idnits 2.17.1 draft-ietf-lisp-predictive-rlocs-02.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 (May 28, 2018) is 2153 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Unused Reference: 'I-D.ietf-lisp-eid-anonymity' is defined on line 574, but no explicit reference was found in the text ** Obsolete normative reference: RFC 6830 (Obsoleted by RFC 9300, RFC 9301) == Outdated reference: A later version (-15) exists of draft-farinacci-lisp-geo-05 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-eid-anonymity-02 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-eid-mobility-02 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-15 Summary: 1 error (**), 0 flaws (~~), 6 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Farinacci 3 Internet-Draft lispers.net 4 Intended status: Experimental P. Pillay-Esnault 5 Expires: November 29, 2018 Huawei Technologies 6 May 28, 2018 8 LISP Predictive RLOCs 9 draft-ietf-lisp-predictive-rlocs-02 11 Abstract 13 This specification describes a method to achieve near-zero packet 14 loss when an EID is roaming quickly across RLOCs. 16 Requirements Language 18 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 19 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 20 document are to be interpreted as described in [RFC2119]. 22 Status of This Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at https://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on November 29, 2018. 39 Copyright Notice 41 Copyright (c) 2018 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (https://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 Table of Contents 56 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 57 2. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 3 58 3. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 4. Design Details . . . . . . . . . . . . . . . . . . . . . . . 5 60 4.1. RLE Encoding . . . . . . . . . . . . . . . . . . . . . . 5 61 4.2. Packet Delivery Optimizations . . . . . . . . . . . . . . 6 62 4.3. Trading Off Replication Cost . . . . . . . . . . . . . . 8 63 5. Directional Paths with Intersections . . . . . . . . . . . . 9 64 6. Multicast Considerations . . . . . . . . . . . . . . . . . . 10 65 7. Multiple Address-Family Considerations . . . . . . . . . . . 11 66 8. Scaling Considerations . . . . . . . . . . . . . . . . . . . 11 67 9. Security Considerations . . . . . . . . . . . . . . . . . . . 12 68 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12 69 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 12 70 11.1. Normative References . . . . . . . . . . . . . . . . . . 12 71 11.2. Informative References . . . . . . . . . . . . . . . . . 13 72 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 13 73 Appendix B. Document Change Log . . . . . . . . . . . . . . . . 13 74 B.1. Changes to draft-ietf-lisp-predictive-rlocs-02 . . . . . 13 75 B.2. Changes to draft-ietf-lisp-predictive-rlocs-01 . . . . . 13 76 B.3. Changes to draft-ietf-lisp-predictive-rlocs-00 . . . . . 14 77 B.4. Changes to draft-farinacci-lisp-predictive-rlocs-02 . . . 14 78 B.5. Changes to draft-farinacci-lisp-predictive-rlocs-01 . . . 14 79 B.6. Changes to draft-farinacci-lisp-predictive-rlocs-00 . . . 14 80 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 82 1. Introduction 84 The LISP architecture [RFC6830] specifies two namespaces, End-Point 85 IDs (EIDs) and Routing Locators (RLOCs). An EID identifies a node in 86 the network and the RLOC indicates the EID's topological location. 87 When an node roams in the network, its EID remains fixed and 88 unchanged but the RLOCs associated with it change to reflect its new 89 topological attachment point. This specification will focus EIDs and 90 RLOCs residing in separate nodes. An EID is assigned to a host node 91 that roams while the RLOCs are assigned to network nodes that stay 92 stationary and are part of the network topology. For example, a set 93 of devices on an aircraft are assigned EIDs, and base stations on the 94 ground attached to the Internet infrastructure are configured as LISP 95 xTRs where their RLOCs are used for the bindings of the EIDs on the 96 aircraft up in the air. 98 The scope of