in this setting remain agnostic 155 of multicast mobility operations. 157 2. Terminology 159 This document uses the terminology as defined for the mobility 160 protocols [RFC6275], [RFC5213] and [RFC5844], as well as the 161 multicast routing [RFC4601] and edge related protocols [RFC3376], 162 [RFC3810] and [RFC4605]. 164 Throught this document, we use the following acronyms 166 HNP Home Network Prefix as defined in [RFC5213]. 168 MAG Mobile Access Gateway as defined in [RFC5213] 170 MLD Multicast Listener Discovery as defined in [RFC2710] and 171 [RFC3810]. 173 PIM Protocol Independent Multicast as defined in [RFC4601]. 175 PMIPv6 Proxy Mobile IPv6 as defined in [RFC5213]. 177 3. Base Solution for Source Mobility and PMIPv6 Routing 179 3.1. Overview 181 The reference scenario for multicast deployment in Proxy Mobile IPv6 182 domains is illustrated in Figure 1. It displays the general setting 183 for source mobility - Mobile Nodes (MNs) with Home Network Prefixes 184 (HNPs) that receive services via tunnels, which are spanned between a 185 Local Mobility Anchor Address (LMAA) and a Proxy Care-of-Address 186 (Proxy-CoA) at a Mobility Access Gateway (MAG). MAGs play the role 187 of first-hop access routers that serve multiple MNs on the downstream 188 while running an MLD/IGMP proxy instance for every LMA upstream 189 tunnel. 191 +-------------+ 192 | Multicast | 193 | Listeners | 194 +-------------+ 195 | 196 *** *** *** *** 197 * ** ** ** * 198 * * 199 * Fixed Internet * 200 * * 201 * ** ** ** * 202 *** *** *** *** 203 / \ 204 +----+ +----+ 205 |LMA1| |LMA2| Multicast Anchor 206 +----+ +----+ 207 LMAA1 | | LMAA2 208 | | 209 \\ //\\ 210 \\ // \\ 211 \\ // \\ Unicast Tunnel 212 \\ // \\ 213 \\ // \\ 214 \\ // \\ 215 Proxy-CoA1 || || Proxy-CoA2 216 +----+ +----+ 217 |MAG1| |MAG2| MLD Proxy 218 +----+ +----+ 219 | | | 220 MN-HNP1 | | MN-HNP2 | MN-HNP3 221 | | | 222 MN1 MN2 MN3 224 Multicast Sender + Listener(s) 226 Figure 1: Reference Network for Multicast Deployment in PMIPv6 228 An MN in a PMIPv6 domain will decide on multicast data transmission 229 completely independent of its current mobility conditions. It will 230 send packets as initiated by applications, using its source address 231 with Home Network Prefix (HNP) and a multicast destination address 232 chosen by application needs. Multicast packets will arrive at the 233 currently active MAG via one of its downstream local (wireless) 234 links. A multicast unaware MAG would simply discard these packets in 235 the absence of instructions for packet processing, i.e., a multicast 236 routing information base (MRIB). 238 An MN can successfully distribute multicast data in PMIPv6, if MLD 239 proxy functions are deployed at the MAG as described in [RFC6224]. 240 In this set-up, the MLD proxy instance serving a mobile multicast 241 source has configured its upstream interface at the tunnel towards 242 MN's corresponding LMA. For each LMA, there will be a separate 243 instance of an MLD proxy. 245 According to the specifications given in [RFC4605], multicast data 246 arriving from a downstream interface of an MLD proxy will be 247 forwarded to the upstream interface and to all but the incoming 248 downstream interfaces that have appropriate forwarding states for 249 this group. Thus multicast streams originating from an MN will 250 arrive at the corresponding LMA and directly at all mobile receivers 251 co-located at the same MAG and MLD proxy instance. Serving as the 252 designated multicast router or an additional MLD proxy, the LMA 253 forwards data to the fixed Internet, whenever forwarding states are 254 maintained by multicast routing. If the LMA is acting as another MLD 255 proxy, it will forward the multicast data to its upstream interface, 256 and to downstream interfaces with matching subscriptions, 257 accordingly. 259 In case of a handover, the MN (being unaware of IP mobility) can 260 continue to send multicast packets as soon as network connectivity is 261 re-established. At this time, the MAG has determined the 262 corresponding LMA, and IPv6 unicast address configuration (including 263 PMIPv6 bindings) has been completed. Still multicast packets 264 arriving at the MAG are discarded (if not buffered) until the MAG has 265 completed the following steps. 267 1. The MAG has determined that the MN is admissible to multicast 268 services. 270 2. The MAG has added the new downstream link to the MLD proxy 271 instance with up-link to the corresponding LMA. 273 As soon as the MN's uplink is associated with the corresponding MLD 274 proxy instance, multicast packets are forwarded again to the LMA and 275 eventually to receivers within the PMIP domain (see the call flow in 276 Figure 2). In this way, multicast source mobility is transparently 277 enabled in PMIPv6 domains that deploy the base scenario for 278 multicast. 280 MN1 MAG1 MN2 MAG2 LMA 281 | | | | | 282 | | Mcast Data | | | 283 | |<---------------+ | | 284 | | Mcast Data | | | 285 | Join(G) +================================================>| 286 +--------------> | | | | 287 | Mcast Data | | | | 288 |<---------------+ | | | 289 | | | | | 290 | < Movement of MN 2 to MAG2 & PMIP Binding Update > | 291 | | | | | 292 | | |--- Rtr Sol -->| | 293 | | |<-- Rtr Adv ---| | 294 | | | | | 295 | | | < MLD Proxy Configuration > | 296 | | | | | 297 | | | (MLD Query) | | 298 | | |<--------------+ | 299 | | | Mcast Data | | 300 | | +-------------->| | 301 | | | | Mcast Data | 302 | | | +===============>| 303 | | | | | 304 | | Mcast Data | | | 305 | |<================================================+ 306 | Mcast Data | | | | 307 |<---------------+ | | | 308 | | | | | 310 Figure 2: Call Flow for Group Communication in Multicast-enabled PMIP 312 These multicast deployment considerations likewise apply for mobile 313 nodes that operate with their IPv4 stack enabled in a PMIPv6 domain. 314 PMIPv6 can provide IPv4 home address mobility support [RFC5844]. 315 IPv4 multicast is handled by an IGMP proxy function at the MAG in an 316 analogous way. 318 Following these deployment steps, multicast traffic distribution 319 transparently inter-operates with PMIPv6. It is worth noting that an 320 MN - while being attached to the same MAG as the mobile source, but 321 associated with a different LMA - cannot receive multicast traffic on 322 a shortest path. Instead, multicast streams flow up to the LMA of 323 the mobile source, are transferred to the LMA of the mobile listener 324 and tunneled downwards to the MAG again (see Section 5 for further 325 optimizations). 