idnits 2.17.1 draft-ietf-lisp-signal-free-multicast-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 8, 2018) is 2234 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 6830 (Obsoleted by RFC 9300, RFC 9301) ** Obsolete normative reference: RFC 6833 (Obsoleted by RFC 9301) == Outdated reference: A later version (-13) exists of draft-ietf-lisp-eid-mobility-01 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-14 Summary: 2 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group V. Moreno 3 Internet-Draft Cisco Systems 4 Intended status: Experimental D. Farinacci 5 Expires: September 9, 2018 lispers.net 6 March 8, 2018 8 Signal-Free LISP Multicast 9 draft-ietf-lisp-signal-free-multicast-09 11 Abstract 13 When multicast sources and receivers are active at LISP sites, the 14 core network is required to use native multicast so packets can be 15 delivered from sources to group members. When multicast is not 16 available to connect the multicast sites together, a signal-free 17 mechanism can be used to allow traffic to flow between sites. The 18 mechanism within here uses unicast replication and encapsulation over 19 the core network for the data-plane and uses the LISP mapping 20 database system so encapsulators at the source LISP multicast site 21 can find decapsulators at the receiver LISP multicast sites. 23 Requirements Language 25 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 26 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 27 document are to be interpreted as described in [RFC2119]. 29 Status of This Memo 31 This Internet-Draft is submitted in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF). Note that other groups may also distribute 36 working documents as Internet-Drafts. The list of current Internet- 37 Drafts is at https://datatracker.ietf.org/drafts/current/. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 This Internet-Draft will expire on September 9, 2018. 46 Copyright Notice 48 Copyright (c) 2018 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (https://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the Simplified BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 64 2. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 4 65 3. Reference Model . . . . . . . . . . . . . . . . . . . . . . . 5 66 4. General Procedures . . . . . . . . . . . . . . . . . . . . . 7 67 4.1. General Receiver-Site Procedures . . . . . . . . . . . . 8 68 4.1.1. Multicast Receiver Detection . . . . . . . . . . . . 8 69 4.1.2. Receiver-Site Registration . . . . . . . . . . . . . 8 70 4.1.3. Consolidation of the Replication-List . . . . . . . . 9 71 4.2. General Source-Site Procedures . . . . . . . . . . . . . 10 72 4.2.1. Multicast Tree Building at the Source-Site . . . . . 10 73 4.2.2. Multicast Destination Resolution . . . . . . . . . . 10 74 4.3. General LISP Notification Procedures . . . . . . . . . . 11 75 5. Source Specific Multicast Trees . . . . . . . . . . . . . . . 11 76 5.1. Source Directly Connected to Source-ITRs . . . . . . . . 12 77 5.2. Source not Directly Connected to Source-ITRs . . . . . . 12 78 6. Multi-Homing Considerations . . . . . . . . . . . . . . . . . 12 79 6.1. Multiple ITRs at a Source-Site . . . . . . . . . . . . . 12 80 6.2. Multiple ETRs at a Receiver-Site . . . . . . . . . . . . 13 81 6.3. Multiple RLOCs for an ETR at a Receiver-Site . . . . . . 13 82 6.4. Multicast RLOCs for an ETR at a Receiver-Site . . . . . . 14 83 7. PIM Any Source Multicast Trees . . . . . . . . . . . . . . . 14 84 8. Signal-Free Multicast for Replication Engineering . . . . . . 15 85 9. Security Considerations . . . . . . . . . . . . . . . . . . . 18 86 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 87 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 19 88 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 19 89 12.1. Normative References . . . . . . . . . . . . . . . . . . 19 90 12.2. Informative References . . . . . . . . . . . . . . . . . 20 91 Appendix A. Document Change Log . . . . . . . . . . . . . . . . 21 92 A.1. Changes to draft-ietf-lisp-signal-free-multicast-09 . . . 21 93 A.2. Changes to draft-ietf-lisp-signal-free-multicast-08 . . . 21 94 A.3. Changes to draft-ietf-lisp-signal-free-multicast-07 . . . 21 95 A.4. Changes to draft-ietf-lisp-signal-free-multicast-06 . . . 21 96 A.5. Changes to draft-ietf-lisp-signal-free-multicast-05 . . . 21 97 A.6. Changes to draft-ietf-lisp-signal-free-multicast-04 . . . 22 98 A.7. Changes to draft-ietf-lisp-signal-free-multicast-03 . . . 22 99 A.8. Changes to draft-ietf-lisp-signal-free-multicast-02 . . . 22 100 A.9. Changes to draft-ietf-lisp-signal-free-multicast-01 . . . 22 101 A.10. Changes to draft-ietf-lisp-signal-free-multicast-00 . . . 23 102 A.11. Changes to draft-farinacci-lisp-signal-free-multicast-04 23 103 A.12. Changes to draft-farinacci-lisp-signal-free-multicast-03 23 104 A.13. Changes to draft-farinacci-lisp-signal-free-multicast-02 23 105 A.14. Changes to draft-farinacci-lisp-signal-free-multicast-01 23 106 A.15. Changes to draft-farinacci-lisp-signal-free-multicast-00 23 107 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 23 109 1. Introduction 111 When multicast sources and receivers are active at LISP sites, and 112 the core network between the sites does not provide multicast 113 support, a signal-free mechanism can be used to create an overlay 114 that will allow multicast traffic to flow between sites and connect 115 the multicast trees at the different sites. 