idnits 2.17.1 draft-ietf-lisp-predictive-rlocs-05.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 (November 4, 2019) is 1635 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 577, 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explicit reference was found in the text ** Obsolete normative reference: RFC 6830 (Obsoleted by RFC 9300, RFC 9301) == Outdated reference: A later version (-15) exists of draft-farinacci-lisp-geo-08 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-eid-anonymity-07 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-eid-mobility-04 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-19 Summary: 1 error (**), 0 flaws (~~), 6 warnings (==), 1 comment (--).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: May 7, 2020 Huawei Technologies 6 November 4, 2019 8 LISP Predictive RLOCs 9 draft-ietf-lisp-predictive-rlocs-05 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 May 7, 2020. 39 Copyright Notice 41 Copyright (c) 2019 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 . . . . . . . . . . . . . . . . . . . . . . 6 61 4.2. Packet Delivery Optimizations . . . . . . . . . . . . . . 7 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 . . . . . . . . . . . . . . . . . . 14 73 Appendix B. Document Change Log . . . . . . . . . . . . . . . . 14 74 B.1. Changes to draft-ietf-lisp-predictive-rlocs-05 . . . . . 14 75 B.2. Changes to draft-ietf-lisp-predictive-rlocs-04 . . . . . 14 76 B.3. Changes to draft-ietf-lisp-predictive-rlocs-03 . . . . . 14 77 B.4. Changes to draft-ietf-lisp-predictive-rlocs-02 . . . . . 14 78 B.5. Changes to draft-ietf-lisp-predictive-rlocs-01 . . . . . 14 79 B.6. Changes to draft-ietf-lisp-predictive-rlocs-00 . . . . . 15 80 B.7. Changes to draft-farinacci-lisp-predictive-rlocs-02 . . . 15 81 B.8. Changes to draft-farinacci-lisp-predictive-rlocs-01 . . . 15 82 B.9. Changes to draft-farinacci-lisp-predictive-rlocs-00 . . . 15 83 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 15 85 1. Introduction 87 The LISP architecture [RFC6830] specifies two namespaces, End-Point 88 IDs (EIDs) and Routing Locators (RLOCs). An EID identifies a node in 89 the network and the RLOC indicates the EID's topological location. 90 When an node roams in the network, its EID remains fixed and 91 unchanged but the RLOCs associated with it change to reflect its new 92 topological attachment point. This specification will focus EIDs and 93 RLOCs residing in separate nodes. An EID is assigned to a host node 94 that roams while the RLOCs are assigned to network nodes that stay 95 stationary and are part of the network topology. For example, a set 96 of devices on an aircraft are assigned EIDs, and base stations on the 97 ground attached to the Internet infrastructure are configured as LISP 98 xTRs where their RLOCs are used for the bindings of the EIDs on the 99 aircraft up in the air. 101 The scope of this specification will not emphasize general physical 102 roaming as an aircraft would do in the sky but in a direction that is 103 more predictable such as a train traveling on a track or vehicle that 104 travels along a road. 106 2. Definition of Terms 108 Roaming-EID - is a network node that moves from one topological 109 location in the network to another. The network node uses the 110 same EID when it is roaming. That is, the EID address does not 111 change for reasons of mobility. A roaming-EID can also be a 112 roaming EID-prefix where a set of EIDs covered by the prefix are 113 all roaming and fate-sharing the same set of RLOCs at the same 114 time. 116 Predictive RLOCs - is a set of ordered RLOCs in a list each assigned 117 to LISP xTRs where the next RLOC in the list has high probability 118 it will be the next LISP xTR in a physical path going in a single 119 predictable direction. 