327 3.2. Base Solution for Source Mobility: Details 329 A support of multicast source mobility in PMIPv6 requires to deploy 330 general multicast functions at PMIPv6 routers and to define their 331 interaction with the PMIPv6 protocol in the following way. 333 3.2.1. Operations of the Mobile Node 335 A Mobile Node willing to send multicast data will proceed as if 336 attached to the fixed Internet. No specific mobility or other 337 multicast related functionalities are required at the MN. 339 3.2.2. Operations of the Mobile Access Gateway 341 A Mobile Access Gateway is required to have MLD proxy instances 342 deployed, one for each tunnel to an LMA, which serves as its unique 343 upstream link (cf., [RFC6224]). On the arrival of an MN, the MAG 344 decides on the mapping of downstream links to a proxy instance and 345 the upstream link to the LMA based on the regular Binding Update List 346 as maintained by PMIPv6 standard operations. When multicast data is 347 received from the MN, the MAG MUST identify the corresponding proxy 348 instance from the incoming interface and forwards multicast data 349 upstream according to [RFC4605]. 351 The MAG MAY apply special admission control to enable multicast data 352 transmission from an MN. It is advisable to take special care that 353 MLD proxy implementations do not redistribute multicast data to 354 downstream interfaces without appropriate subscriptions in place. 356 3.2.3. Operations of the Local Mobility Anchor 358 For any MN, the Local Mobility Anchor acts as the persistent Home 359 Agent and at the same time as the default multicast upstream for the 360 corresponding MAG. It will manage and maintain a multicast 361 forwarding information base for all group traffic arriving from its 362 mobile sources. It SHOULD participate in multicast routing functions 363 that enable traffic redistribution to all adjacent LMAs within the 364 PMIPv6 domain and thereby ensure a continuous receptivity while the 365 source is in motion. 367 3.2.3.1. Local Mobility Anchors Operating PIM 369 Local Mobility Anchors that operate the PIM-SM routing protocol 370 [RFC4601] will require sources to be directly connected for sending 371 PIM registers to the RP. This does not hold in a PMIPv6 domain, as 372 MAGs are routers intermediate to MN and the LMA. In this sense, MNs 373 are multicast sources external to the PIM-SM domain. 375 To mitigate this incompatibility common to all subsidiary MLD proxy 376 domains, the LMA MUST act as a PIM Border Router and activate the 377 Border-bit. In this case, the DirectlyConnected(S) is treated as 378 being TRUE for mobile sources and the PIM-SM forwarding rule "iif == 379 RPF_interface(S)" is relaxed to be TRUE, as the incoming tunnel 380 interface from MAG to LMA is considered as not part of the PIM-SM 381 component of the LMA (see A.1 of [RFC4601] ). 383 In addition, an LMA serving as PIM Designated Router (DR) is 384 connected to MLD proxies via individual IP-tunnel interfaces and will 385 experience changing PIM source states on handover. As the incoming 386 interface connects to a point-to-point link, PIM Assert contention is 387 not active, and incoming interface validation is only performed by 388 Reverse Path Forwarding (RPF) checks. Consequently, a PIM DR SHOULD 389 update incoming source states, as soon as RPF inspection succeeds, 390 i.e., after PMIPv6 forwarding state update. Consequently, PIM 391 routers SHOULD be able to manage these state changes, but some 392 implementations are expected to incorrectly refuse packets until the 393 previous state has timed out. 395 Notably, running BIDIR-PIM [RFC5015] on LMAs remains robust with 396 respect to source location and does not require special 397 configurations or state management for sources. 399 3.2.4. IPv4 Support 401 An MN in a PMIPv6 domain may use an IPv4 address transparently for 402 communication as specified in [RFC5844]. For this purpose, an LMA 403 can register an IPv4-Proxy-CoA in its Binding Cache and the MAG can 404 provide IPv4 support in its access network. Correspondingly, 405 multicast membership management will be performed by the MN using 406 IGMP. For multicast support on the network side, an IGMP proxy 407 function needs to be deployed at MAGs in exactly the same way as for 408 IPv6. [RFC4605] defines IGMP proxy behaviour in full agreement with 409 IPv6/MLD. Thus IPv4 support can be transparently provided following 410 the obvious deployment analogy. 412 For a dual-stack IPv4/IPv6 access network, the MAG proxy instances 413 SHOULD choose multicast signaling according to address configurations 414 on the link, but MAY submit IGMP and MLD queries in parallel, if 415 needed. It should further be noted that the infrastructure cannot 416 identify two data streams as identical when distributed via an IPv4 417 and IPv6 multicast group. Thus duplicate data may be forwarded on a 418 heterogeneous network layer. 420 A particular note is worth giving the scenario of [RFC5845] in which 421 overlapping private address spaces of different operators can be 422 hosted in a PMIP domain by using GRE encapsulation with key 423 identification. This scenario implies that unicast communication in 424 the MAG-LMA tunnel can be individually identified per MN by the GRE 425 keys. This scenario still does not impose any special treatment of 426 multicast communication for the following reasons. 428 Multicast streams from and to MNs arrive at a MAG on point-to-point 429 links (identical to unicast). Multicast data transmission from the 430 MAG to the corresponding LMA is link-local between the routers and 431 routing/forwarding remains independent of any individual MN. So the 432 MAG-proxy and the LMA SHOULD NOT use GRE key identifiers, but plain 433 GRE encapsulation in multicast communication (including MLD queries 434 and reports). Multicast traffic is transmitted as router-to-router 435 forwarding via the MAG-to-LMA tunnels and according to the multicast 436 routing information base (MRIB) of the MAG or the LMA. It remains 437 independent of MN's unicast addresses, while the MAG proxy instance 438 redistributes multicast data down the point-to-point links 439 (interfaces) according to its local subscription states, independent 440 of IP addresses of the MN. 442 3.2.5. Efficiency of the Distribution System 444 The distribution system of the base solution directly follows PMIPv6 445 routing rules, and organizes multicast domains with respect to LMAs. 446 Thus, no coordination between address spaces or services is required 447 between the different instances, provided their associated LMAs 448 belong to disjoint multicast domains. Routing is optimal for 449 communication between MNs of the same domain, or stationary 450 subscribers. 