117 The signal-free mechanism proposed here does not extend PIM [RFC7761] 118 over the overlay as proposed in [RFC6831], nor does the mechanism 119 utilize direct signaling between the Receiver-ETRs and Sender-ITRs as 120 described in [I-D.farinacci-lisp-mr-signaling]. The signal-free 121 mechanism proposed reduces the amount of signaling required between 122 sites to a minimum and is centered around the registration of 123 Receiver-sites for a particular multicast-group or multicast-channel 124 with the LISP Mapping System. 126 Registrations from the different receiver-sites will be merged at the 127 Mapping System to assemble a multicast-replication-list inclusive of 128 all RLOCs that lead to receivers for a particular multicast-group or 129 multicast-channel. The replication-list for each specific multicast- 130 entry is maintained as a database mapping entry in the LISP Mapping 131 System. 133 When the ITR at the source-site receives multicast traffic from 134 sources at its site, the ITR can query the mapping system by issuing 135 Map-Request messages for the (S,G) source and destination addresses 136 in the packets received. The Mapping System will return the RLOC 137 replication-list to the ITR, which the ITR will cache as per standard 138 LISP procedure. Since the core is assumed to not support multicast, 139 the ITR will replicate the multicast traffic for each RLOC on the 140 replication-list and will unicast encapsulate the traffic to each 141 RLOC. The combined function or replicating and encapsulating the 142 traffic to the RLOCs in the replication-list is referred to as "rep- 143 encapsulation" in this document. 145 The document describes the General Procedures (Section 4) and 146 information encoding that are required at the Receiver-sites and 147 Source-sites to achieve signal-free multicast interconnectivity. The 148 General Procedures for Mapping System Notifications to different 149 sites are also described. A section dedicated to the specific case 150 of SSM trees discusses the implications to the General Procedures for 151 SSM multicast trees over different topological scenarios. A section 152 on ASM support is included to identify the constraints that come 153 along with supporting it using LISP Signal-Free multicast. 155 There is a section dedicated to Replication Engineering. A mechanism 156 to reduce the impact of head-end replication. The mapping system, 157 via LISP Signal-Free mechanisms, can be used to build a layer of 158 RTRs. 160 2. Definition of Terms 162 LISP related terms, notably Map-Request, Map-Reply, Ingress Tunnel 163 Router (ITR), Egress Tunnel Router (ETR), Map-Server (MS) and Map- 164 Resolver (MR) are defined in the LISP specification [RFC6830]. 166 Extensions to the definitions in [RFC6830] for their application to 167 multicast routing are documented in [RFC6831]. 169 Terms defining interactions with the LISP Mapping System are defined 170 in [RFC6833]. 172 The following terms are consistent with the definitions in [RFC6830] 173 and [RFC6831]. The terms are specific cases of the general terms and 174 are here defined to facilitate the descriptions and discussions 175 within this particular document. 177 Source: Multicast source end-point. Host originating multicast 178 packets. 180 Receiver: Multicast group member end-point. Host joins multicast 181 group as a receiver of multicast packets sent to the group. 183 Receiver-site: LISP site where multicast receivers are located. 185 Source-site: LISP site where multicast sources are located. 187 RP-site: LISP site where an ASM PIM Rendezvous Point [RFC7761] is 188 located. The RP-site and the Source-site MAY be the same in some 189 situations. 191 Receiver-ETR: LISP decapsulating xTR at the Receiver-site. This is a 192 multicast ETR. 194 Source-ITR: LISP encapsulating xTR at the Source-site. This is a 195 multicast ITR. 197 RP-xTR: LISP xTR at the RP-site. This is typically a multicast ITR. 199 Replication-list: Mapping-entry containing the list of RLOCs that 200 have registered Receivers for a particular multicast-entry. 202 Multicast-entry: A tuple identifying a multicast tree. Multicast- 203 entries are in the form of (S-prefix, G-prefix). 205 Rep-encapsulation: The process of replicating and then encapsulating 206 traffic to multiple RLOCs. 208 Re-encapsulating Tunnel Router (RTR): An RTR is a router that 209 implements the re-encapsulating tunnel function detailed in Section 8 210 of the main LISP specification [RFC6830]. A LISP RTR performs packet 211 re-routing by chaining ETR and ITR functions, whereby it first 212 removes the LISP header of an ingress packet and then prepends a new 213 LISP header to an egress packet. 215 RTR Level: An RTR level is encoded in a Replication-List-Entry (RLE) 216 LCAF Type detailed in [RFC8060]. Each entry in the replication list 217 contains an address of an xTR and a level value. Level values are 218 used to create a replication hierarchy so that ITRs at source LISP 219 sites replicate to the lowest (smaller value) level number RTRs in a 220 RLE entry. And then RTRs at a given level replicate to the next 221 higher level of RTRs. The number of RTRs at each level are 222 engineered to control the fan-out or replication factor so a tradeoff 223 between the width of the level versus the number of levels can be 224 selected. 226 ASM: Any-Source Multicast as defined in [RFC3569] and [RFC7761] where 227 multicast distribution trees are built with a Rendezvous Point. 229 SSM: Source Specific Multicast as defined in [RFC3569] where 230 multicast distribution trees are built and rooted at the multicast 231 router(s) directly connected to the multicast source. 