121 Road-Side-Units (RSUs) - is a network node that acts as a router, 122 more specifically as a LISP xTR. The xTR automatically discovers 123 roaming-EIDs that come into network connectivity range and relays 124 packets to and from the roaming-EID. RSUs are typically deployed 125 along a directional path like a train track or road and are in 126 connectivity range of devices that travel along the directional 127 path. 129 3. Overview 131 The goal of this specification is to describe a make-before-break 132 EID-mobility mechanism that offers near-zero packet loss. Offering 133 minimal packet loss, not only allows transport layers to operate more 134 efficiently, but because an EID does not change while moving, 135 transport layer session continuity is maintained. To achieve these 136 requirements, a mechanism that reacts to the mobility event is 137 necessary but not sufficient. So the question is not that there 138 isn't a reaction but when it happens. By using some predictive 139 algorithms, we can guess with high probability where the EID will 140 roam to next. We can achieve this to a point where packet data will 141 be at the new location when the EID arrives. 143 First we should examine both the send and receive directions with 144 respect to the roaming-EID. Refer to Figure 1 for discussion. We 145 show a network node with a fixed EID address assigned to a roaming- 146 EID moving along a train track. And there are LISP xTRs deployed as 147 Road-Side-Units to support the connectivity between the roaming-EID 148 and the infrastructure or to another roaming-EID. 150 Roaming-EID ----> 152 ====//====//====//====//====//====//====//====//===//====//====//==== 153 // // // // // // // // // // // 154 ====//====//====//====//====//====//====//====//===//====//====//==== 156 xTR xTR xTR xTR xTR xTR 157 A B C D E F 159 Figure 1: Directional Mobility 161 For the send direction from roaming-EID to any destination can be 162 accomplish as a local decision. As long as the roaming-EID is in 163 signal range to any xTR along the path, it can use it to forward 164 packets. The LISP xTR, acting as an ITR, can forward packets to 165 destinations in non-LISP sites as well as to stationary and roaming 166 EIDs in LISP sites. This is accomplished by using the LISP overlay 167 via dynamic packet encapsulation. When the roaming-EID sends 168 packets, the LISP xTR must discover the EID and MAY register the EID 169 with a set of RLOCs to the mapping system 170 [I-D.ietf-lisp-eid-mobility]. The discovery process is important 171 because the LISP xTR, acting as an ETR for decapsulating packets that 172 arrive, needs to know what local ports or radios to send packets to 173 the roaming-EID. 175 Much of the focus of this design is on the packet direction to the 176 roaming-EID. And how remote LISP ITRs find the current location 177 (RLOCs) quickly when the roaming-EID is moving at high speed. This 178 specification solves the fast roaming with the introduction of the 179 Predictive-RLOCs algorithm. 181 Since a safe assumption is that the roaming-EID is going in one 182 direction and cannot deviate from it allows us to know a priori the 183 next set of RLOCs the roaming-EID will pass by. Referring to 184 Figure 1, if the roaming-EID is in range near xTR-A, then as it 185 moves, it will at some point pass by xTR-B and xTR-C, and so on. As 186 the roaming-EID moves, one could time when the EID is mapped to RLOC 187 A, and when it should change to RLOC B and so on. However, the speed 188 of movement of the roaming-EID won't be constant and the variables 189 involved in consistent timing cannot be relied on. Furthermore, 190 timing the move is not a make-before-break algorithm, meaning the 191 reaction of the binding happens at the time the roaming-EID is 192 discovered by an xTR. One cannot achieve fast hand-offs when message 193 signaling will be required to inform remote ITRs of the new binding. 195 The Predictive RLOCs algorithm allows a set of RLOCs, in an ordered 196 list, to be provided to remote ITRs so they have the information 197 available and local for when they need to use it. Therefore, no 198 control-plane message signaling occurs when the roaming-EID is 199 discovered by LISP xTRs. 201 4. Design Details 203 Predictive RLOCs accommodates for encapsulated packets to be 204 delivered to Road-Side-Unit LISP xTRs regardless where the roaming- 205 EID is currently positioned. 