452 In the following, efficiency-related issues remain. 454 Multicast reception at LMA In the current deployment scenario, the 455 LMA will receive all multicast traffic originating from its 456 associated MNs. There is no mechanism to suppress upstream 457 forwarding in the absence of receivers. 459 MNs on the same MAG using different LMAs For a mobile receiver and a 460 source that use different LMAs, the traffic has to go up to one 461 LMA, cross over to the other LMA, and then be tunneled back to the 462 same MAG, causing redundant flows in the access network and at the 463 MAG. 465 These remaining deficits in routing efficiency can be resolved by 466 adding peering functions to MLD proxies as described in Section 5. 468 4. Direct Multicast Routing 470 There are deployment scenarios, where multicast services are 471 available throughout the access network independent of the PMIPv6 472 routing system [RFC7028]. In these cases, the visited networks grant 473 a local content distribution service (in contrast to LMA-based home 474 subscription) with locally optimized traffic flows. It is also 475 possible to deploy a mixed service model of local and LMA-based 476 subscriptions, provided a unique way of service selection is 477 implemented. For example, access routers (MAGs) could decide on 478 service access based on the multicast address G or the SSM channel 479 (S,G) under request (see Appendix A for further discussions). 481 4.1. Overview 483 Direct multicast access can be supported by 485 o native multicast routing provided by one multicast router that is 486 neighboring MLD proxies deployed at MAGs within a flat access 487 network, or via tunnel uplinks, 489 o a multicast routing protocol such as PIM-SM [RFC4601] or BIDIR-PIM 490 [RFC5015] deployed at the MAGs. 492 *** *** *** *** 493 * ** ** ** * 494 * * 495 * Multicast * 496 +----+ * Infrastructure * +----+ 497 |LMA | * ** ** ** * |LMA | 498 +----+ *** *** *** *** +----+ 499 | // \\ | 500 \\ // \\ PMIP (unicast) | 501 PMIP \\ // \\ // \\ ** *** *** ** // 502 (unicast) \\ // \\ // \\ * ** ** ** // 503 \\ // \\ // \\* Multicast *// 504 || || || || * || Routing || * 505 +----+ +----+ * +----+ +----+ * 506 MLD Proxy |MAG1| |MAG2| * |MAG1| |MAG2| * 507 +----+ +----+ *+----+ ** ** +----+* 508 | | | | |*** *** ***| 509 | | | | | | 510 MN1 MN2 MN3 MN1 MN2 MN3 512 (a) Multicast Access at Proxy Uplink (b) Multicast Routing at MAG 514 Figure 3: Reference Networks for (a) Proxy-assisted Direct Multicast 515 Access and (b) Dynamic Multicast Routing at MAGs 517 Figure 3 displays the corresponding deployment scenarios, which 518 separate multicast from PMIPv6 unicast routing. It is assumed 519 throughout these scenarios that all MAGs (MLD proxies) are linked to 520 a single multicast routing domain. Notably, this scenario requires 521 coordination of multicast address utilization and service bindings. 523 Multicast traffic distribution can be simplified in these scenarios. 524 A single proxy instance at MAGs with up-link into the multicast 525 domain will serve as a first hop multicast gateway and avoid traffic 526 duplication or detour routing. Multicast routing functions at MAGs 527 will seamlessly embed access gateways within a multicast cloud. 528 However, mobility of the multicast source in this scenario will 529 require some multicast routing protocols to rebuild distribution 530 trees. This can cause significant service disruptions or delays (see 531 [RFC5757] for further aspects). Deployment details are specific to 532 the multicast routing protocol in use, in the following described for 533 common protocols. 535 4.2. MLD Proxies at MAGs 537 In a PMIPv6 domain, single MLD proxy instances can be deployed at 538 each MAG that enable multicast service at the access via an uplink to 539 a multicast service infrastructure (see Figure 3 (a) ). To avoid 540 service disruptions on handovers, the uplinks of all proxies SHOULD 541 be adjacent to the same next-hop multicast router. This can either 542 be achieved by arranging proxies within a flat access network, or by 543 upstream tunnels that terminate at a common multicast router. 545 Multicast data submitted by a mobile source will reach the MLD proxy 546 at the MAG that subsequently forwards flows to the upstream, and all 547 downstream interfaces with appropriate subscriptions. Traversing the 548 upstream will transfer traffic into the multicast infrastructure 549 (e.g., to a PIM Designated Router) which will route packets to all 550 local MAGs that have joined the group, as well as further upstream 551 according to protocol procedures and forwarding states. 553 On handover, a mobile source will reattach to a new MAG and can 554 continue to send multicast packets as soon as PMIPv6 unicast 555 configurations have been completed. Like at the previous MAG, the 556 new MLD proxy will forward data upstream and downstream to 557 subscribers. Listeners local to the previous MAG will continue to 558 receive group traffic via the local multicast distribution 559 infrastructure following aggregated listener reports of the previous 560 proxy. In general, traffic from the mobile source continues to be 561 transmitted via the same next-hop multicast router using the same 562 source address and thus remains unchanged when seen from the wider 563 multicast infrastructure. 565 4.2.1. Considerations for PIM-SM on the Upstream 567 A mobile source that transmits data via an MLD proxy will not be 568 directly connected to a PIM Designated Router as discussed in 569 Section 3.2.3.1. Countermeasures apply correspondingly. 