this specification will not emphasize general physical 99 roaming as an aircraft would do in the sky but in a direction that is 100 more predictable such as a train traveling on a track or vehicle that 101 travels along a road. 103 2. Definition of Terms 105 Roaming-EID - is a network node that moves from one topological 106 location in the network to another. The network node uses the 107 same EID when it is roaming. That is, the EID address does not 108 change for reasons of mobility. A roaming-EID can also be a 109 roaming EID-prefix where a set of EIDs covered by the prefix are 110 all roaming and fate-sharing the same set of RLOCs at the same 111 time. 113 Predictive RLOCs - is a set of ordered RLOCs in a list each assigned 114 to LISP xTRs where the next RLOC in the list has high probability 115 it will be the next LISP xTR in a physical path going in a single 116 predictable direction. 118 Road-Side-Units (RSUs) - is a network node that acts as a router, 119 more specifically as a LISP xTR. The xTR automatically discovers 120 roaming-EIDs that come into network connectivity range and relays 121 packets to and from the roaming-EID. RSUs are typically deployed 122 along a directional path like a train track or road and are in 123 connectivity range of devices that travel along the directional 124 path. 126 3. Overview 128 The goal of this specification is to describe a make-before-break 129 EID-mobility mechanism that offers near-zero packet loss. Offering 130 minimal packet loss, not only allows transport layers to operate more 131 efficiently, but because an EID does not change while moving, 132 transport layer session continuity is maintained. To achieve these 133 requirements, a mechanism that reacts to the mobility event is 134 necessary but not sufficient. So the question is not that there 135 isn't a reaction but when it happens. By using some predictive 136 algorithms, we can guess with high probability where the EID will 137 roam to next. We can achieve this to a point where packet data will 138 be at the new location when the EID arrives. 140 First we should examine both the send and receive directions with 141 respect to the roaming-EID. Refer to Figure 1 for discussion. We 142 show a network node with a fixed EID address assigned to a roaming- 143 EID moving along a train track. And there are LISP xTRs deployed as 144 Road-Side-Units to support the connectivity between the roaming-EID 145 and the infrastructure or to another roaming-EID. 147 Roaming-EID ----> 149 ====//====//====//====//====//====//====//====//===//====//====//==== 150 // // // // // // // // // // // 151 ====//====//====//====//====//====//====//====//===//====//====//==== 153 xTR xTR xTR xTR xTR xTR 154 A B C D E F 156 Figure 1: Directional Mobility 158 For the send direction from roaming-EID to any destination can be 159 accomplish as a local decision. As long as the roaming-EID is in 160 signal range to any xTR along the path, it can use it to forward 161 packets. The LISP xTR, acting as an ITR, can forward packets to 162 destinations in non-LISP sites as well as to stationary and roaming 163 EIDs in LISP sites. This is accomplished by using the LISP overlay 164 via dynamic packet encapsulation. When the roaming-EID sends 165 packets, the LISP xTR must discover the EID and MAY register the EID 166 with a set of RLOCs to the mapping system 167 [I-D.ietf-lisp-eid-mobility]. The discovery process is important 168 because the LISP xTR, acting as an ETR for decapsulating packets that 169 arrive, needs to know what local ports or radios to send packets to 170 the roaming-EID. 172 Much of the focus of this design is on the packet direction to the 173 roaming-EID. And how remote LISP ITRs find the current location 174 (RLOCs) quickly when the roaming-EID is moving at high speed. This 175 specification solves the fast roaming with the introduction of the 176 Predictive-RLOCs algorithm. 