207 Referring to Figure 1, the following sequence is performed: 209 1. The Predictive RLOCs are registered to the mapping system as a 210 LCAF encoded Replication List Entry (RLE) Type [RFC8060]. The 211 registration can happen by one or more RSUs or by a third-party. 212 When registered by an RSU, and when no coordination is desired, 213 they each register their own RLOC with merge-semantics so the 214 list can be created and maintained in the LISP Map-Server. When 215 registered by a third-party, the complete list of RLOCs can be 216 included in the RLE. 218 2. There can be multiple RLEs present each as different RLOC- 219 records so a remote ITR can select one RLOC-record versus the 220 other based in priority and weight policy [RFC6830]. 222 3. When a remote ITR receives a packet destined for a roaming-EID, 223 it encapsulates and replicates to each RLOC in the RLE thereby 224 delivering the packet to the locations the roaming-EID is about 225 to appear. There are some cases where packets will go to 226 locations where the roaming-EID has already been, but see 227 Section 4.2 for packet delivery optimizations. 229 4. When the ETR resident RSU receives an encapsulated packet, it 230 decapsulates the packet and then determines if the roaming-EID 231 had been previously discovered. If the EID has not been 232 discovered, the ETR drops the packet. Otherwise, the ETR 233 delivers the decapsulated packet on the port interface the 234 roaming-EID was discovered on. 236 4.1. RLE Encoding 238 The LCAF [RFC8060] Replication List Entry (RLE) will be used to 239 encode the Predictive RLOCs in an RLOC-record for Map-Registers, Map- 240 Reply, and Map-Notify messages [RFC6830]. 242 0 1 2 3 243 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 1 244 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 245 | AFI = 16387 | Rsvd1 | Flags | 246 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 247 | Type = 13 | Rsvd2 | 4 + n | 248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 249 | Rsvd3 | Rsvd4 | Level Value | 250 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 251 | AFI = x | RTR/ETR #1 ... | 252 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 253 | Rsvd3 | Rsvd4 | Level Value | 254 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 255 | AFI = x | RTR/ETR #n ... | 256 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 258 When the RLOC-record contains an RLE with RLOC entries all with the 259 same level value, it means the physical order listed is the 260 directional path of the RSUs. This will typically be the result of a 261 third-party doing the registration where it knows ahead of time the 262 RSU deployment. 264 When each RSU is registering with merge-semantics on their own, the 265 level number is used to place them in an ordered list. Since the 266 registrations come at different times and therefore arrive in 267 different order than the physical RSU path, the level number creates 268 the necessary sequencing. Each RSU needs to know its position in the 269 path relative to other RSUs. For example, in xTR-B, it would 270 register with level 1 since it is after xTR-A (and before xTR-C). So 271 if the registration order was xTR-B with level 1, xTR-C with level 2, 272 and xTR-A with level 0, the RLE list stored in the mapping system 273 would be (xTR-A, xTR-B, xTR-C). It is recommended that level numbers 274 be assigned in increments of 10 so latter insertion is possible. 276 The use of Geo-Prefixes and Geo-Points [I-D.farinacci-lisp-geo] can 277 be used to compare the physical presence of each RSU with respect to 278 each other, so they can choose level numbers to sequence themselves. 