571 A PIM Designated Router that is connected to MLD proxies via 572 individual IP-tunnel interfaces will experience invalid PIM source 573 states on handover. In some implementations of PIM-SM this could 574 lead to an interim packet loss (see Section 3.2.3.1). This problem 575 can be mitigated by aggregating proxies on a lower layer. 577 4.2.2. SSM Considerations 579 Source-specific subscriptions invalidate with routes, whenever the 580 source moves from or to the MAG/proxy of a subscriber. Multicast 581 forwarding states will rebuild with unicast route changes. However, 582 this may lead to noticeable service disruptions for locally 583 subscribed nodes. 585 4.3. PIM-SM at MAGs 587 The full-featured multicast routing protocol PIM-SM MAY be deployed 588 in the access network for providing multicast services in parallel to 589 unicast routes (see Figure 3 b). Throughout this section, it is 590 assumed that the PMIPv6 mobility domain is part of a single PIM-SM 591 multicast routing domain with PIM-SM routing functions present at all 592 MAGs and all LMAs. The PIM routing instance at a MAG SHALL then 593 serve as the Designated Router (DR) for all directly attached Mobile 594 Nodes. For expediting handover operations, it is advisable to 595 position PIM Rendezvous Points (RPs) in the core of the PMIPv6 596 network domain. However, regular IP routing tables need not be 597 present in a PMIPv6 deployment, and additional effort is required to 598 establish reverse path forwarding rules as required by PIM-SM. 600 4.3.1. Routing Information Base for PIM-SM 602 In this scenario, PIM-SM will rely on a Multicast Routing Information 603 Base (MRIB) that is generated independently of the policy-based 604 routing rules of PMIPv6. The granularity of mobility-related routing 605 locators required in PIM depends on the complexity (specific phase) 606 of its deployment. 608 For all three phases in the operation of PIM (see [RFC4601]), the 609 following information is needed. 611 o All routes to networks and nodes (including RPs) that are not 612 mobile members of the PMIPv6 domain MUST be defined consistently 613 among PIM routers and MUST remain unaffected by node mobility. 614 The setup of these general routes is expected to follow the 615 topology of the operator network and is beyond the scope of this 616 document. 618 The following route entries are required at a PIM-operating MAG when 619 phases two or three of PIM, or PIM-SSM are in operation. 621 o Local routes to the Home Network Prefixes (HNPs) of all MNs 622 associated with their corresponding point-to-point attachments 623 that MUST be included in the local MRIB. 625 o All routes to MNs that are attached to distant MAGs of the PMIPv6 626 domain point towards their corresponding LMAs. These routes MUST 627 be made available in the MRIB of all PIM routers (except for the 628 local MAG of attachment), but MAY be eventually expressed by an 629 appropriate default entry. 631 4.3.2. Operations of PIM in Phase One (RP Tree) 633 A new mobile source S will transmit multicast data of group G towards 634 its MAG of attachment. Acting as a PIM DR, the access gateway will 635 unicast-encapsulate the multicast packets and forward the data to the 636 Virtual Interface (VI) with encapsulation target RP(G), a process 637 known as PIM source registering. The RP will decapsulate and 638 natively forward the packets down the RP-based distribution tree 639 towards (mobile and stationary) subscribers. 641 On handover, the point-to-point link connecting the mobile source to 642 the old MAG will go down and all (S,*) flows terminate. In response, 643 the previous DR (MAG) deactivates the data encapsulation channels for 644 the transient source (i.e., all DownstreamJPState(S,*,VI) are set to 645 NoInfo state). After reattaching and completing unicast handover 646 negotiations, the mobile source can continue to transmit multicast 647 packets, while being treated as a new source at its new DR (MAG). 648 Source register encapsulation will be immediately initiated, and 649 (S,G) data continue to flow natively down the (*,G) RP-based tree. 651 Source handover management in PIM phase one admits low complexity and 652 remains transparent to receivers. In addition, the source register 653 tunnel management of PIM is a fast protocol operation that introduces 654 little overhead. In a PMIPv6 deployment, PIM RPs MAY be configured 655 to not initiated (S,G) shortest path trees for mobile sources, and 656 thus remain in phase one of the protocol. The price to pay for such 657 simplified deployment lies in possible routing detours by an overall 658 RP-based packet distribution. 660 4.3.3. Operations of PIM in Phase Two (Register-Stop) 662 After receiving source register packets, a PIM RP eventually will 663 initiate a source-specific Join for creating a shortest path tree to 664 the (mobile) source S, and issue a source register stop at the native 665 arrival of data from S. For initiating an (S,G) tree, the RP, as well 666 as all intermediate routers, require route entries for the HNP of the 667 MN that - unless the RP coincides with the MAG of S - point towards 668 the corresponding LMA of S. Consequently, the (S,G) tree will proceed 669 from the RP via the (stable) LMA, down the LMA-MAG tunnel to the 670 mobile source. This tree can be of lower routing efficiency than the 671 PIM source register tunnel established in phase one. 673 On handover, the mobile source reattaches to a new MAG (DR), and 674 PMIPv6 unicast management will transfer the LMA-MAG tunnel to the new 675 point of attachment. However, in the absence of a corresponding 676 multicast forwarding state, the new DR will treat S as a new source 677 and initiate a source registering of PIM phase one with the RP. In 678 response, the PIM RP will recognize the known source at a new 679 (tunnel) interface and immediately responds with a register stop. As 680 the RP had previously joined the shortest path tree towards the 681 source via the LMA, it will see an RPF change when data arrives at a 682 new interface. Implementation-dependent, this can trigger an update 683 of the PIM MRIB and trigger a new PIM Join message that will install 684 the multicast forwarding state missing at the new MAG. Otherwise, 685 the tree is periodically updated by Joins transmitted towards the new 686 MAG on a path via the LMA. In proceeding this way, a quick recovery 687 of PIM transition from phase one to two will be performed per 688 handover. 