idnits 2.17.1 draft-ietf-lisp-22.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == There are 10 instances of lines with private range IPv4 addresses in the document. If these are generic example addresses, they should be changed to use any of the ranges defined in RFC 6890 (or successor): 192.0.2.x, 198.51.100.x or 203.0.113.x. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to use 'NOT RECOMMENDED' as an RFC 2119 keyword, but does not include the phrase in its RFC 2119 key words list. -- The document date (February 12, 2012) is 4450 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Outdated reference: A later version (-10) exists of draft-ietf-lisp-alt-09 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-ms-12 ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-12) exists of draft-ietf-6man-udpzero-04 == Outdated reference: A later version (-09) exists of draft-ietf-lisp-map-versioning-05 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-interworking-02 == Outdated reference: A later version (-10) exists of draft-ietf-karp-design-guide-06 == Outdated reference: A later version (-10) exists of draft-farinacci-lisp-lcaf-06 == Outdated reference: A later version (-12) exists of draft-ietf-lisp-deployment-02 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-mib-02 == Outdated reference: A later version (-16) exists of draft-meyer-lisp-mn-06 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-00 -- Unexpected draft version: The latest known version of draft-meyer-loc-id-implications is -01, but you're referring to -02. == Outdated reference: A later version (-14) exists of draft-ietf-lisp-multicast-10 == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-08 Summary: 2 errors (**), 0 flaws (~~), 16 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Farinacci 3 Internet-Draft V. Fuller 4 Intended status: Experimental D. Meyer 5 Expires: August 15, 2012 D. Lewis 6 cisco Systems 7 February 12, 2012 9 Locator/ID Separation Protocol (LISP) 10 draft-ietf-lisp-22 12 Abstract 14 This draft describes a network layer based protocol that enables 15 separation of IP addresses into two new numbering spaces: Endpoint 16 Identifiers (EIDs) and Routing Locators (RLOCs). No changes are 17 required to either host protocol stacks or to the "core" of the 18 Internet infrastructure. LISP can be incrementally deployed, without 19 a "flag day", and offers traffic engineering, multi-homing, and 20 mobility benefits to early adopters, even when there are relatively 21 few LISP-capable sites. 23 Design and development of LISP was largely motivated by the problem 24 statement produced by the October 2006 IAB Routing and Addressing 25 Workshop. 27 Status of this Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at http://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on August 15, 2012. 44 Copyright Notice 46 Copyright (c) 2012 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (http://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Requirements Notation . . . . . . . . . . . . . . . . . . . . 5 62 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 6 63 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 8 64 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . . 14 65 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . . 16 66 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . . 18 67 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 19 68 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 19 69 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . . 21 70 5.4. Dealing with Large Encapsulated Packets . . . . . . . . . 25 71 5.4.1. A Stateless Solution to MTU Handling . . . . . . . . . 25 72 5.4.2. A Stateful Solution to MTU Handling . . . . . . . . . 26 73 5.5. Using Virtualization and Segmentation with LISP . . . . . 26 74 6. EID-to-RLOC Mapping . . . . . . . . . . . . . . . . . . . . . 28 75 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats . . . . . 28 76 6.1.1. LISP Packet Type Allocations . . . . . . . . . . . . . 30 77 6.1.2. Map-Request Message Format . . . . . . . . . . . . . . 30 78 6.1.3. EID-to-RLOC UDP Map-Request Message . . . . . . . . . 33 79 6.1.4. Map-Reply Message Format . . . . . . . . . . . . . . . 34 80 6.1.5. EID-to-RLOC UDP Map-Reply Message . . . . . . . . . . 38 81 6.1.6. Map-Register Message Format . . . . . . . . . . . . . 40 82 6.1.7. Map-Notify Message Format . . . . . . . . . . . . . . 42 83 6.1.8. Encapsulated Control Message Format . . . . . . . . . 43 84 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 45 85 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 47 86 6.3.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . 49 87 6.3.2. RLOC Probing Algorithm . . . . . . . . . . . . . . . . 