279 Also if the xTRs register with a Geo-Point in an RLOC-record, then 280 perhaps the Map-Server could sequence the RLE list. 282 4.2. Packet Delivery Optimizations 284 Since the remote ITR will replicate to all RLOCs in the RLE, a 285 situation is created where packets go to RLOCs that don't need to. 286 For instance, if the roaming-EID is along side of xTR-B and the RLE 287 is (xTR-A, xTR-B, xTR-C), there is no reason to replicate to xTR-A 288 since the roaming-EID has passed it and the the signal range is weak 289 or lost. However, replicating to xTR-B and xTR-C is important to 290 deliver packets to where the roaming-EID resides and where it is 291 about to go to. 293 A simple data-plane option, which converges fairly quickly is to have 294 the remote xTR, acting as an ETR, when packets are sent from the 295 roaming-EID, examine the source RLOC in the outer header of the 296 encapsulated packet. If the source RLOC is xTR-B, the remote xTR can 297 determine that the roaming-EID has moved past xTR-A and no longer 298 needs to encapsulate packets to xTR-A's RLOC. 300 In addition, the remote ITR can use RLOC-probing to determine if each 301 RLOC in the RLE is reachable. And if not reachable, exclude from the 302 list of RLOCs to replicate to. 304 This solution also handles the case where xTR-A and xTR-B may overlap 305 in radio signal range, but the signal is weak from the roaming-EID to 306 xTR-A but stronger to xTR-B. In this case, the roaming-EID selects 307 xTR-B to send packets that inform the remote xTR that return packets 308 should not be encapsulated to xTR-A. 310 There are also situations where the RSUs are in signal range of each 311 other in which case they could report reachability status of each 312 other. The use of the Locator-Status-Bits of the LISP encapsulation 313 header could be used to convey this information to the remote xTR. 314 This would only occur when the roaming-EID was discovered by both 315 xTR-A and xTR-B so it was possible for either xTR to reach the 316 roaming-EID. Either an IGP like routing protocol would be required 317 to allow each xTR to know the other could reach the roaming-EID or a 318 path trace tool (i.e. traceroute) could be originated by one xTR 319 targeted for the roaming-EID but MAC-forwarded through the other xTR. 320 These and other roaming-EID reachability mechanisms are work in 321 progress and for further study. 323 When a remote ITR is doing "Directional Mobility" and replicating to 324 the last RLOC in the RLE list, it has a decision to guess where the 325 roaming-EID will move to next. Conservatively, an ITR can replicate 326 to the entire set of RLOCs in the RLE list and wait to see if the 327 roaming-EID moves to one of the RLOCs in the RLE list. 329 Or more liberally, the remote ITR can assume the new roaming 330 direction. For example, with an RLE list of (xTR-A, xTR-B, xTR-C, 331 xTR-D) and the roaming-EID is at D, the remote ITR can replicate to 332 all of A, B, C and D to determine where the roaming-EID will move to 333 next. If the roaming-EID moves to C after it was at D, it is 334 possible that the EID is moving in the opposite direction from C to B 335 to A. This would be known as "Back-n-Forth Mobility". If an 336 implementation is configured to support this for a particluar EID, 337 the remote ITR could replicate in this sequence as the roaming-EID 338 moves from A to D and back to A: (A, B, C, D), (B, C, D), (C, D), (D, 339 C, B, A), (C, B, A), (B, A), and again (A, B, C, D). 341 The roaming-EID could be doing "Circular Mobility" where it moves 342 from A to D directionally, next from D-to-A, and then back to A to D 343 directionally again. This form of mobility is just as predicatable 344 as "Back-n-Forth Mobility" since a consistent pattern can be relied 345 on. Both of these mobility forms can be achieved with near-zero 346 packet loss. 348 On the other hand, the roaming-EID can be roaming arbitrarily using 349 "Random Mobility" where it could roam in the following combinations: 350 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- 351 D, D-to-A, D-to-B, or D-to-C. In this situation, when a return 352 packet arrives at the ITR, it could then just replicate to where the 353 roaming-EID is at rest and do so for a short period of time before it 354 replciates to the entire RLE list again. Using the wrong time period 355 could lead to packet loss. All these types of mobility can be 356 supported by the remote ITR in a local manner without consulting or 357 depending on any other LISP system. It is left for further study, if 358 any of the mobility types above should be encoded in the mapping 359 system. 