233 3. Reference Model 235 The reference model that will be used for the discussion of the 236 Signal-Free multicast tree interconnection is illustrated in 237 Figure 1. 239 MS/MR 240 +---+ 241 | | 242 +---+ +---+ +---+ +---+ +---+ 243 Src-1 ----| R1|-----|ITR| | |ETR|------| R2|------ Rcv-2 244 +---+ +---+ | +---+ +---+ 245 \ | / 246 Source-site-1 \ | / Receiver-site-2 247 \ | / 248 \ | / 249 \ | / 250 Core 251 / \ 252 / \ 253 / \ 254 / \ 255 / \ 256 +---+ +---+ 257 Src-3 --------------|ITR| |ETR|----------------- Rcv-4 258 +---+ +---+ 260 Source-site-3 Receiver-site-4 262 Figure 1: LISP Multicast Generic Reference Model 264 Sites 1 and 3 are Source-sites. 266 Source-site-3 presents a Source (Src-3) that is directly connected to 267 the Source-ITR 269 Source-site-1 presents a Source (Src-1) that is one hop or more away 270 from the Source-ITR 272 Receiver-site-2 and 4 are receiver sites with not-directly connected 273 and directly connected Receiver end-points respectively 275 R1 is a multicast router in Source-site-1. 277 R2 is a multicast router at the Receiver-site. 279 The Map-Servers and Resolvers are reachable in the RLOC space in the 280 Core, only one is shown for illustration purposes, but these can be 281 many or even part of a Distributed Mapping System, such as a DDT 282 Tree. 284 The procedures for interconnecting multicast Trees over an overlay 285 can be broken down into three functional areas: 287 o Receiver-site procedures 289 o Source-site procedures 291 o LISP notification procedures 293 The receiver site procedures will be common for most tree types and 294 topologies. 296 The procedures at the source site can vary depending on the type of 297 trees being interconnected as well as based on the topological 298 relation between sources and source-site xTRs. For ASM trees, a 299 special case of the Source-site is the RP-site for which a variation 300 of the Source-site procedures MAY be necessary if ASM trees are to be 301 supported in future specifications of LISP Signal-Free multicast. 303 The LISP notification procedures between sites are normalized for the 304 different possible scenarios. Certain scenarios MAY benefit from a 305 simplified notification mechanism or no notification requirement at 306 all. 308 4. General Procedures 310 The interconnection of multicast trees across different LISP sites 311 involves the following procedures to build the necessary multicast 312 distribution trees across sites. 314 1. The presence of multicast Receiver end-points is detected by the 315 Receiver-ETRs at the Receiver-sites. 317 2. Receiver-ETRs register their RLOCs as part of the replication- 318 list for the multicast-entry the detected Receivers subscribe to. 320 3. The Mapping-system merges all receiver-ETR or delivery-group 321 RLOCs to build a comprehensive replication-list inclusive of all 322 Receiver-sites for each multicast-entry. 324 4. LISP Map-Notify messages MUST be sent to the Source-ITR informing 325 of any changes in the replication-list. 327 5. Multicast-tree building at the Source-site is initiated when the 328 Source-ITR receives the LISP Notification. 330 Once the multicast distribution trees are built, the following 331 forwarding procedures may take place: 333 1. The Source sends multicast packets to the multicast group 334 destination address. 336 2. Multicast traffic follows the multicast tree built at the Source- 337 site and makes its way to the Source-ITRs. 339 3. The Source-ITR will issue a map-request to resolve the 340 replication-list for the multicast-entry. 342 4. The Mapping System responds to the Source-ITR with a map-reply 343 containing the replication-list for the multicast group 344 requested. 346 5. The Source-ITR caches the replication-list received in the map- 347 reply for the multicast-entry. 349 6. Multicast traffic is rep-encapsulated. That is, the packet is 350 replicated for each RLOC in the replication-list and then 351 encapsulated to each one. 353 4.1. General Receiver-Site Procedures 355 4.1.1. Multicast Receiver Detection 357 When the Receiver-ETRs are directly connected to the Receivers (e.g. 358 Receiver-site-4 in Figure 1), the Receiver-ETRs will receive IGMP 359 Reports from the Receivers indicating which group the Receivers wish 360 to subscribe to. Based on these IGMP Reports, the receiver-ETR is 361 made aware of the presence of Receivers as well as which group they 362 are interested in. 364 When the Receiver-ETRs are several hops away from the Receivers (e.g. 365 Receiver-site-2 in Figure 1), the Receiver-ETRs will receive PIM join 366 messages which will allow the Receiver-ETR to know that there are 367 multicast Receivers at the site and also learn which multicast group 368 the Receivers are for. 370 4.1.2. Receiver-Site Registration 372 Once the Receiver-ETRs detect the presence of Receivers at the 373 Receiver-site, the Receiver-ETRs MUST issue Map-Register messages to 374 include the Receiver-ETR RLOCs in the replication-list for the 375 multicast-entry the Receivers joined. 377 The Map-Register message MUST use the multicast-entry (Source, Group) 378 tuple as its EID record type with the Receiver-ETR RLOCs conforming 379 the locator set. 381 The EID in the Map-Register message MUST be encoded using the 382 Multicast Information LCAF type defined in [RFC8060]. 384 The RLOC in the Map-Register message MUST be encoded using the 385 Replication List Entry (RLE) LCAF type defined in [RFC8060] with the 386 Level Value fields for all entries set to 128 (decimal). 