50 88 6.4. EID Reachability within a LISP Site . . . . . . . . . . . 51 89 6.5. Routing Locator Hashing . . . . . . . . . . . . . . . . . 52 90 6.6. Changing the Contents of EID-to-RLOC Mappings . . . . . . 53 91 6.6.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 54 92 6.6.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 54 93 6.6.3. Database Map Versioning . . . . . . . . . . . . . . . 56 94 7. Router Performance Considerations . . . . . . . . . . . . . . 57 95 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 58 96 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 59 97 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 59 98 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 60 99 8.4. LISP Functionality with Conventional NATs . . . . . . . . 60 100 8.5. Packets Egressing a LISP Site . . . . . . . . . . . . . . 61 101 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 62 102 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 63 103 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 63 104 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 63 105 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 65 106 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 65 107 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 65 108 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 65 109 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 67 110 10.5. LISP Mobile Node Mobility . . . . . . . . . . . . . . . . 67 111 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 69 112 12. Security Considerations . . . . . . . . . . . . . . . . . . . 70 113 13. Network Management Considerations . . . . . . . . . . . . . . 72 114 14. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 73 115 14.1. LISP ACT and Flag Fields . . . . . . . . . . . . . . . . . 73 116 14.2. LISP Address Type Codes . . . . . . . . . . . . . . . . . 73 117 14.3. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . . 73 118 14.4. LISP Key ID Numbers . . . . . . . . . . . . . . . . . . . 74 119 15. Known Open Issues and Areas of Future Work . . . . . . . . . . 75 120 16. References . . . . . . . . . . . . . . . . . . . . . . . . . . 77 121 16.1. Normative References . . . . . . . . . . . . . . . . . . . 77 122 16.2. Informative References . . . . . . . . . . . . . . . . . . 78 123 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 82 124 Appendix B. Document Change Log . . . . . . . . . . . . . . . . . 83 125 B.1. Changes to draft-ietf-lisp-22.txt . . . . . . . . . . . . 83 126 B.2. Changes to draft-ietf-lisp-21.txt . . . . . . . . . . . . 83 127 B.3. Changes to draft-ietf-lisp-20.txt . . . . . . . . . . . . 83 128 B.4. Changes to draft-ietf-lisp-19.txt . . . . . . . . . . . . 83 129 B.5. Changes to draft-ietf-lisp-18.txt . . . . . . . . . . . . 83 130 B.6. Changes to draft-ietf-lisp-17.txt . . . . . . . . . . . . 83 131 B.7. Changes to draft-ietf-lisp-16.txt . . . . . . . . . . . . 83 132 B.8. Changes to draft-ietf-lisp-15.txt . . . . . . . . . . . . 84 133 B.9. Changes to draft-ietf-lisp-14.txt . . . . . . . . . . . . 84 134 B.10. Changes to draft-ietf-lisp-13.txt . . . . . . . . . . . . 84 135 B.11. Changes to draft-ietf-lisp-12.txt . . . . . . . . . . . . 85 136 B.12. Changes to draft-ietf-lisp-11.txt . . . . . . . . . . . . 86 137 B.13. Changes to draft-ietf-lisp-10.txt . . . . . . . . . . . . 87 138 B.14. Changes to draft-ietf-lisp-09.txt . . . . . . . . . . . . 88 139 B.15. Changes to draft-ietf-lisp-08.txt . . . . . . . . . . . . 88 140 B.16. Changes to draft-ietf-lisp-07.txt . . . . . . . . . . . . 90 141 B.17. Changes to draft-ietf-lisp-06.txt . . . . . . . . . . . . 91 142 B.18. Changes to draft-ietf-lisp-05.txt . . . . . . . . . . . . 92 143 B.19. Changes to draft-ietf-lisp-04.txt . . . . . . . . . . . . 93 144 B.20. Changes to draft-ietf-lisp-03.txt . . . . . . . . . . . . 95 145 B.21. Changes to draft-ietf-lisp-02.txt . . . . . . . . . . . . 95 146 B.22. Changes to draft-ietf-lisp-01.txt . . . . . . . . . . . . 95 147 B.23. Changes to draft-ietf-lisp-00.txt . . . . . . . . . . . . 96 148 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 97 150 1. Requirements Notation 152 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 153 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 154 document are to be interpreted as described in [RFC2119]. 156 2. Introduction 158 This document describes the Locator/Identifier Separation Protocol 159 (LISP), which provides a set of functions for routers to exchange 160 information used to map from non globally routeable Endpoint 161 Identifiers (EIDs) to routeable Routing Locators (RLOCs). It also 162 defines a mechanism for these LISP routers to encapsulate IP packets 163 addressed with EIDs for transmission across the< Internet that uses 164 RLOCs for routing and forwarding. 