178 Since a safe assumption is that the roaming-EID is going in one 179 direction and cannot deviate from it allows us to know a priori the 180 next set of RLOCs the roaming-EID will pass by. Referring to 181 Figure 1, if the roaming-EID is in range near xTR-A, then as it 182 moves, it will at some point pass by xTR-B and xTR-C, and so on. As 183 the roaming-EID moves, one could time when the EID is mapped to RLOC 184 A, and when it should change to RLOC B and so on. However, the speed 185 of movement of the roaming-EID won't be constant and the variables 186 involved in consistent timing cannot be relied on. Furthermore, 187 timing the move is not a make-before-break algorithm, meaning the 188 reaction of the binding happens at the time the roaming-EID is 189 discovered by an xTR. One cannot achieve fast hand-offs when message 190 signaling will be required to inform remote ITRs of the new binding. 192 The Predictive RLOCs algorithm allows a set of RLOCs, in an ordered 193 list, to be provided to remote ITRs so they have the information 194 available and local for when they need to use it. Therefore, no 195 control-plane message signaling occurs when the roaming-EID is 196 discovered by LISP xTRs. 198 4. Design Details 200 Predictive RLOCs accommodates for encapsulated packets to be 201 delivered to Road-Side-Unit LISP xTRs regardless where the roaming- 202 EID is currently positioned. 204 Referring to Figure 1, the following sequence is performed: 206 1. The Predictive RLOCs are registered to the mapping system as a 207 LCAF encoded Replication List Entry (RLE) Type [RFC8060]. The 208 registration can happen by one or more RSUs or by a third-party. 209 When registered by an RSU, and when no coordination is desired, 210 they each register their own RLOC with merge-semantics so the 211 list can be created and maintained in the LISP Map-Server. When 212 registered by a third-party, the complete list of RLOCs can be 213 included in the RLE. 215 2. There can be multiple RLEs present each as different RLOC- 216 records so a remote ITR can select one RLOC-record versus the 217 other based in priority and weight policy [RFC6830]. 219 3. When a remote ITR receives a packet destined for a roaming-EID, 220 it encapsulates and replicates to each RLOC in the RLE thereby 221 delivering the packet to the locations the roaming-EID is about 222 to appear. There are some cases where packets will go to 223 locations where the roaming-EID has already been, but see 224 Section 4.2 for packet delivery optimizations. 226 4. When the ETR resident RSU receives an encapsulated packet, it 227 decapsulates the packet and then determines if the roaming-EID 228 had been previously discovered. If the EID has not been 229 discovered, the ETR drops the packet. Otherwise, the ETR 230 delivers the decapsulated packet on the port interface the 231 roaming-EID was discovered on. 233 4.1. RLE Encoding 235 The LCAF [RFC8060] Replication List Entry (RLE) will be used to 236 encode the Predictive RLOCs in an RLOC-record for Map-Registers, Map- 237 Reply, and Map-Notify messages [RFC6830]. 239 0 1 2 3 240 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 241 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 242 | AFI = 16387 | Rsvd1 | Flags | 243 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 244 | Type = 13 | Rsvd2 | 4 + n | 245 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 246 | Rsvd3 | Rsvd4 | Level Value | 247 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 248 | AFI = x | RTR/ETR #1 ... | 249 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 250 | Rsvd3 | Rsvd4 | Level Value | 251 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 252 | AFI = x | RTR/ETR #n ... | 253 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 255 When the RLOC-record contains an RLE with RLOC entries all with the 256 same level value, it means the physical order listed is the 257 directional path of the RSUs. This will typically be the result of a 258 third-party doing the registration where it knows ahead of time the 259 RSU deployment. 