361 4.3. Trading Off Replication Cost 363 If RLE lists are large, packet replication can occur to locations 364 well before the roaming-EID arrives. Making RLE lists small is 365 useful without sacrificing hand-off issues or incurring packet loss 366 to the application. By having overlapping RLEs in separate RLOC- 367 records we a simple mechanism to solve this problem. Here is an 368 example mapping entry to illustrate the point: 370 EID = , RLOC-records: 371 RLOC = (RLE: xTR-A, xTR-B) 372 RLOC = (RLE: xTR-B, xTR-C, xTR-D, xTR-E) 373 RLOC = (RLE: xTR-E, xTR-F) 375 When the remote ITR is encapsulating to xTR-B as a decision to use 376 the first RLOC-record, it can decide to move to use the second RLOC- 377 record because xTR-B is the last entry in the first RLOC-record and 378 the first entry in the second RLOC-record. When there are 379 overlapping RLEs, the remote ITR can decide when it is more efficient 380 to switch over. For example, when the roaming-EID is in range of 381 xTR-A, the remote ITR uses the first RLOC-record so the wasted 382 replication cost is to xTR-B only versus a worse cost when using the 383 second RLOC-record. But when the roaming-EID is in range of xTR-B, 384 then replicating to the other xTRs in the second RLOC-record may be 385 crucial if the roaming-EID has increased speed. And when the 386 roaming-EID may be at rest in a parked mode, then the remote ITR 387 encapsulates to only xTR-F using the third RLOC-record since the 388 roaming-EID has moved past xTR-E. 390 In addition, to eliminate unnecessary replication to xTRs further 391 down a directional path, GEO-prefixes [I-D.farinacci-lisp-geo] can be 392 used so only nearby xTRs that the roaming-EID is about to come in 393 contact with are the only ones to receive encapsulated packets. 395 Even when replication lists are not large, we can reduce the cost of 396 replication that the entire network bears by moving the replicator 397 away from the the source (i.e. the ITR) and closer to the RSUs (i.e. 398 the ETRs). See the use of RTRs for Replication Engineering 399 techniques in [RFC8378]. 401 5. Directional Paths with Intersections 403 A roaming-EID could be registered to the mapping system with the 404 following nested RLE mapping: 406 EID = , RLOC-records: 407 RLOC = (RLE: xTR-A, xTR-B, xTR-C, (RLE: xTR-X, xTR-Y, xTR-Z), 408 (RLE: xTR-I, xTR-J, xTR-K), xTR-D, xTR-E) 410 The mapping entry above describes 3 directional paths where the 411 ordered list has encoded one-level of two nested RLEs to denote 412 intersections in a horizontal path. Which is drawn as: 414 | | xTR-K 415 | | 416 | | 417 | | xTR-J 418 | | 419 | | 420 Roaming | | xTR-I 421 EID ----> | | 422 --------------------------------------- ------------------------------ 423 --------------------------------------- ------------------------------ 424 xTR-A xTR-B xTR-C | | xTR-D xTR-E 425 | | 426 | | xTR-X 427 | | 428 | | 429 | | xTR-Y 430 | | 431 | | 432 | | xTR-Z 434 When the roaming-EID is on the horizontal path, the remote-ITRs 435 typically replicate to the rest the of the xTRs in the ordered list. 436 When a list has nested RLEs, the replication should occur to at least 437 the first RLOC in a nested RLE list. So if the remote-ITR is 438 replicating to xTR-C, xTR-D, and xTR-E, it should also replicate to 439 xTR-X and xTR-I anticipating a possible turn at the intersection. 