388 The encoding described above MUST be used consistently for Map- 389 Register messages, entries in the Mapping System, Map-reply messages 390 as well as the map-cache at the Source-ITRs. 392 The Map-Register messages [RFC6830] sent by the receiver-ETRs MUST 393 have the following bits set as here specified: 395 1. merge-request-bit set to 1. The Map-Register messages are sent 396 with "Merge Semantics". The Map-Server will receive 397 registrations from a multitude of Receiver-ETRs. The Map-Server 398 will merge the registrations for common EIDs and maintain a 399 consolidated replication-list for each multicast-entry. 401 2. want-map-notify-bit (M) set to 0. This tells the Mapping System 402 that the receiver-ETR does not expect to receive Map-Notify 403 messages as it does not need to be notified of all changes to the 404 replication-list. 406 3. proxy-reply-bit (P) set to 1. The merged replication-list is 407 kept in the Map-Servers. By setting the proxy-reply bit, the 408 receiver-ETRs instruct the Mapping-system to proxy reply to map- 409 requests issued for the multicast entries. 411 Map-Register messages for a particular multicast-entry MAY be sent 412 for every receiver detected, even if previous receivers have been 413 detected for the particular multicast-entry. This allows the 414 replication-list to remain up to date. 416 Receiver-ETRs MUST be configured to know what Map-Servers Map- 417 Register messages are sent to. The configuration is likely to be 418 associated with an S-prefix that multiple (S,G) entries match to and 419 are more specific for. Therefore, the S-prefix determines the Map- 420 Server set in the least number of configuration statements. 422 4.1.3. Consolidation of the Replication-List 424 The Map-Server will receive registrations from a multitude of 425 Receiver-ETRs. The Map-Server will merge the registrations for 426 common EIDs and consolidate a replication-list for each multicast- 427 entry. 429 When an ETR sends an RLE RLOC-record in a Map-Register and the RLE 430 entry already exists in the Map-Server's RLE merged list, the Map- 431 Server will replace the single RLE entry with the information from 432 the Map-Register RLOC-record. The Map-Server MUST NOT merge 433 duplicate RLOCs in the consolidated replication-list. 435 4.2. General Source-Site Procedures 437 Source-ITRs MUST register the unicast EIDs of any Sources or 438 Rendezvous Points that may be present on the Source-site. In other 439 words, it is assumed that the Sources and RPs are LISP EIDs. 441 The registration of the unicast EIDs for the Sources or Rendezvous 442 Points allows the Map-Server to know where to send Map-Notify 443 messages to. Therefore, the Source-ITR MUST register the unicast 444 S-prefix EID with the want-map-notify-bit set in order to receive 445 Map-Notify messages whenever there is a change in the replication- 446 list. 448 4.2.1. Multicast Tree Building at the Source-Site 450 When the source site receives the Map-Notify messages from the 451 mapping system as described in Section 4.3, it will initiate the 452 process of building a multicast distribution tree that will allow the 453 multicast packets from the Source to reach the Source-ITR. 455 The Source-ITR MUST issue a PIM join for the multicast-entry for 456 which it received the Map-Notify message. The join will be issued in 457 the direction of the source or in the direction of the RP for the SSM 458 and ASM cases respectively. 460 4.2.2. Multicast Destination Resolution 462 On reception of multicast packets, the source-ITR obtains the 463 replication-list for the (S,G) addresses in the packets. 465 In order to obtain the replication-list, the Source-ITR MUST issue a 466 Map-Request message in which the EID is the (S,G) multicast tuple 467 which is encoded using the Multicast Info LCAF type defined in 468 [RFC8060]. 470 The Mapping System (most likely the Map-Server) will Map-reply with 471 the merged replication-list maintained in the Mapping System. The 472 Map-reply message MUST follow the format defined in [RFC6830], its 473 EID is encoded using the Multicast Info LCAF type and the 474 corresponding RLOC-records are encoded using the RLE LCAF type. Both 475 LCAF types defined in [RFC8060]. 477 4.3. General LISP Notification Procedures 479 The Map-Server will issue LISP Map-Notify messages to inform the 480 Source-site of the presence of receivers for a particular multicast 481 group over the overlay. 483 Updated Map-Notify messages SHOULD be issued every time a new 484 registration is received from a Receiver-site. This guarantees that 485 the source-sites are aware of any potential changes in the multicast- 486 distribution-list membership. 488 The Map-Notify messages carry (S,G) multicast EIDs encoded using the 489 Multicast Info LCAF type defined in [RFC8060]. 491 Map-Notify messages will be sent by the Map-Server to the RLOCs with 492 which the unicast S-prefix EID was registered. In the case when 493 sources are discovered dynamically [I-D.ietf-lisp-eid-mobility], xTRs 494 MUST register sources explicitly with the want-map-notify-bit set. 495 This is so the ITR in the site the source has moved to can get the 496 most current replication list. 498 When both the Receiver-sites and the Source-sites register to the 499 same Map-Server, the Map-Server has all the necessary information to 500 send the Map-Notify messages to the Source-site. 502 When the Map-Servers are distributed (when using LISP-DDT [RFC8111]), 503 the Receiver-sites MAY register to one Map-Server while the Source- 504 site registers to a different Map-Server. In this scenario, the Map- 505 Server for the receiver sites MUST resolve the unicast S-prefix EID 506 across a distributed mapping transport system, per standard LISP 507 lookup procedures and obtain the necessary information to send the 508 Map-Notify messages to the Source-site. The Map-Notify messages are 509 sent with an authentication length of 0 as they would not be 510 authenticated. 