261 When each RSU is registering with merge-semantics on their own, the 262 level number is used to place them in an ordered list. Since the 263 registrations come at different times and therefore arrive in 264 different order than the physical RSU path, the level number creates 265 the necessary sequencing. Each RSU needs to know its position in the 266 path relative to other RSUs. For example, in xTR-B, it would 267 register with level 1 since it is after xTR-A (and before xTR-C). So 268 if the registration order was xTR-B with level 1, xTR-C with level 2, 269 and xTR-A with level 0, the RLE list stored in the mapping system 270 would be (xTR-A, xTR-B, xTR-C). It is recommended that level numbers 271 be assigned in increments of 10 so latter insertion is possible. 273 The use of Geo-Prefixes and Geo-Points [I-D.farinacci-lisp-geo] can 274 be used to compare the physical presence of each RSU with respect to 275 each other, so they can choose level numbers to sequence themselves. 276 Also if the xTRs register with a Geo-Point in an RLOC-record, then 277 perhaps the Map-Server could sequence the RLE list. 279 4.2. Packet Delivery Optimizations 281 Since the remote ITR will replicate to all RLOCs in the RLE, a 282 situation is created where packets go to RLOCs that don't need to. 283 For instance, if the roaming-EID is along side of xTR-B and the RLE 284 is (xTR-A, xTR-B, xTR-C), there is no reason to replicate to xTR-A 285 since the roaming-EID has passed it and the the signal range is weak 286 or lost. However, replicating to xTR-B and xTR-C is important to 287 deliver packets to where the roaming-EID resides and where it is 288 about to go to. 290 A simple data-plane option, which converges fairly quickly is to have 291 the remote xTR, acting as an ETR, when packets are sent from the 292 roaming-EID, examine the source RLOC in the outer header of the 293 encapsulated packet. If the source RLOC is xTR-B, the remote xTR can 294 determine that the roaming-EID has moved past xTR-A and no longer 295 needs to encapsulate packets to xTR-A's RLOC. 297 In addition, the remote ITR can use RLOC-probing to determine if each 298 RLOC in the RLE is reachable. And if not reachable, exclude from the 299 list of RLOCs to replicate to. 301 This solution also handles the case where xTR-A and xTR-B may overlap 302 in radio signal range, but the signal is weak from the roaming-EID to 303 xTR-A but stronger to xTR-B. In this case, the roaming-EID selects 304 xTR-B to send packets that inform the remote xTR that return packets 305 should not be encapsulated to xTR-A. 307 There are also situations where the RSUs are in signal range of each 308 other in which case they could report reachability status of each 309 other. The use of the Locator-Status-Bits of the LISP encapsulation 310 header could be used to convey this information to the remote xTR. 311 This would only occur when the roaming-EID was discovered by both 312 xTR-A and xTR-B so it was possible for either xTR to reach the 313 roaming-EID. Either an IGP like routing protocol would be required 314 to allow each xTR to know the other could reach the roaming-EID or a 315 path trace tool (i.e. traceroute) could be originated by one xTR 316 targeted for the roaming-EID but MAC-forwarded through the other xTR. 317 These and other roaming-EID reachability mechanisms are work in 318 progress and for further study. 320 When a remote ITR is doing "Directional Mobility" and replicating to 321 the last RLOC in the RLE list, it has a decision to guess where the 322 roaming-EID will move to next. Conservatively, an ITR can replicate 323 to the entire set of RLOCs in the RLE list and wait to see if the 324 roaming-EID moves to one of the RLOCs in the RLE list. 326 Or more liberally, the remote ITR can assume the new roaming 327 direction. For example, with an RLE list of (xTR-A, xTR-B, xTR-C, 328 xTR-D) and the roaming-EID is at D, the remote ITR can replicate to 329 all of A, B, C and D to determine where the roaming-EID will move to 330 next. If the roaming-EID moves to C after it was at D, it is 331 possible that the EID is moving in the opposite direction from C to B 332 to A. This would be known as "Back-n-Forth Mobility". If an 333 implementation is configured to support this for a particluar EID, 334 the remote ITR could replicate in this sequence as the roaming-EID 335 moves from A to D and back to A: (A, B, C, D), (B, C, D), (C, D), (D, 336 C, B, A), (C, B, A), (B, A), and again (A, B, C, D). 