690 4.3.4. Operations of PIM in Phase Three (Shortest-Path Tree) 692 In response to an exceeded threshold of packet transmission, DRs of 693 receivers eventually will initiate a source-specific Join for 694 creating a shortest path tree to the (mobile) source S, thereby 695 transitioning PIM into the final short-cut phase three. For all 696 receivers not sharing a MAG with S, this (S,G) tree will range from 697 the receiving DR via the (stable) LMA, the LMA-MAG tunnel, and the 698 serving MAG to the mobile source. This tree is of higher routing 699 efficiency than that established in the previous phase two, but need 700 not outperform the PIM source register tunnel established in phase 701 one. It provides the advantage of immediate data delivery to 702 receivers that share a MAG with S. 704 On handover, the mobile source reattaches to a new MAG (DR), and 705 PMIPv6 unicast management will transfer the LMA-MAG tunnel to the new 706 point of attachment. However, in the absence of a corresponding 707 multicast forwarding state, the new DR will treat S as a new source 708 and initiate a source registering of PIM phase one. A PIM 709 implementation compliant with this change can recover phase three 710 states in the following way. First, the RP recovers to phase two as 711 described in the previous section, and will not forward data arriving 712 via the source register tunnel. Tree maintenance eventually 713 triggered by the RPF change (see Section 4.3.3) will generate proper 714 states for a native forwarding from the new MAG via the LMA. 715 Thereafter, packets arriving at the LMA without source register 716 encapsulation are forwarded natively along the shortest path tree 717 towards receivers. 719 In consequence, the PIM transitions from phase one to two and to 720 three will be quickly recovered per handover, but still lead to an 721 enhanced signaling load and intermediate packet loss. 723 4.3.5. PIM-SSM Considerations 725 Source-specific Joins of receivers will guide PIM to operate in SSM 726 mode and lead to an immediate establishment of source-specific 727 shortest path trees. Such (S,G) trees will equal the distribution 728 system of PIM's final phase three (see Section 4.3.4). However, on 729 handover and in the absence of RP-based data distribution, SSM data 730 delivery cannot be resumed via source registering as in PIM phase 731 one. Consequently, data packets transmitted after a handover will be 732 discarded at the MAG until regular tree maintenance has reestablished 733 the (S,G) forwarding state at the new MAG. 735 4.3.6. Handover Optimizations for PIM 737 Source-specific shortest path trees are constructed in PIM-SM (phase 738 two and three), and in PIM-SSM that follow LMA-MAG tunnels towards a 739 source. As PIM remains unaware of source mobility management, these 740 trees invalidate under handovers with each tunnel re-establishment at 741 a new MAG. Regular tree maintenance of PIM will recover the states, 742 but remains unsynchronized and too slow to seamlessly preserve PIM 743 data distribution services. 745 A method to quickly recover PIM (S,G) trees under handover SHOULD 746 synchronize multicast state maintenance with unicast handover 747 operations and can proceed as follows. On handover, an LMA reads all 748 (S,G) Join states from its corresponding tunnel interface and 749 identifies those source addresses S_i that match moving HNPs. After 750 re-establishing the new tunnel, it SHOULD associate the (S_i,*) Join 751 states with the new tunnel endpoint and immediately trigger a state 752 maintenance (PIM Join) message. In proceeding this way, the source- 753 specific PIM states are transferred to the new tunnel end point and 754 propagated to the new MAG in synchrony with unicast handover 755 procedures. 757 4.4. BIDIR-PIM 759 BIDIR-PIM MAY be deployed in the access network for providing 760 multicast services in parallel to unicast routes. Throughout this 761 section, it is assumed that the PMIPv6 mobility domain is part of a 762 single BIDIR-PIM multicast routing domain with BIDIR-PIM routing 763 functions present at all MAGs and all LMAs. The PIM routing instance 764 at a MAG SHALL then serve as the Designated Forwarder (DF) for all 765 directly attached Mobile Nodes. For expediting handover operations, 766 it is advisable to position BIDIR-PIM Rendezvous Point Addresses 767 (RPAs) in the core of the PMIPv6 network domain. As regular IP 768 routing tables need not be present in a PMIPv6 deployment, reverse 769 path forwarding rules as required by BIDIR-PIM need to be 770 established. 772 4.4.1. Routing Information Base for BIDIR-PIM 774 In this scenario, BIDIR-PIM will rely on a Multicast Routing 775 Information Base (MRIB) that is generated independently of the 776 policy-based routing rules of PMIPv6. The following information is 777 needed. 779 o All routes to networks and nodes (including RPAs) that are not 780 mobile members of the PMIPv6 domain MUST be defined consistently 781 among BIDIR-PIM routers and remain unaffected by node mobility. 782 The setup of these general routes is expected to follow the 783 topology of the operator network and is beyond the scope of this 784 document. 786 4.4.2. Operations of BIDIR-PIM 788 BIDIR-PIM will establish spanning trees across its network domain in 789 conformance to its pre-configured RPAs and the routing information 790 provided. Multicast data transmitted by a mobile source will 791 immediately be forwarded by its DF (MAG) onto the spanning tree for 792 the multicast group without further protocol operations. 794 On handover, the mobile source reattaches to a new MAG (DF), which 795 completes unicast network configurations. Thereafter, the source can 796 immediately proceed with multicast packet transmission onto the pre- 797 established distribution tree. BIDIR-PIM does neither require 798 protocol signaling nor additional reconfiguration delays to adapt to 799 source mobility and can be considered the protocol of choice for 800 mobile multicast operations in the access. As multicast streams 801 always flow up to the Rendezvous Point Link, some care should be 802 taken to configure RPAs compliant with network capacities. 