512 When the Map-Servers are distributed, different Receiver-sites MAY 513 register to different Map-Servers. However, this is not supported 514 with the currently defined mechanisms. 516 5. Source Specific Multicast Trees 518 The interconnection of Source Specific Multicast (SSM) Trees across 519 sites will follow the General Receiver-site Procedures described in 520 Section 4.1 on the Receiver-sites. 522 The Source-site Procedures will vary depending on the topological 523 location of the Source within the Source-site as described in 524 Section 5.1 and Section 5.2 . 526 5.1. Source Directly Connected to Source-ITRs 528 When the Source is directly connected to the source-ITR, it is not 529 necessary to trigger signaling to build a local multicast tree at the 530 Source-site. Therefore Map-Notify messages are not required to 531 initiate building of the multicast tree at the Source-site. 533 Map-Notify messages are still required to ensure that any changes to 534 the replication-list are communicated to the Source-site so that the 535 map-cache at the Source-ITRs is kept updated. 537 5.2. Source not Directly Connected to Source-ITRs 539 The General LISP Notification Procedures described in Section 4.3 540 MUST be followed when the Source is not directly connected to the 541 source-ITR. On reception of Map-Notify messages, local multicast 542 signaling MUST be initiated at the Source-site per the General Source 543 Site Procedures for Multicast Tree building described in 544 Section 4.2.1. 546 In the SSM case, the IP address of the Source is known and it is also 547 registered with the LISP mapping system. Thus, the mapping system 548 MAY resolve the mapping for the Source address in order to send Map- 549 Notify messages to the correct source-ITR. 551 6. Multi-Homing Considerations 553 6.1. Multiple ITRs at a Source-Site 555 When multiple ITRs exist at a source multicast site, care MUST be 556 taken that more than one ITR does not head-end replicate packets else 557 receiver multicast sites will receive duplicate packets. The 558 following procedures will be used for each topology scenarios: 560 o When more than one ITR is directly connected to the source host, 561 either the PIM DR or the IGMP querier (when PIM is not enabled on 562 the ITRs) is responsible for packet replication. All other ITRs 563 silently drop the packet. In the IGMP querier case, one or more 564 ITRs on the source LAN MUST be IGMP querier candidates. 565 Therefore, it is required they are configured as such. 567 o When more than one ITR is multiple hops away from the source host 568 and one of the ITRs is the PIM Rendezvous Point, then the PIM RP 569 is responsible for packet replication. 571 o When more than one ITR is multiple hops away from the source host 572 and the PIM Rendezvous Point is not one of the ITRs, then one of 573 the ITRs MUST join to the RP. When a Map-Notify is received from 574 the Map-Server by an ITR, only the highest RLOC addressed ITR will 575 join toward the PIM RP or toward the source. 577 6.2. Multiple ETRs at a Receiver-Site 579 When multiple ETRs exist in a receiver multicast site, and each 580 create multicast join state, they each Map-Register their RLOC 581 addresses to the mapping system. In this scenario, the replication 582 happens on the overlay causing multiple ETR entry points to replicate 583 to all receivers versus a single ETR entry point replicating to all 584 receivers. If an ETR does not create join state, because it has not 585 received PIM joins or IGMP reports, it will not Map-Register its RLOC 586 addresses to the mapping system. The same procedures in Section 4.1 587 are followed. 589 When multiple ETRs exist on the same LAN as a receiver host, then the 590 PIM DR, when PIM is enabled, or the IGMP querier is responsible for 591 sending a Map-Register for its RLOC. In the IGMP case, one or more 592 ETRs on LAN MUST be IGMP querier candidates. Therefore, it is 593 required they are configured as such. 595 6.3. Multiple RLOCs for an ETR at a Receiver-Site 597 It MAY be desirable to have multiple underlay paths to an ETR for 598 multicast packet delivery. This can be done by having multiple RLOCs 599 assigned to an ETR and having the ETR send Map-Registers for all its 600 RLOCs. By doing this, an ITR can choose a specific path based on 601 underlay performance and/or RLOC reachability. 603 It is recommended that an ETR sends a Map-Register with a single 604 RLOC-record that uses the ELP LCAF type [RFC8060] that is nested 605 inside RLE entry LCAF. For example say ETR1 has assigned RLOC1 and 606 RLOC2 for a LISP receiver site. And there is ETR2 in another LISP 607 receiver site, that has RLOC3. The two receiver sites have the same 608 (S,G) being joined. Here is how the RLOC-record is encoded on each 609 ETR: 611 ETR1: EID-record: (S,G) 612 RLOC-record: RLE[ ELP{ (RLOC1,s,p), (RLOC2,s,p) } ] 614 ETR2: EID-record: (S,G) 615 RLOC-record: RLE[ RLOC3 ] 617 And here is how the entry is merged and stored on the Map-Server 618 since the Map-Registers have an RLE encoded RLOC-record: 620 MS: EID-record: (S,G) 621 RLOC-record: RLE[ RLOC3, ELP{ (RLOC1,s,p), (RLOC2,s,p) } ] 623 When the ITR receives a packet from a multicast source S for group G, 624 it uses the merged RLOC-record, returned from the Map-Server. The 625 ITR replicates the packet to (RLOC3 and RLOC1) or (RLOC3 and RLOC2). 