338 The roaming-EID could be doing "Circular Mobility" where it moves 339 from A to D directionally, next from D-to-A, and then back to A to D 340 directionally again. This form of mobility is just as predicatable 341 as "Back-n-Forth Mobility" since a consistent pattern can be relied 342 on. Both of these mobility forms can be achieved with near-zero 343 packet loss. 345 On the other hand, the roaming-EID can be roaming arbitrarily using 346 "Random Mobility" where it could roam in the following combinations: 347 A-to-B, A-to-C, A-to-D, B-to-A, B-to-C, B-to-D, C-to-A, C-to-B, C-to- 348 D, D-to-A, D-to-B, or D-to-C. In this situation, when a return 349 packet arrives at the ITR, it could then just replicate to where the 350 roaming-EID is at rest and do so for a short period of time before it 351 replciates to the entire RLE list again. Using the wrong time period 352 could lead to packet loss. All these types of mobility can be 353 supported by the remote ITR in a local manner without consulting or 354 depending on any other LISP system. It is left for further study, if 355 any of the mobility types above should be encoded in the mapping 356 system. 358 4.3. Trading Off Replication Cost 360 If RLE lists are large, packet replication can occur to locations 361 well before the roaming-EID arrives. Making RLE lists small is 362 useful without sacrificing hand-off issues or incurring packet loss 363 to the application. By having overlapping RLEs in separate RLOC- 364 records we a simple mechanism to solve this problem. Here is an 365 example mapping entry to illustrate the point: 367 EID = , RLOC-records: 368 RLOC = (RLE: xTR-A, xTR-B) 369 RLOC = (RLE: xTR-B, xTR-C, xTR-D, xTR-E) 370 RLOC = (RLE: xTR-E, xTR-F) 372 When the remote ITR is encapsulating to xTR-B as a decision to use 373 the first RLOC-record, it can decide to move to use the second RLOC- 374 record because xTR-B is the last entry in the first RLOC-record and 375 the first entry in the second RLOC-record. When there are 376 overlapping RLEs, the remote ITR can decide when it is more efficient 377 to switch over. For example, when the roaming-EID is in range of 378 xTR-A, the remote ITR uses the first RLOC-record so the wasted 379 replication cost is to xTR-B only versus a worse cost when using the 380 second RLOC-record. But when the roaming-EID is in range of xTR-B, 381 then replicating to the other xTRs in the second RLOC-record may be 382 crucial if the roaming-EID has increased speed. And when the 383 roaming-EID may be at rest in a parked mode, then the remote ITR 384 encapsulates to only xTR-F using the third RLOC-record since the 385 roaming-EID has moved past xTR-E. 387 In addition, to eliminate unnecessary replication to xTRs further 388 down a directional path, GEO-prefixes [I-D.farinacci-lisp-geo] can be 389 used so only nearby xTRs that the roaming-EID is about to come in 390 contact with are the only ones to receive encapsulated packets. 392 Even when replication lists are not large, we can reduce the cost of 393 replication that the entire network bears by moving the replicator 394 away from the the source (i.e. the ITR) and closer to the RSUs (i.e. 395 the ETRs). See the use of RTRs for Replication Engineering 396 techniques in [RFC8378]. 398 5. Directional Paths with Intersections 400 A roaming-EID could be registered to the mapping system with the 401 following nested RLE mapping: 403 EID = , RLOC-records: 404 RLOC = (RLE: xTR-A, xTR-B, xTR-C, (RLE: xTR-X, xTR-Y, xTR-Z), 405 (RLE: xTR-I, xTR-J, xTR-K), xTR-D, xTR-E) 407 The mapping entry above describes 3 directional paths where the 408 ordered list has encoded one-level of two nested RLEs to denote 409 intersections in a horizontal path. Which is drawn as: 411 | | xTR-K 412 | | 413 | | 414 | | xTR-J 415 | | 416 | | 417 Roaming | | xTR-I 418 EID ----> | | 419 --------------------------------------- ------------------------------ 420 --------------------------------------- ------------------------------ 421 xTR-A xTR-B xTR-C | | xTR-D xTR-E 422 | | 423 | | xTR-X 424 | | 425 | | 426 | | xTR-Y 427 | | 428 | | 429 | | xTR-Z 431 When the roaming-EID is on the horizontal path, the remote-ITRs 432 typically replicate to the rest the of the xTRs in the ordered list. 433 When a list has nested RLEs, the replication should occur to at least 434 the first RLOC in a nested RLE list. So if the remote-ITR is 435 replicating to xTR-C, xTR-D, and xTR-E, it should also replicate to 436 xTR-X and xTR-I anticipating a possible turn at the intersection. 