166 Creation of LISP was initially motivated by discussions during the 167 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 168 October, 2006 (see [RFC4984]). A key conclusion of the workshop was 169 that the Internet routing and addressing system was not scaling well 170 in the face of the explosive growth of new sites; one reason for this 171 poor scaling is the increasing number of multi-homed and other sites 172 that cannot be addressed as part of topologically- or provider-based 173 aggregated prefixes. Additional work that more completely described 174 the problem statement may be found in [RADIR]. 176 A basic observation, made many years ago in early networking research 177 such as that documented in [CHIAPPA] and [RFC4984], is that using a 178 single address field for both identifying a device and for 179 determining where it is topologically located in the network requires 180 optimization along two conflicting axes: for routing to be efficient, 181 the address must be assigned topologically; for collections of 182 devices to be easily and effectively managed, without the need for 183 renumbering in response to topological change (such as that caused by 184 adding or removing attachment points to the network or by mobility 185 events), the address must explicitly not be tied to the topology. 187 The approach that LISP takes to solving the routing scalability 188 problem is to replace IP addresses with two new types of numbers: 189 Routing Locators (RLOCs), which are topologically assigned to network 190 attachment points (and are therefore amenable to aggregation) and 191 used for routing and forwarding of packets through the network; and 192 Endpoint Identifiers (EIDs), which are assigned independently from 193 the network topology, are used for numbering devices, and are 194 aggregated along administrative boundaries. LISP then defines 195 functions for mapping between the two numbering spaces and for 196 encapsulating traffic originated by devices using non-routeable EIDs 197 for transport across a network infrastructure that routes and 198 forwards using RLOCs. Both RLOCs and EIDs are syntactically- 199 identical to IP addresses; it is the semantics of how they are used 200 that differs. 202 This document describes the protocol that implements these functions. 203 The database which stores the mappings between EIDs and RLOCs is 204 explicitly a separate "module" to facilitate experimentation with a 205 variety of approaches. One database design that is being developed 206 for experimentation as part of the LISP working group work is [ALT]. 207 Others that have been described include [CONS], [EMACS], [NERD]. 208 Finally, [LISP-MS], documents a general-purpose service interface for 209 accessing a mapping database; this interface is intended to make the 210 mapping database modular so that different approaches can be tried 211 without the need to modify installed LISP capable devices in LISP 212 sites. 214 This experimental specification has areas that require additional 215 experience and measurement. It is NOT RECOMMENDED for deployment 216 beyond experimental situations. Results of experimentation may lead 217 to modifications and enhancements of protocol mechanisms defined in 218 this document. See Section 15 for specific, known issues that are in 219 need of further work during development, implementation, and 220 experimentation. 222 An examination of the implications of LISP on Internet traffic, 223 applications, routers, and security is for future study. This 224 analysis will explain what role LISP can play in scalable routing and 225 will also look at scalability and levels of state required for 226 encapsulation, decapsulation, liveness, and so on. 228 3. Definition of Terms 230 Provider Independent (PI) Addresses: PI addresses are an address 231 block assigned from a pool where blocks are not associated with 232 any particular location in the network (e.g. from a particular 233 service provider), and is therefore not topologically aggregatable 234 in the routing system. 236 Provider Assigned (PA) Addresses: PA addresses are an address block 237 assigned to a site by each service provider to which a site 238 connects. Typically, each block is sub-block of a service 239 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 240 is aggregated into the larger block before being advertised into 241 the global Internet. Traditionally, IP multihoming has been 242 implemented by each multi-homed site acquiring its own, globally- 243 visible prefix. LISP uses only topologically-assigned and 244 aggregatable address blocks for RLOCs, eliminating this 245 demonstrably non-scalable practice. 