804 5. MLD Proxy Peering Function for Optimized Source Mobility in PMIPv6 806 A deployment of MLD Proxies (see [RFC4605]) at MAGs has proven a 807 useful and appropriate approach to multicast in PMIPv6, see 808 [RFC6224], [RFC7028]. However, deploying unmodified standard proxies 809 can go along with significant performance degradation for mobile 810 senders as discussed along the lines of this document. To overcome 811 these deficits, an optimized approach to multicast source mobility 812 based on extended peering functions among proxies is defined in this 813 section. Based on such direct data exchange between proxy instances 814 at MAGs, triangular routing is avoided and multicast streams can be 815 disseminated directly within a PMIPv6 access network, and in 816 particular within MAG routing machines. Prior to presenting the 817 solution, we will summarize the relevant requirements. 819 5.1. Requirements 821 Solutions that extend MLD Proxies by additional uplinking functions 822 need to comply to the following requirements. 824 Prevention of Routing Loops In the absence of a full-featured 825 routing logic at an MLD Proxy, simple and locally decidable rules 826 need to prevent source traffic from traversing the network in 827 loops as potentially enabled by multiple uplinks. 829 Unique coverage of receivers Listener functions at Proxies require 830 simple, locally decidable rules to initiate a unique delivery of 831 multicast packets to all receivers. 833 Following local filter techniques, these requirements are met in the 834 following solution. 836 5.2. Overview 838 A peering interface for MLD proxies allows for a direct data exchange 839 of locally attached multicast sources. Such peering interfaces can 840 be configured - as a direct link or a bidirectional tunnel - between 841 any two proxy instances (locally deployed as in [RFC6224] or 842 remotely). Peerings remain as silent virtual links in regular proxy 843 operations. Data is exchanged on such links only in cases, where one 844 peering proxy on its downstream directly connects to a source of 845 multicast traffic, which the other peering proxy actively subscribes 846 to. In such cases, the proxy connected to the source will receive a 847 listener report on its peering interface and forwards traffic from 848 its local source accordingly. It is worth noting that multicast 849 traffic distribution on peering links does not follow reverse unicast 850 paths to sources. In the following, operations are defined for ASM 851 and SSM, but provide superior performance in the presence of source- 852 specific signaling (IGMPv3/MLDv2) [RFC4604]. 854 5.3. Operations in Support of Multicast Senders 856 An MLD proxy in the perspective of a sender will see peering 857 interfaces as restricted downstream interfaces. It will install and 858 maintain source filters at its peering links that will restrict data 859 transmission to those packets that originate from a source that is 860 locally attached at one of its downstream interfaces. 862 In detail, a proxy will extract from its configuration the network 863 prefixes attached to its downstream interfaces and MUST implement a 864 source filter base at its peering interfaces that restricts data 865 transmission to IP source addresses from its local prefixes. This 866 filter base MUST be updated, if and only if the downstream 867 configuration changes (e.g., due to mobility). Multicast packets 868 that arrive from the upstream interface of the proxy are thus 869 prevented from traversing any peering link, but are only forwarded to 870 regular downstream interfaces with appropriate subscription states. 871 In this way, a multihop forwarding on peering links is prevented. 873 Multicast traffic arriving from a locally attached source will be 874 forwarded to the regular upstream interface and all downstreams with 875 appropriate subscription states (i.e., regular proxy operations). In 876 addition, multicast packets of local origin are transferred to those 877 peering interfaces with appropriate subscription states. 879 5.4. Operations in Support of Multicast Listeners 881 At the listener side, peering interfaces appear as preferred upstream 882 links. The multicast proxy will attempt to receive multicast 883 services on peering links for as many groups (channels) as possible. 884 The general upstream interface configured according to [RFC4605] will 885 be used only for retrieving those groups (channels) that remain 886 unavailable from peerings. From this extension of [RFC4605], an MLD 887 proxy with peering interconnects will exhibit several interfaces for 888 pulling remote traffic: the regular upstream and the peerings. 889 Traffic available from any of the peering links will be mutually 890 disjoint, but normally also available from the upstream. To prevent 891 duplicate traffic from arriving at the listener side, the proxy 893 o MAY delay aggregated reports to the upstream, and 895 o MUST apply appropriate filters to exclude duplicate streams. 897 In detail, an MLD proxy instance at a MAG first issues listener 898 reports (in parallel) to all of its peering links. These links span 899 at most one (virtual) hop. Whenever certain group traffic (SSM 900 channels) does not arrive from the peerings after a waiting time 901 (default: 10 ms (node-local) and 25 ms (remote)), additional 902 (complementary, in the case of SSM) reports are sent to the standard 903 upstream interface. 905 Whenever traffic from a peering link arrives, an MLD proxy MUST 906 install source filters at its RFC 4605 upstream in the following way. 908 ASM with IGMPv2/MLDv1 In the presence of Any Source Multicast using 909 IGMPv2/MLDv1, only, the proxy cannot signal source filtering to 910 its upstream. Correspondingly, it applies (S,*) ingress filters 911 at its upstream interface for all sources S seen in traffic on the 912 peering links. It is noteworthy that unwanted traffic is still 913 replicated to the proxy via the (wired) provider backbone, but it 914 is not forwarded into the wireless access network. 916 ASM with IGMPv3/MLDv2 In the presence of source-specific signaling 917 (IGMPv3/MLDv2), the upstream interface is set to (S,*) exclude 918 mode for all sources S seen in traffic of the peering links. The 919 corresponding source-specific signaling will prevent forwarding of 920 duplicate traffic throughout the access network. 