437 But when the roaming-EID is known to be at xTR-D (a left or right 438 hand turn was not taken), replication should only occur to xTR-D and 439 xTR-E. Once either xTR-I or xTR-X is determined to be where the 440 roaming-EID resides, then the replication occurs on the respective 441 directional path only. 443 When nested RLEs are used it may be difficult to get merge-semantics 444 to work when each xTR registers itself. So it is suggested a third- 445 party registers nested RLEs. It is left to further study to 446 understand better how to automate this. 448 6. Multicast Considerations 450 In this design, the remote ITR is receiving a unicast packet from an 451 EID and replicating and encapsulating to each RLOC in an RLE list. 452 This form of replication is no different than a traditional multicast 453 replication function. So replicating multicast packets in the same 454 fashion is a fallout from this design. 456 If there are multiple roaming-EIDs joined to the same multicast group 457 but reside at different RSUs, a merge has to be done of any pruned 458 RLEs used for forwarding. So if roaming-EID-1 resides at xTR-A and 459 roaming-EID-2 resides at xTR-B and the RLE list is (xTR-A, xTR-B, 460 xTR-C), and they are joined to the same multicast group, then 461 replication occurs to all of xTR-A, xTR-B, and xTR-C. Even since 462 roaming-EID-2 is past xTR-A, packets need to be delivered to xTR-A 463 for roaming-EID-1. In addition, packets need to be delivered to 464 xTR-C because roaming-EID-1 and roaming-EID-2 will get to xTR-C (and 465 roaming-EID-1 may get there sooner if it is traveling faster than 466 roaming-EID-2). 468 When a roaming-EID is a multicast source, procedures from [RFC8378] 469 are used to deliver packets to multicast group members anywhere in 470 the network. The solution requires no signaling to the RSUs. When 471 RSUs receive multicast packets from a roaming-EID, they do a 472 (roaming-EID,G) mapping database lookup to find the replication list 473 of ETRs to encapsulate to. 475 7. Multiple Address-Family Considerations 477 Note that roaming-EIDs can be assigned IPv6 EID addresses while the 478 RSU xTRs could be using IPv4 RLOC addresses. Any combination of 479 address-families can be supported as well as for multicast packet 480 forwarding, where (S,G) are IPv6 addresses entries and replication is 481 done with IPv4 RLOCs in the outer header. 483 8. Scaling Considerations 485 One can imagine there will be a large number of roaming-EIDs. So 486 there is a strong desire to efficiently store state in the mapping 487 database and the in remote ITRs map-caches. It is likely, that 488 roaming-EIDs may share the same path and move at the same speed (EID 489 devices on a train) and therefore share the same Predictive RLOCs. 490 And since EIDs are not reassigned for mobility purposes or may be 491 temporal , they will not be topologically aggregatable, so they 492 cannot compress into a single EID-prefix mapping entry that share the 493 same RLOC-set. 495 By using a level of indirection with the mapping system this problem 496 can be solved. The following mapping entries could exist in the 497 mapping database: 499 EID = , RLOC-records: 500 RLOC = (afi=: "am-train-to-paris") 501 EID = , RLOC-records: 502 RLOC = (afi=: "am-train-to-paris") 503 EID = , RLOC-records: 504 RLOC = (afi=: "am-train-to-paris") 506 EID = "am-train-to-paris", RLOC-records: 507 RLOC = (afi=lcaf/RLE-type: xTR-A, xTR-B, xTR-C) 509 EID = "am-train-to-paris-passengers", RLOC-records: 510 RLOC = (afi=lcaf/afi-list-type: , , ) 512 Each passenger that boards a train has their EID registered to point 513 to the name of the train "am-train-to-paris". And then the train 514 with EID "am-train-to-paris" stores the Predictive RLOC-set. When a 515 remote-ITR wants to encapsulate packets for an EID, it looks up the 516 EID in the mapping database gets the name "am-train-to-paris" 517 returned. Then the remote-ITR does another lookup for the name "am- 518 train-to-paris" to get the RLE list returned. 