247 Routing Locator (RLOC): A RLOC is an IPv4 [RFC0791] or IPv6 248 [RFC2460] address of an egress tunnel router (ETR). A RLOC is the 249 output of an EID-to-RLOC mapping lookup. An EID maps to one or 250 more RLOCs. Typically, RLOCs are numbered from topologically- 251 aggregatable blocks that are assigned to a site at each point to 252 which it attaches to the global Internet; where the topology is 253 defined by the connectivity of provider networks, RLOCs can be 254 thought of as PA addresses. Multiple RLOCs can be assigned to the 255 same ETR device or to multiple ETR devices at a site. 257 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 258 IPv6) value used in the source and destination address fields of 259 the first (most inner) LISP header of a packet. The host obtains 260 a destination EID the same way it obtains an destination address 261 today, for example through a Domain Name System (DNS) [RFC1034] 262 lookup or Session Invitation Protocol (SIP) [RFC3261] exchange. 263 The source EID is obtained via existing mechanisms used to set a 264 host's "local" IP address. An EID used on the public Internet 265 must have the same properties as any other IP address used in that 266 manner; this means, among other things, that it must be globally 267 unique. An EID is allocated to a host from an EID-prefix block 268 associated with the site where the host is located. An EID can be 269 used by a host to refer to other hosts. EIDs MUST NOT be used as 270 LISP RLOCs. Note that EID blocks MAY be assigned in a 271 hierarchical manner, independent of the network topology, to 272 facilitate scaling of the mapping database. In addition, an EID 273 block assigned to a site may have site-local structure 274 (subnetting) for routing within the site; this structure is not 275 visible to the global routing system. In theory, the bit string 276 that represents an EID for one device can represent an RLOC for a 277 different device. As the architecture is realized, if a given bit 278 string is both an RLOC and an EID, it must refer to the same 279 entity in both cases. When used in discussions with other 280 Locator/ID separation proposals, a LISP EID will be called a 281 "LEID". Throughout this document, any references to "EID" refers 282 to an LEID. 284 EID-prefix: An EID-prefix is a power-of-two block of EIDs which are 285 allocated to a site by an address allocation authority. EID- 286 prefixes are associated with a set of RLOC addresses which make up 287 a "database mapping". EID-prefix allocations can be broken up 288 into smaller blocks when an RLOC set is to be associated with the 289 larger EID-prefix block. A globally routed address block (whether 290 PI or PA) is not inherently an EID-prefix. A globally routed 291 address block MAY be used by its assignee as an EID block. The 292 converse is not supported. That is, a site which receives an 293 explicitly allocated EID-prefix may not use that EID-prefix as a 294 globally routed prefix. This would require coordination and 295 cooperation with the entities managing the mapping infrastructure. 296 Once this has been done, that block could be removed from the 297 globally routed IP system, if other suitable transition and access 298 mechanisms are in place. Discussion of such transition and access 299 mechanisms can be found in [INTERWORK] and [LISP-DEPLOY]. 301 End-system: An end-system is an IPv4 or IPv6 device that originates 302 packets with a single IPv4 or IPv6 header. The end-system 303 supplies an EID value for the destination address field of the IP 304 header when communicating globally (i.e. outside of its routing 305 domain). An end-system can be a host computer, a switch or router 306 device, or any network appliance. 308 Ingress Tunnel Router (ITR): An ITR is a router that resides in a 309 LISP site. Packets sent by sources inside of the LISP site to 310 destinations outside of the site are candidates for encapsulation 311 by the ITR. The ITR treats the IP destination address as an EID 312 and performs an EID-to-RLOC mapping lookup. The router then 313 prepends an "outer" IP header with one of its globally-routable 314 RLOCs in the source address field and the result of the mapping 315 lookup in the destination address field. Note that this 316 destination RLOC MAY be an intermediate, proxy device that has 317 better knowledge of the EID-to-RLOC mapping closer to the 318 destination EID. In general, an ITR receives IP packets from site 319 end-systems on one side and sends LISP-encapsulated IP packets 320 toward the Internet on the other side. 322 Specifically, when a service provider prepends a LISP header for 323 Traffic Engineering purposes, the router that does this is also 324 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 325 on the outer destination address (the originating ITR's supplied 326 RLOC) or the inner destination address (the originating hosts 327 supplied EID). 