626 Since it is required for the s-bit to be set for RLOC1, the ITR MUST 627 replicate to RLOC1 if it is reachable. When the required p-bit is 628 also set, the RLOC-reachability mechanisms from [RFC6830] are 629 followed. If the ITR determines that RLOC1 is unreachable, it uses 630 RLOC2, as long as RLOC2 is reachable. 632 6.4. Multicast RLOCs for an ETR at a Receiver-Site 634 This specification is focused on underlays without multicast support, 635 but does not preclude the use of multicast RLOCs in RLE entries. 636 ETRs MAY register multicast EID entries using multicast RLOCs. In 637 such cases the ETRs will get joined to underlay multicast 638 distribution trees by using IGMP as a multicast host using mechanisms 639 in [RFC2236] and [RFC3376]. 641 7. PIM Any Source Multicast Trees 643 LISP signal-free multicast can support ASM Trees in limited but 644 acceptable topologies. It is suggested for the simplification of 645 building ASM trees across the LISP overlay to have PIM-ASM run 646 independently in each LISP site. What this means, is that a PIM 647 Rendezvous Point (RP) is configured in each LISP site so PIM Register 648 procedures and (*,G) state maintenance is contained within the LISP 649 site. 651 The following procedure will be used to support ASM in each LISP 652 site: 654 1. In a Receiver-site, the RP is colocated with the ETR. RPs for 655 different groups can be spread across each ETR, but is not 656 required. 658 2. When (*,G) state is created in an ETR, the procedures in 659 Section 4.1.2 are followed. In addition, the ETR registers 660 (S-prefix,G), where S-prefix is 0/0 (the respective unicast 661 default route for the address-family) to the mapping system. 663 3. In a Source-site, the RP is colocated with the ITR. RPs for 664 different groups can be spread across each ITR, but is not 665 required. 667 4. When a multicast source sends a packet, a PIM Register message is 668 delivered to the ITR and the procedures in Section 4.2 are 669 followed. 671 5. When the ITR sends a Map-Request for (S,G) and no Receiver-site 672 has registered for (S,G), the mapping system will return the 673 (0/0,G) entry to the ITR so it has a replication list of all the 674 ETRs that have received (*,G) state. 676 6. The ITR stores the replication-list in its map-cache for (S,G). 677 It replicates packets to all ETRs in the list. 679 7. ETRs decapsulate packets and forward based on (*,G) state in 680 their site. 682 8. When last-hop PIM routers join the newly discovered (S,G), the 683 ETR will store the state and follow the procedures in 684 Section 4.1.2. 686 8. Signal-Free Multicast for Replication Engineering 688 The mechanisms in this draft can be applied to the LISP Replication- 689 Engineering [I-D.coras-lisp-re] design. Rather than having the 690 layered LISP-RE RTR hierarchy use signaling mechanisms, the RTRs can 691 register their availability for multicast tree replication via the 692 mapping database system. 694 As stated in [I-D.coras-lisp-re], the RTR layered hierarchy is used 695 to avoid head-end replication in replicating nodes closest to a 696 multicast source. Rather than have multicast ITRs replicate to each 697 ETR in an RLE entry of a (S,G) mapping database entry, it could 698 replicate to one or more layer-0 RTRs in the LISP-RE hierarchy. 700 This draft documents how the RTR hierarchy is determined but not what 701 are the optimal layers of RTRs to use. Methods for determining 702 optimal paths or RTR topological closeness are out of scope for his 703 document. 705 There are two formats an (S,G) mapping database entry could have. 706 One format is a 'complete-format' and the other is a 'filtered- 707 format'. A 'complete-format' entails an (S,G) entry having multiple 708 RLOC records which contain both ETRs that have registered as well as 709 the RTRs at the first level of the LISP-RE hierarchy for the ITR to 710 replicate to. When using 'complete-format', the ITR has the ability 711 to select if it replicates to RTRs or to the registered ETRs at the 712 receiver sites. A 'filtered-format' (S,G) entry is one where the 713 Map-Server returns the RLOC-records that it decides the ITR SHOULD 714 use. So replication policy is shifted from the ITRs to the mapping 715 system. The Map-Servers can also decide for a given ITR, if it uses 716 a different set of replication targets per (S,G) entry for which the 717 ITR is replicating for. 719 The procedure for the LISP-RE RTRs to make themselves available for 720 replication can occur before or after any receivers join an (S,G) 721 entry or any sources send for a particular (S,G) entry. Therefore, 722 newly configured RTR state will be used to create new (S,G) state and 723 inherited into existing (S,G) state. A set of RTRs can register 724 themselves to the mapping system or a third-party can do so on their 725 behalf. When RTR registration occurs, it is done with an (S-prefix, 726 G-prefix) entry so it can advertise its replication services for a 727 wide-range of source/group combinations. 729 When a Map-Server receives (S,G) registrations from ETRs and 730 (S-prefix, G-prefix) registrations from RTRs, it has the option of 731 merging the RTR RLOC-records for each (S,G) that is more-specific for 732 the (S-prefix, G-prefix) entry or keep them separate. When merging, 733 a Map-Server is ready to return a 'complete-format' Map-Reply. When 734 keeping the entries separate, the Map-Server can decide what to 735 include in a Map-Reply when a Map-Request is received. It can 736 include a combination of RLOC-records from each entry or decide to 737 use one or the other depending on policy configured. 