520 When new EIDs board the train, the RLE mapping entry does not need to 521 be modified. Only an EID-to-name mapping is registered for the 522 specific new EID. Optionally, another name "am-train-to-paris- 523 passengers" can be registered as an EID to allow mapping to all 524 specific EIDs which are on the train. This can be used for 525 inventory, billing, or security purposes. 527 This optimization comes at a cost of a 2-stage lookup. However, if 528 both sets of mapping entries are registered to the same Map-Server, a 529 combined RLOC-set could be returned. This idea is for further study. 531 9. Security Considerations 533 LISP has procedures for supporting both control-plane security 534 [I-D.ietf-lisp-sec] and data-plane security [RFC8061]. 536 10. IANA Considerations 538 At this time there are no requests for IANA. 540 11. References 542 11.1. Normative References 544 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 545 Requirement Levels", BCP 14, RFC 2119, 546 DOI 10.17487/RFC2119, March 1997, 547 . 549 [RFC6830] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "The 550 Locator/ID Separation Protocol (LISP)", RFC 6830, 551 DOI 10.17487/RFC6830, January 2013, 552 . 554 [RFC8060] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 555 Address Format (LCAF)", RFC 8060, DOI 10.17487/RFC8060, 556 February 2017, . 558 [RFC8061] Farinacci, D. and B. Weis, "Locator/ID Separation Protocol 559 (LISP) Data-Plane Confidentiality", RFC 8061, 560 DOI 10.17487/RFC8061, February 2017, 561 . 563 [RFC8378] Moreno, V. and D. Farinacci, "Signal-Free Locator/ID 564 Separation Protocol (LISP) Multicast", RFC 8378, 565 DOI 10.17487/RFC8378, May 2018, 566 . 568 11.2. Informative References 570 [I-D.farinacci-lisp-geo] 571 Farinacci, D., "LISP Geo-Coordinate Use-Cases", draft- 572 farinacci-lisp-geo-05 (work in progress), April 2018. 574 [I-D.ietf-lisp-eid-anonymity] 575 Farinacci, D., Pillay-Esnault, P., and W. Haddad, "LISP 576 EID Anonymity", draft-ietf-lisp-eid-anonymity-02 (work in 577 progress), April 2018. 579 [I-D.ietf-lisp-eid-mobility] 580 Portoles-Comeras, M., Ashtaputre, V., Moreno, V., Maino, 581 F., and D. Farinacci, "LISP L2/L3 EID Mobility Using a 582 Unified Control Plane", draft-ietf-lisp-eid-mobility-02 583 (work in progress), May 2018. 585 [I-D.ietf-lisp-sec] 586 Maino, F., Ermagan, V., Cabellos-Aparicio, A., and D. 587 Saucez, "LISP-Security (LISP-SEC)", draft-ietf-lisp-sec-15 588 (work in progress), April 2018. 590 Appendix A. Acknowledgments 592 The author would like to thank the LISP WG for their review and 593 acceptance of this draft. 595 Appendix B. Document Change Log 597 [RFC Editor: Please delete this section on publication as RFC.] 599 B.1. Changes to draft-ietf-lisp-predictive-rlocs-02 601 o Posted May 2018. 603 o Update document timer and references. 605 B.2. Changes to draft-ietf-lisp-predictive-rlocs-01 607 o Posted November 2017. 609 o Update document timer and references. 611 o Reflect comments from Prague meeting presentation. There is no 612 need for "RLE Usage Types" as suggested. The ITR can treat what 613 RLOCs it replicates to as a local matter via implementation 614 configuration. RLE Directional is default. Circular rotation, 615 back-n-forth, and random selection of RLOCs is up to the ITR. 617 B.3. Changes to draft-ietf-lisp-predictive-rlocs-00 619 o Posted June 2017. 621 o Make this specification a working group document. It is a copy of 622 draft-farinacci-lisp-predictive-rlocs-02. 624 B.4. Changes to draft-farinacci-lisp-predictive-rlocs-02 626 o Posted May 2017 to update document timer. 628 B.5. Changes to draft-farinacci-lisp-predictive-rlocs-01 630 o Posted November 2016 to update document timer. 632 B.6. Changes to draft-farinacci-lisp-predictive-rlocs-00 634 o Initial post April 2016. 636 Authors' Addresses 638 Dino Farinacci 639 lispers.net 640 San Jose, CA 641 USA 643 Email: farinacci@gmail.com 645 Padma Pillay-Esnault 646 Huawei Technologies 647 San Clara, CA 648 USA 650 Email: padma@huawei.com