329 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 330 network that prepends an additional LISP header for Traffic 331 Engineering purposes. 333 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 334 packet where the destination address in the "outer" IP header is 335 one of its own RLOCs. The router strips the "outer" header and 336 forwards the packet based on the next IP header found. In 337 general, an ETR receives LISP-encapsulated IP packets from the 338 Internet on one side and sends decapsulated IP packets to site 339 end-systems on the other side. ETR functionality does not have to 340 be limited to a router device. A server host can be the endpoint 341 of a LISP tunnel as well. 343 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 344 network that strips an outer LISP header for Traffic Engineering 345 purposes. 347 xTR: A xTR is a reference to an ITR or ETR when direction of data 348 flow is not part of the context description. xTR refers to the 349 router that is the tunnel endpoint. Used synonymously with the 350 term "Tunnel Router". For example, "An xTR can be located at the 351 Customer Edge (CE) router", meaning both ITR and ETR functionality 352 is at the CE router. 354 LISP Router: A LISP router is a router that performs the functions 355 of any or all of ITR, ETR, PITR, or PETR. 357 EID-to-RLOC Cache: The EID-to-RLOC cache is a short-lived, on- 358 demand table in an ITR that stores, tracks, and is responsible for 359 timing-out and otherwise validating EID-to-RLOC mappings. This 360 cache is distinct from the full "database" of EID-to-RLOC 361 mappings, it is dynamic, local to the ITR(s), and relatively small 362 while the database is distributed, relatively static, and much 363 more global in scope. 365 EID-to-RLOC Database: The EID-to-RLOC database is a global 366 distributed database that contains all known EID-prefix to RLOC 367 mappings. Each potential ETR typically contains a small piece of 368 the database: the EID-to-RLOC mappings for the EID prefixes 369 "behind" the router. These map to one of the router's own, 370 globally-visible, IP addresses. The same database mapping entries 371 MUST be configured on all ETRs for a given site. In a steady 372 state the EID-prefixes for the site and the locator-set for each 373 EID-prefix MUST be the same on all ETRs. Procedures to enforce 374 and/or verify this are outside the scope of this document. Note 375 that there MAY be transient conditions when the EID-prefix for the 376 site and locator-set for each EID-prefix may not be the same on 377 all ETRs. This has no negative implications since a partial set 378 of locators can be used. 380 Recursive Tunneling: Recursive tunneling occurs when a packet has 381 more than one LISP IP header. Additional layers of tunneling MAY 382 be employed to implement traffic engineering or other re-routing 383 as needed. When this is done, an additional "outer" LISP header 384 is added and the original RLOCs are preserved in the "inner" 385 header. Any references to tunnels in this specification refers to 386 dynamic encapsulating tunnels and they are never statically 387 configured. 389 Reencapsulating Tunnels: Reencapsulating tunneling occurs when an 390 ETR removes a LISP header, then acts as an ITR to prepend another 391 LISP header. Doing this allows a packet to be re-routed by the 392 re-encapsulating router without adding the overhead of additional 393 tunnel headers. Any references to tunnels in this specification 394 refers to dynamic encapsulating tunnels and they are never 395 statically configured. When using multiple mapping database 396 systems, care must be taken to not create reencapsulation loops 397 through misconfiguration. 399 LISP Header: a term used in this document to refer to the outer 400 IPv4 or IPv6 header, a UDP header, and a LISP-specific 8-octet 401 header that follows the UDP header, an ITR prepends or an ETR 402 strips. 404 Address Family Identifier (AFI): a term used to describe an address 405 encoding in a packet. An address family currently pertains to an 406 IPv4 or IPv6 address. See [AFI]/[AFI-REGISTRY] and [RFC3232] for 407 details. An AFI value of 0 used in this specification indicates 408 an unspecified encoded address where the length of the address is 409 0 octets following the 16-bit AFI value of 0. 