739 +---+ +----+ 740 Src-1 --------------|ITR| |ETR1|---------------- Rcv-1 741 +---+ +----+ 742 \ / 743 Source-site-1 \ / Receiver-site-1 744 \ / 745 \ / 746 +----+ \ / +----+ 747 |RTR1| \ / |RTR2| Level-0 748 +----+ \ / +----+ 749 \ <^^^^^^^^^^^^^^> / 750 \ < > / 751 < Core-Network > 752 < > 753 754 / / \ \ 755 / / \ \ 756 +----+ / / \ \ +----+ 757 |RTR3| / \ |RTR4| Level-1 758 +----+ / \ +----+ 759 / \ 760 / \ 761 +----+ +----+ 762 Rcv-2 --------------|ETR2| |ETR3|---------------- Rcv-3 763 +----+ +----+ 765 Receiver-site-2 Receiver-site-3 767 Figure 2: LISP-RE Reference Model 769 Here is a specific example, illustrated in Figure 2, of (S,G) and 770 (S-prefix, G-prefix) mapping database entries when a source S is 771 behind an ITR and there are receiver sites joined to (S,G) via ETR1, 772 ETR2, and ETR3. And there exists a LISP-RE hierarchy of RTR1 and 773 RTR2 at level-0 and RTR3 and RTR4 at level-1: 775 EID-record: (S,G) 776 RLOC-record: RLE: (ETR1, ETR2, ETR3), p1 777 EID-record: (S-prefix, G-prefix) 778 RLOC-record: RLE: (RTR1(L0), RTR2(L0), RTR3(L1), RTR4(L1)), p1 780 The above entries are in the form of how they were registered and 781 stored in a Map-Server. When a Map-Server uses 'complete-format', a 782 Map-Reply it originates has the mapping record encoded as: 784 EID-record: (S,G) 785 RLOC-record: RLE: (RTR1(L0), RTR3(L1)), p1 786 RLOC-record: RLE: (ETR1, ETR2, ETR3), p1 788 The above Map-Reply allows the ITR to decide if it replicates to the 789 ETRs or if it SHOULD replicate only to level-0 RTR1. This decision 790 is left to the ITR since both RLOC-records have priority 1. If the 791 Map-Server wanted to force the ITR to replicate to RTR1, it would set 792 the ETRs RLOC-record to priority greater than 1. 794 When a Map_server uses "filtered-format', a Map-Reply it originates 795 has the mapping record encoded as: 797 EID-record: (S,G) 798 RLOC-record: RLE: (RTR1(L0), RTR3(L1)), p1 800 An (S,G) entry can contain alternate RTRs. So rather than 801 replicating to multiple RTRs, one of a RTR set MAY be used based on 802 the RTR reachability status. An ITR can test reachability status to 803 any layer-0 RTR using RLOC-probing so it can choose one RTR from a 804 set to replicate to. When this is done the RTRs are encoded in 805 different RLOC-records versus together in one RLE RLOC-record. This 806 moves the replication load off the ITRs at the source site to the 807 RTRs inside the network infrastructure. This mechanism can also be 808 used by level-n RTRs to level-n+1 RTRs. 810 The following mapping would be encoded in a Map-Reply sent by a Map- 811 Server and stored in the ITR. The ITR would use RTR1 until it went 812 unreachable and then switch to use RTR2: 814 EID-record: (S,G) 815 RLOC-record: RTR1, p1 816 RLOC-record: RTR2, p2 818 9. Security Considerations 820 [I-D.ietf-lisp-sec] defines a set of security mechanisms that provide 821 origin authentication, integrity and anti-replay protection to LISP's 822 EID-to-RLOC mapping data conveyed via mapping lookup process. LISP- 823 SEC also enables verification of authorization on EID-prefix claims 824 in Map-Reply messages. 826 Additional security mechanisms to protect the LISP Map-Register 827 messages are defined in [RFC6833]. 829 The security of the Mapping System Infrastructure depends on the 830 particular mapping database used. The [RFC8111] specification, as an 831 example, defines a public-key based mechanism that provides origin 832 authentication and integrity protection to the LISP DDT protocol. 834 Map-Replies received by the source-ITR can be signed (by the Map- 835 Server) so the ITR knows the replication-list is from a legit source. 837 Data-plane encryption can be used when doing unicast rep- 838 encapsulation as described in [RFC8061]. 840 10. IANA Considerations 842 This document has no IANA implications 844 11. Acknowledgements 846 The authors want to thank Greg Shepherd, Joel Halpern and Sharon 847 Barkai for their insightful contribution to shaping the ideas in this 848 document. A special thanks to Luigi Iannone, LISP WG co-chair, for 849 shepherding this working group document. Thanks also goes to Jimmy 850 Kyriannis, Paul Vinciguerra, Florin Coras, and Yan Filyurin for 851 testing an implementation of this draft. 853 12. References 855 12.1. Normative References 857 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 858 Requirement Levels", BCP 14, RFC 2119, 859 DOI 10.17487/RFC2119, March 1997, 860 . 862 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 863 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 864 . 866 [RFC3376] Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A. 867 Thyagarajan, "Internet Group Management Protocol, Version 868 3", RFC 3376, DOI 10.17487/RFC3376, October 2002, 869 . 871 [RFC3569] Bhattacharyya, S., Ed., "An Overview of Source-Specific 872 Multicast (SSM)", RFC 3569, DOI 10.17487/RFC3569, July 873 2003, . 875 [RFC6830] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "The 876 Locator/ID Separation Protocol (LISP)", RFC 6830, 877 DOI 10.17487/RFC6830, January 2013, 878 . 880 [RFC6831] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, "The 881 Locator/ID Separation Protocol (LISP) for Multicast 882 Environments", RFC 6831, DOI 10.17487/RFC6831, January 883 2013, . 