922 SSM In the presence of Source Specific Multicast, the proxy will 923 subscribe on its uplink interface to those (S,G) channels, only, 924 that do not arrive via the peering links. 926 MLD proxies will install data-driven timers (source-timeout) for each 927 source but common to all peering interfaces to detect interruptions 928 of data services from individual sources at proxy peers. Termination 929 of source-specific flows may be application-specific, but also due to 930 a source handover, or transmission failures. After a handover, a 931 mobile source may reattach at another MLD proxy with peering relation 932 to the listener, or at a proxy that does not peer. While in the 933 first case, traffic reappears on another peering link, in the second 934 case data can only be retrieved via the regular upstream. To account 935 for the latter, the MLD proxy revokes the source-specific filter(s) 936 at its upstream interface, after the source-timeout fires (default: 937 50 ms). Corresponding traffic will then be pulled from the regular 938 upstream. Source-specific filters MUST be reinstalled, whenever 939 traffic of this source arrives at any peering interface. 941 There is a noteworthy trade-off between traffic minimization and 942 available traffic at the upstream that is locally filtered at the 943 proxy. Implementors can use this relation to optimize for service- 944 specific requirements. 946 In proceeding this way, multicast group data will arrive from peering 947 interfaces first, while only peer-wise unavailable traffic is 948 retrieved from the regular upstream interface. 950 6. IANA Considerations 952 This document makes no request to IANA.. 954 Note to RFC Editor: this section may be removed on publication as an 955 RFC. 957 7. Security Considerations 959 This document defines multicast sender mobility based on PMIPv6 and 960 common multicast routing protocols. Consequently, threats identified 961 as security concerns of [RFC2236], [RFC2710], , [RFC3810], [RFC4605], 962 [RFC5213], and [RFC5844] are inherited by this document. 964 In addition, particular attention should be paid to implications of 965 combining multicast and mobility management at network entities. As 966 this specification allows mobile nodes to initiate the creation of 967 multicast forwarding states at MAGs and LMAs while changing 968 attachments, threats of resource exhaustion at PMIP routers and 969 access networks arrive from rapid state changes, as well as from high 970 volume data streams routed into access networks of limited 971 capacities. In cases of PIM-SM deployment, handover operations of 972 the MNs include re-registering sources at the Rendezvous Points at 973 possibly high frequency. In addition to proper authorization checks 974 of MNs, rate controls at routing agents and replicators may be needed 975 to protect the agents and the downstream networks. In particular, 976 MLD proxy implementations at MAGs SHOULD automatically extinct 977 multicast state on the departure of MNs, as mobile multicast 978 listeners in the PMIPv6 domain will in general not actively terminate 979 group membership prior to departure. 981 The deployment of IGMP/MLD proxies for multicast routing requires 982 particular care, as routing loops on the upstream are not 983 automatically detected. Peering functions between proxies extend 984 this threat in the following way. Routing loops among peering and 985 upstream interfaces are prevented by filters on local sources. Such 986 filtering can fail, whenever prefix configurations for downstream 987 interfaces at a proxy are incorrect or inconsistent. Consequently, 988 implementations of peering-enabled proxies SHOULD take particular 989 care on keeping IP configurations consistent at the downstream in a 990 reliable and timely manner (see [RFC6224] for requirements on 991 PMIPv6-compliant implementations of MLD proxies). 993 8. Acknowledgements 995 The authors would like to thank (in alphabetical order) David Black, 996 Luis M. Contreras, Spencer Dawkins, Muhamma Omer Farooq, Bohao Feng, 997 Sri Gundavelli, Dirk von Hugo, Ning Kong, Jouni Korhonen, He-Wu Li, 998 Cong Liu, Radia Perlman, Akbar Rahman, Behcet Sarikaya, Stig Venaas, 999 Li-Li Wang, Sebastian Woelke, Qian Wu, Zhi-Wei Yan for advice, help 1000 and reviews of the document. Funding by the German Federal Ministry 1001 of Education and Research within the G-LAB Initiative (projects 1002 HAMcast, Mindstone and SAFEST) is gratefully acknowledged. 1004 9. References 1006 9.1. Normative References 1008 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1009 Requirement Levels", BCP 14, RFC 2119, March 1997. 1011 [RFC2710] Deering, S., Fenner, W., and B. Haberman, "Multicast 1012 Listener Discovery (MLD) for IPv6", RFC 2710, October 1013 1999. 1015 [RFC3376] Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A. 1016 Thyagarajan, "Internet Group Management Protocol, Version 1017 3", RFC 3376, October 2002. 1019 [RFC3810] Vida, R. and L. Costa, "Multicast Listener Discovery 1020 Version 2 (MLDv2) for IPv6", RFC 3810, June 2004. 1022 [RFC4601] Fenner, B., Handley, M., Holbrook, H., and I. Kouvelas, 1023 "Protocol Independent Multicast - Sparse Mode (PIM-SM): 1024 Protocol Specification (Revised)", RFC 4601, August 2006. 1026 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 1027 "Internet Group Management Protocol (IGMP) / Multicast 1028 Listener Discovery (MLD)-Based Multicast Forwarding ("IGMP 1029 /MLD Proxying")", RFC 4605, August 2006. 1031 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 1032 "Bidirectional Protocol Independent Multicast (BIDIR- 1033 PIM)", RFC 5015, October 2007. 1035 [RFC5213] Gundavelli, S., Leung, K., Devarapalli, V., Chowdhury, K., 1036 and B. Patil, "Proxy Mobile IPv6", RFC 5213, August 2008. 1038 [RFC5844] Wakikawa, R. and S. Gundavelli, "IPv4 Support for Proxy 1039 Mobile IPv6", RFC 5844, May 2010. 1041 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 1042 in IPv6", RFC 6275, July 2011. 1044 9.2. Informative References 1046 [I-D.ietf-multimob-fmipv6-pfmipv6-multicast] 1047 Schmidt, T., Waehlisch, M., Koodli, R., Fairhurst, G., and 1048 D. Liu, "Multicast Listener Extensions for MIPv6 and 1049 PMIPv6 Fast Handovers", draft-ietf-multimob- 1050 fmipv6-pfmipv6-multicast-05 (work in progress), March 1051 2014. 