411 Negative Mapping Entry: A negative mapping entry, also known as a 412 negative cache entry, is an EID-to-RLOC entry where an EID-prefix 413 is advertised or stored with no RLOCs. That is, the locator-set 414 for the EID-to-RLOC entry is empty or has an encoded locator count 415 of 0. This type of entry could be used to describe a prefix from 416 a non-LISP site, which is explicitly not in the mapping database. 417 There are a set of well defined actions that are encoded in a 418 Negative Map-Reply (Section 6.1.5). 420 Data Probe: A data-probe is a LISP-encapsulated data packet where 421 the inner header destination address equals the outer header 422 destination address used to trigger a Map-Reply by a decapsulating 423 ETR. In addition, the original packet is decapsulated and 424 delivered to the destination host if the destination EID is in the 425 EID-prefix range configured on the ETR. Otherwise, the packet is 426 discarded. A Data Probe is used in some of the mapping database 427 designs to "probe" or request a Map-Reply from an ETR; in other 428 cases, Map-Requests are used. See each mapping database design 429 for details. When using Data Probes, by sending Map-Requests on 430 the underlying routing system, EID-prefixes must be advertised. 431 However, this is discouraged if the core is to scale by having 432 less EID-prefixes stored in the core router's routing tables. 434 Proxy ITR (PITR): A PITR is defined and described in [INTERWORK], a 435 PITR acts like an ITR but does so on behalf of non-LISP sites 436 which send packets to destinations at LISP sites. 438 Proxy ETR (PETR): A PETR is defined and described in [INTERWORK], a 439 PETR acts like an ETR but does so on behalf of LISP sites which 440 send packets to destinations at non-LISP sites. 442 Route-returnability: is an assumption that the underlying routing 443 system will deliver packets to the destination. When combined 444 with a nonce that is provided by a sender and returned by a 445 receiver, this limits off-path data insertion. A route- 446 returnability check is verified when a message is sent with a 447 nonce, another message is returned with the same nonce, and the 448 destination of the original message appears as the source of the 449 returned message. 451 LISP site: is a set of routers in an edge network that are under a 452 single technical administration. LISP routers which reside in the 453 edge network are the demarcation points to separate the edge 454 network from the core network. 456 Client-side: a term used in this document to indicate a connection 457 initiation attempt by an EID. The ITR(s) at the LISP site are the 458 first to get involved in obtaining database map cache entries by 459 sending Map-Request messages. 461 Server-side: a term used in this document to indicate a connection 462 initiation attempt is being accepted for a destination EID. The 463 ETR(s) at the destination LISP site are the first to send Map- 464 Replies to the source site initiating the connection. The ETR(s) 465 at this destination site can obtain mappings by gleaning 466 information from Map-Requests, Data-Probes, or encapsulated 467 packets. 469 Locator Status Bits (LSBs): Locator status bits are present in the 470 LISP header. They are used by ITRs to inform ETRs about the up/ 471 down status of all ETRs at the local site. These bits are used as 472 a hint to convey up/down router status and not path reachability 473 status. The LSBs can be verified by use of one of the Locator 474 Reachability Algorithms described in Section 6.3. 476 Anycast Address: a term used in this document to refer to the same 477 IPv4 or IPv6 address configured and used on multiple systems at 478 the same time. An EID or RLOC can be an anycast address in each 479 of their own address spaces. 481 4. Basic Overview 483 One key concept of LISP is that end-systems (hosts) operate the same 484 way they do today. The IP addresses that hosts use for tracking 485 sockets, connections, and for sending and receiving packets do not 486 change. In LISP terminology, these IP addresses are called Endpoint 487 Identifiers (EIDs). 489 Routers continue to forward packets based on IP destination 490 addresses. When a packet is LISP encapsulated, these addresses are 491 referred to as Routing Locators (RLOCs). Most routers along a path 492 between two hosts will not change; they continue to perform routing/ 493 forwarding lookups on the destination addresses. For routers between 494 the source host and the ITR as well as routers from the ETR to the 495 destination host, the destination address is an EID. For the routers 496 between the ITR and the ETR, the destination address is an RLOC. 498 Another key LISP concept is the "Tunnel Router". A tunnel router 499 prepends LISP headers on host-originated packets and strips them 500 prior to final delivery to their destination. The IP addresses in 501 this "outer header" are RLOCs. During end-to-end packet exchange 502 between two Internet hosts, an ITR prepends a new LISP header to each 503 packet and an egress tunnel router strips the new header. The ITR 504 performs EID-to-RLOC lookups to determine the routing path to the 505 ETR, which has the RLOC as one of its IP addresses. 