885 [RFC6833] Fuller, V. and D. Farinacci, "Locator/ID Separation 886 Protocol (LISP) Map-Server Interface", RFC 6833, 887 DOI 10.17487/RFC6833, January 2013, 888 . 890 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 891 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 892 Multicast - Sparse Mode (PIM-SM): Protocol Specification 893 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 894 2016, . 896 [RFC8060] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 897 Address Format (LCAF)", RFC 8060, DOI 10.17487/RFC8060, 898 February 2017, . 900 [RFC8111] Fuller, V., Lewis, D., Ermagan, V., Jain, A., and A. 901 Smirnov, "Locator/ID Separation Protocol Delegated 902 Database Tree (LISP-DDT)", RFC 8111, DOI 10.17487/RFC8111, 903 May 2017, . 905 12.2. Informative References 907 [I-D.coras-lisp-re] 908 Coras, F., Cabellos-Aparicio, A., Domingo-Pascual, J., 909 Maino, F., and D. Farinacci, "LISP Replication 910 Engineering", draft-coras-lisp-re-08 (work in progress), 911 November 2015. 913 [I-D.farinacci-lisp-mr-signaling] 914 Farinacci, D. and M. Napierala, "LISP Control-Plane 915 Multicast Signaling", draft-farinacci-lisp-mr-signaling-06 916 (work in progress), February 2015. 918 [I-D.ietf-lisp-eid-mobility] 919 Portoles-Comeras, M., Ashtaputre, V., Moreno, V., Maino, 920 F., and D. Farinacci, "LISP L2/L3 EID Mobility Using a 921 Unified Control Plane", draft-ietf-lisp-eid-mobility-01 922 (work in progress), November 2017. 924 [I-D.ietf-lisp-sec] 925 Maino, F., Ermagan, V., Cabellos-Aparicio, A., and D. 926 Saucez, "LISP-Security (LISP-SEC)", draft-ietf-lisp-sec-14 927 (work in progress), October 2017. 929 [RFC8061] Farinacci, D. and B. Weis, "Locator/ID Separation Protocol 930 (LISP) Data-Plane Confidentiality", RFC 8061, 931 DOI 10.17487/RFC8061, February 2017, 932 . 934 Appendix A. Document Change Log 936 [RFC Editor: Please delete this section on publication as RFC.] 938 A.1. Changes to draft-ietf-lisp-signal-free-multicast-09 940 o Posted March 2018. 942 o Fixed idnits in draft. 944 A.2. Changes to draft-ietf-lisp-signal-free-multicast-08 946 o Posted February 2018. 948 o Fixed last call editorial comments. 950 A.3. Changes to draft-ietf-lisp-signal-free-multicast-07 952 o Posted November 2017. 954 o Changes after shepherd review and RFC1918 terminology compliant. 956 A.4. Changes to draft-ietf-lisp-signal-free-multicast-06 958 o Posted July 2017. 960 o Stig made a comment about referencing RFC6831 when an RLOC is a 961 multicast address. It opens up a lot of assumptions on what parts 962 of RFC6831 is performed and which parts should not be performed. 963 In the case of signal-free-multicast, join the underlay trees as a 964 multicast host by using IGMP. 966 A.5. Changes to draft-ietf-lisp-signal-free-multicast-05 968 o Posted July 2017. 970 o Make it clear that when a RLE is sent by an ETR and it is already 971 in the merged RLE list on the Map-Server, that the Map-Server 972 replaces the RLE entry (versus adding a duplicate RLE entry to the 973 list). 975 o Make it clear that an RLOC can be a unicast or multicast address. 976 Then make a reference to RFC6831 about mechanisms to support 977 multicast RLOCs. 979 o Fix some typos. 981 A.6. Changes to draft-ietf-lisp-signal-free-multicast-04 983 o Posted May 2017. 985 o Make it clear that recieiver-ETRs need configuraiton information 986 for what Map-Servers (S,G) entries are registered to. 988 o Make it clear this document indicates what RTR layered hierarchy 989 to use and not if its the best hierarchy to use. 991 A.7. Changes to draft-ietf-lisp-signal-free-multicast-03 993 o Posted April 2017. 995 o Add "Multi-Homing Considerations" section to describe the case 996 where a source LISP site has multiple ITRs and the multicast 997 distribution tree at the source site branches to more than one 998 ITR. And at receiver sites where there are multiple ETRs and 999 multiple RLOCs per ETR. 1001 A.8. Changes to draft-ietf-lisp-signal-free-multicast-02 1003 o Posted October 2016. 1005 o Updated document expiration timer. 1007 A.9. Changes to draft-ietf-lisp-signal-free-multicast-01 1009 o Posted April 2016. 1011 o Add text to define RTRs and indicate how RTR level number is used 1012 for LISP-RE. 1014 o Draw figure 2 that shows a LISP-RE topology. 1016 o Indicate that PIM-ASM or (*,G) trees can be supported in LISP 1017 Signal-Free Multicast. 1019 A.10. Changes to draft-ietf-lisp-signal-free-multicast-00 1021 o Posted late December 2015. 1023 o Converted draft-farinacci-lisp-signal-free-multicast-04 into LISP 1024 working group draft. 1026 A.11. Changes to draft-farinacci-lisp-signal-free-multicast-04 1028 o Posted early December 2015. 1030 o Update references and document timer. 1032 A.12. Changes to draft-farinacci-lisp-signal-free-multicast-03 1034 o Posted June 2015. 1036 o Update references and document timer. 1038 A.13. Changes to draft-farinacci-lisp-signal-free-multicast-02 1040 o Posted December 2014. 1042 o Added section about how LISP-RE can use the mechanisms from 1043 signal-free-multicast so we can avoid head-end replication and 1044 avoid signalling across a layered RE topology. 1046 A.14. Changes to draft-farinacci-lisp-signal-free-multicast-01 1048 o Posted June 2014. 1050 o Changes based on implementation experience of this draft. 1052 A.15. Changes to draft-farinacci-lisp-signal-free-multicast-00 1054 o Posted initial draft February 2014. 1056 Authors' Addresses 1058 Victor Moreno 1059 Cisco Systems 1060 170 Tasman Drive 1061 San Jose, California 95134 1062 USA 1064 Email: vimoreno@cisco.com 1065 Dino Farinacci 1066 lispers.net 1067 San Jose, CA 95120 1068 USA 1070 Email: farinacci@gmail.com