440 But when the roaming-EID is known to be at xTR-D (a left or right 441 hand turn was not taken), replication should only occur to xTR-D and 442 xTR-E. Once either xTR-I or xTR-X is determined to be where the 443 roaming-EID resides, then the replication occurs on the respective 444 directional path only. 446 When nested RLEs are used it may be difficult to get merge-semantics 447 to work when each xTR registers itself. So it is suggested a third- 448 party registers nested RLEs. It is left to further study to 449 understand better how to automate this. 451 6. Multicast Considerations 453 In this design, the remote ITR is receiving a unicast packet from an 454 EID and replicating and encapsulating to each RLOC in an RLE list. 455 This form of replication is no different than a traditional multicast 456 replication function. So replicating multicast packets in the same 457 fashion is a fallout from this design. 459 If there are multiple roaming-EIDs joined to the same multicast group 460 but reside at different RSUs, a merge has to be done of any pruned 461 RLEs used for forwarding. So if roaming-EID-1 resides at xTR-A and 462 roaming-EID-2 resides at xTR-B and the RLE list is (xTR-A, xTR-B, 463 xTR-C), and they are joined to the same multicast group, then 464 replication occurs to all of xTR-A, xTR-B, and xTR-C. Even since 465 roaming-EID-2 is past xTR-A, packets need to be delivered to xTR-A 466 for roaming-EID-1. In addition, packets need to be delivered to 467 xTR-C because roaming-EID-1 and roaming-EID-2 will get to xTR-C (and 468 roaming-EID-1 may get there sooner if it is traveling faster than 469 roaming-EID-2). 471 When a roaming-EID is a multicast source, procedures from [RFC8378] 472 are used to deliver packets to multicast group members anywhere in 473 the network. The solution requires no signaling to the RSUs. When 474 RSUs receive multicast packets from a roaming-EID, they do a 475 (roaming-EID,G) mapping database lookup to find the replication list 476 of ETRs to encapsulate to. 478 7. Multiple Address-Family Considerations 480 Note that roaming-EIDs can be assigned IPv6 EID addresses while the 481 RSU xTRs could be using IPv4 RLOC addresses. Any combination of 482 address-families can be supported as well as for multicast packet 483 forwarding, where (S,G) are IPv6 addresses entries and replication is 484 done with IPv4 RLOCs in the outer header. 486 8. Scaling Considerations 488 One can imagine there will be a large number of roaming-EIDs. So 489 there is a strong desire to efficiently store state in the mapping 490 database and the in remote ITRs map-caches. It is likely, that 491 roaming-EIDs may share the same path and move at the same speed (EID 492 devices on a train) and therefore share the same Predictive RLOCs. 493 And since EIDs are not reassigned for mobility purposes or may be 494 temporal , they will not be topologically aggregatable, so they 495 cannot compress into a single EID-prefix mapping entry that share the 496 same RLOC-set. 498 By using a level of indirection with the mapping system this problem 499 can be solved. The following mapping entries could exist in the 500 mapping database: 502 EID = , RLOC-records: 503 RLOC = (afi=: "am-train-to-paris") 504 EID = , RLOC-records: 505 RLOC = (afi=: "am-train-to-paris") 506 EID = , RLOC-records: 507 RLOC = (afi=: "am-train-to-paris") 509 EID = "am-train-to-paris", RLOC-records: 510 RLOC = (afi=lcaf/RLE-type: xTR-A, xTR-B, xTR-C) 512 EID = "am-train-to-paris-passengers", RLOC-records: 513 RLOC = (afi=lcaf/afi-list-type: , , ) 515 Each passenger that boards a train has their EID registered to point 516 to the name of the train "am-train-to-paris". And then the train 517 with EID "am-train-to-paris" stores the Predictive RLOC-set. When a 518 remote-ITR wants to encapsulate packets for an EID, it looks up the 519 EID in the mapping database gets the name "am-train-to-paris" 520 returned. Then the remote-ITR does another lookup for the name "am- 521 train-to-paris" to get the RLE list returned. 523 When new EIDs board the train, the RLE mapping entry does not need to 524 be modified. Only an EID-to-name mapping is registered for the 525 specific new EID. Optionally, another name "am-train-to-paris- 526 passengers" can be registered as an EID to allow mapping to all 527 specific EIDs which are on the train. This can be used for 528 inventory, billing, or security purposes. 