1053 [I-D.ietf-multimob-handover-optimization] 1054 Contreras, L., Bernardos, C., and I. Soto, "PMIPv6 1055 multicast handover optimization by the Subscription 1056 Information Acquisition through the LMA (SIAL)", draft- 1057 ietf-multimob-handover-optimization-07 (work in progress), 1058 December 2013. 1060 [Peering-Analysis] 1061 Schmidt, TC., Woelke, S., and M. Waehlisch, "Peer my Proxy 1062 - A Performance Study of Peering Extensions for Multicast 1063 in Proxy Mobile IP Domains", Proc. of 7th IFIP Wireless 1064 and Mobile Networking Conference (WMNC 2014) IEEEPress, 1065 May 2014. 1067 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 1068 2", RFC 2236, November 1997. 1070 [RFC4604] Holbrook, H., Cain, B., and B. Haberman, "Using Internet 1071 Group Management Protocol Version 3 (IGMPv3) and Multicast 1072 Listener Discovery Protocol Version 2 (MLDv2) for Source- 1073 Specific Multicast", RFC 4604, August 2006. 1075 [RFC5757] Schmidt, T., Waehlisch, M., and G. Fairhurst, "Multicast 1076 Mobility in Mobile IP Version 6 (MIPv6): Problem Statement 1077 and Brief Survey", RFC 5757, February 2010. 1079 [RFC5845] Muhanna, A., Khalil, M., Gundavelli, S., and K. Leung, 1080 "Generic Routing Encapsulation (GRE) Key Option for Proxy 1081 Mobile IPv6", RFC 5845, June 2010. 1083 [RFC6224] Schmidt, T., Waehlisch, M., and S. Krishnan, "Base 1084 Deployment for Multicast Listener Support in Proxy Mobile 1085 IPv6 (PMIPv6) Domains", RFC 6224, April 2011. 1087 [RFC7028] Zuniga, JC., Contreras, LM., Bernardos, CJ., Jeon, S., and 1088 Y. Kim, "Multicast Mobility Routing Optimizations for 1089 Proxy Mobile IPv6", RFC 7028, September 2013. 1091 Appendix A. Multiple Upstream Interface Proxy 1093 In this section, we document upstream extensions for an MLD proxy 1094 that were originally developed during the work on this document. 1095 Multiple proxy instances deployed at a single MAG (see Section 3) can 1096 be avoided by adding multiple upstream interfaces to a single MLD 1097 Proxy. In a typical PMIPv6 deployment, each upstream of a single 1098 proxy instance can interconnect to one of the LMAs. With such 1099 ambiguous upstream options, appropriate forwarding rules MUST be 1100 supplied to 1102 o unambiguously guide traffic forwarding from directly attached 1103 mobile sources, and 1105 o lead listener reports to initiating unique traffic subscriptions. 1107 This can be achieved by a complete set of source- and group-specific 1108 filter rules (e.g., (S,*), (*,G)) installed at proxy interfaces. 1109 These filters MAY be derived in parts from PMIPv6 routing policies, 1110 and can include a default behavior (e.g., (*,*)). 1112 A.1. Operations for Local Multicast Sources 1114 Packets from a locally attached multicast source will be forwarded to 1115 all downstream interfaces with appropriate subscriptions, as well as 1116 up the interface with the matching source-specific filter. 1118 Typically, the upstream interface for a mobile multicast source is 1119 chosen based on the policy routing (e.g., the MAG-LMA tunnel 1120 interface for LMA-based routing or the interface towards the 1121 multicast router for direct routing), but alternate configurations 1122 MAY be applied. Packets from a locally attached multicast source 1123 will be forwarded to the corresponding upstream interface with the 1124 matching source-specific filter, as well as all the downstream 1125 interfaces with appropriate subscriptions. 1127 A.2. Operations for Local Multicast Subscribers 1129 Multicast listener reports are group-wise aggregated by the MLD 1130 proxy. The aggregated report is issued to the upstream interface 1131 with matching group/channel-specific filter. The choice of the 1132 corresponding upstream interface for aggregated group membership 1133 reports MAY be additionally based on some administrative scoping 1134 rules for scoped multicast group addresses. 1136 In detail, a Multiple Upstream Interface proxy will provide and 1137 maintain a Multicast Subscription Filter Table that maps source- and 1138 group-specific filters to upstream interfaces. The forwarding 1139 decision for an aggregated MLD listener report is based on the first 1140 matching entry from this table, with the understanding that for 1141 IGMPv3/MLDv2 the MLD proxy performs a state decomposition, if needed 1142 (i.e., a (*,G) subscription is split into (S,G) and (* \ S,G) in the 1143 presence of (*,G) after (S,G) interface entries), and that 1144 (S,*)-filters are always false in the absence of source-specific 1145 signaling, i.e. in IGMPv2/MLDv1 only domains. 1147 In typical deployment scenarios, specific group services (channels) 1148 could be either associated with selected uplinks to remote LMAs, 1149 while a (*,*) default subscription entry (in the last table line) is 1150 bound to a local routing interface, or selected groups are configured 1151 as local services first, while a (*,*) default entry (in the last 1152 table line) points to a remote uplink that provides the general 1153 multicast support. 1155 Appendix B. Implementation 1157 An implementation of the extended IGMP/MLD proxy has been provided 1158 within the MCPROXY project http://mcproxy.realmv6.org/. This open 1159 source software is written in C++ and uses forwarding capabilities of 1160 the Linux kernel. It supports all regular operations according to 1161 [RFC4605], allows for multiple proxy instances on one node, 1162 dynamically changing downstream links, as well as proxy-to-proxy 1163 peerings and multiple upstream links with individual configurations. 1164 The software can be downloaded from Github at https://github.com/ 1165 mcproxy/mcproxy. Based on this software, an experimental performance 1166 evaluation of the proxy peering function has been reported in 1167 [Peering-Analysis]. 1169 Authors' Addresses 1171 Thomas C. Schmidt (editor) 1172 HAW Hamburg 1173 Berliner Tor 7 1174 Hamburg 20099 1175 Germany 1177 Email: schmidt@informatik.haw-hamburg.de 1178 URI: http://inet.cpt.haw-hamburg.de/members/schmidt 1179 Shuai Gao 1180 Beijing Jiaotong University 1181 Beijing 1182 China 1184 Email: shgao@bjtu.edu.cn 1186 Hong-Ke Zhang 1187 Beijing Jiaotong University 1188 Beijing 1189 China 1191 Email: hkzhang@bjtu.edu.cn 1193 Matthias Waehlisch 1194 link-lab & FU Berlin 1195 Hoenower Str. 35 1196 Berlin 10318 1197 Germany 1199 Email: mw@link-lab.net