530 This optimization comes at a cost of a 2-stage lookup. However, if 531 both sets of mapping entries are registered to the same Map-Server, a 532 combined RLOC-set could be returned. This idea is for further study. 534 9. Security Considerations 536 LISP has procedures for supporting both control-plane security 537 [I-D.ietf-lisp-sec] and data-plane security [RFC8061]. 539 10. IANA Considerations 541 At this time there are no requests for IANA. 543 11. References 545 11.1. Normative References 547 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 548 Requirement Levels", BCP 14, RFC 2119, 549 DOI 10.17487/RFC2119, March 1997, 550 . 552 [RFC6830] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "The 553 Locator/ID Separation Protocol (LISP)", RFC 6830, 554 DOI 10.17487/RFC6830, January 2013, 555 . 557 [RFC8060] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 558 Address Format (LCAF)", RFC 8060, DOI 10.17487/RFC8060, 559 February 2017, . 561 [RFC8061] Farinacci, D. and B. Weis, "Locator/ID Separation Protocol 562 (LISP) Data-Plane Confidentiality", RFC 8061, 563 DOI 10.17487/RFC8061, February 2017, 564 . 566 [RFC8378] Moreno, V. and D. Farinacci, "Signal-Free Locator/ID 567 Separation Protocol (LISP) Multicast", RFC 8378, 568 DOI 10.17487/RFC8378, May 2018, 569 . 571 11.2. Informative References 573 [I-D.farinacci-lisp-geo] 574 Farinacci, D., "LISP Geo-Coordinate Use-Cases", draft- 575 farinacci-lisp-geo-08 (work in progress), October 2019. 577 [I-D.ietf-lisp-eid-anonymity] 578 Farinacci, D., Pillay-Esnault, P., and W. Haddad, "LISP 579 EID Anonymity", draft-ietf-lisp-eid-anonymity-07 (work in 580 progress), October 2019. 582 [I-D.ietf-lisp-eid-mobility] 583 Portoles-Comeras, M., Ashtaputre, V., Moreno, V., Maino, 584 F., and D. Farinacci, "LISP L2/L3 EID Mobility Using a 585 Unified Control Plane", draft-ietf-lisp-eid-mobility-04 586 (work in progress), May 2019. 588 [I-D.ietf-lisp-sec] 589 Maino, F., Ermagan, V., Cabellos-Aparicio, A., and D. 590 Saucez, "LISP-Security (LISP-SEC)", draft-ietf-lisp-sec-19 591 (work in progress), July 2019. 593 Appendix A. Acknowledgments 595 The author would like to thank the LISP WG for their review and 596 acceptance of this draft. 598 Appendix B. Document Change Log 600 [RFC Editor: Please delete this section on publication as RFC.] 602 B.1. Changes to draft-ietf-lisp-predictive-rlocs-05 604 o Posted November 2019. 606 o Update document timer and references. 608 B.2. Changes to draft-ietf-lisp-predictive-rlocs-04 610 o Posted May 2019. 612 o Update document timer and references. 614 B.3. Changes to draft-ietf-lisp-predictive-rlocs-03 616 o Posted November 2018. 618 o Update document timer and references. 620 B.4. Changes to draft-ietf-lisp-predictive-rlocs-02 622 o Posted May 2018. 624 o Update document timer and references. 626 B.5. Changes to draft-ietf-lisp-predictive-rlocs-01 628 o Posted November 2017. 630 o Update document timer and references. 632 o Reflect comments from Prague meeting presentation. There is no 633 need for "RLE Usage Types" as suggested. The ITR can treat what 634 RLOCs it replicates to as a local matter via implementation 635 configuration. RLE Directional is default. Circular rotation, 636 back-n-forth, and random selection of RLOCs is up to the ITR. 638 B.6. Changes to draft-ietf-lisp-predictive-rlocs-00 640 o Posted June 2017. 642 o Make this specification a working group document. It is a copy of 643 draft-farinacci-lisp-predictive-rlocs-02. 645 B.7. Changes to draft-farinacci-lisp-predictive-rlocs-02 647 o Posted May 2017 to update document timer. 649 B.8. Changes to draft-farinacci-lisp-predictive-rlocs-01 651 o Posted November 2016 to update document timer. 653 B.9. Changes to draft-farinacci-lisp-predictive-rlocs-00 655 o Initial post April 2016. 657 Authors' Addresses 659 Dino Farinacci 660 lispers.net 661 San Jose, CA 662 USA 664 Email: farinacci@gmail.com 666 Padma Pillay-Esnault 667 Huawei Technologies 668 San Clara, CA 669 USA 671 Email: padma@huawei.com