507 Some basic rules governing LISP are: 509 o End-systems (hosts) only send to addresses which are EIDs. They 510 don't know addresses are EIDs versus RLOCs but assume packets get 511 to their intended destinations. In a system where LISP is 512 deployed, LISP routers intercept EID addressed packets and assist 513 in delivering them across the network core where EIDs cannot be 514 routed. The procedure a host uses to send IP packets does not 515 change. 517 o EIDs are always IP addresses assigned to hosts. 519 o LISP routers mostly deal with Routing Locator addresses. See 520 details later in Section 4.1 to clarify what is meant by "mostly". 522 o RLOCs are always IP addresses assigned to routers; preferably, 523 topologically-oriented addresses from provider CIDR (Classless 524 Inter-Domain Routing) blocks. 526 o When a router originates packets it may use as a source address 527 either an EID or RLOC. When acting as a host (e.g. when 528 terminating a transport session such as SSH, TELNET, or SNMP), it 529 may use an EID that is explicitly assigned for that purpose. An 530 EID that identifies the router as a host MUST NOT be used as an 531 RLOC; an EID is only routable within the scope of a site. A 532 typical BGP configuration might demonstrate this "hybrid" EID/RLOC 533 usage where a router could use its "host-like" EID to terminate 534 iBGP sessions to other routers in a site while at the same time 535 using RLOCs to terminate eBGP sessions to routers outside the 536 site. 538 o Packets with EIDs in them are not expected to be delivered end-to- 539 end in the absence of an EID-to-RLOC mapping operation. They are 540 expected to be used locally for intra-site communication or to be 541 encapsulated for inter-site communication. 543 o EID prefixes are likely to be hierarchically assigned in a manner 544 which is optimized for administrative convenience and to 545 facilitate scaling of the EID-to-RLOC mapping database. The 546 hierarchy is based on a address allocation hierarchy which is 547 independent of the network topology. 549 o EIDs may also be structured (subnetted) in a manner suitable for 550 local routing within an autonomous system. 552 An additional LISP header MAY be prepended to packets by a TE-ITR 553 when re-routing of the path for a packet is desired. A potential 554 use-case for this would be an ISP router that needs to perform 555 traffic engineering for packets flowing through its network. In such 556 a situation, termed Recursive Tunneling, an ISP transit acts as an 557 additional ingress tunnel router and the RLOC it uses for the new 558 prepended header would be either a TE-ETR within the ISP (along 559 intra-ISP traffic engineered path) or a TE-ETR within another ISP (an 560 inter-ISP traffic engineered path, where an agreement to build such a 561 path exists). 563 In order to avoid excessive packet overhead as well as possible 564 encapsulation loops, this document mandates that a maximum of two 565 LISP headers can be prepended to a packet. For initial LISP 566 deployments, it is assumed two headers is sufficient, where the first 567 prepended header is used at a site for Location/Identity separation 568 and second prepended header is used inside a service provider for 569 Traffic Engineering purposes. 571 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 572 For example, the ITR for a particular end-to-end packet exchange 573 might be the first-hop or default router within a site for the source 574 host. Similarly, the egress tunnel router might be the last-hop 575 router directly-connected to the destination host. Another example, 576 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 577 could be the site's border router at the service provider attachment 578 point. Mixing and matching of site-operated, ISP-operated, and other 579 tunnel routers is allowed for maximum flexibility. See Section 8 for 580 more details. 582 4.1. Packet Flow Sequence 584 This section provides an example of the unicast packet flow with the 585 following conditions: 587 o Source host "host1.abc.example.com" is sending a packet to 588 "host2.xyz.example.com", exactly what host1 would do if the site 589 was not using LISP. 591 o Each site is multi-homed, so each tunnel router has an address 592 (RLOC) assigned from the service provider address block for each 593 provider to which that particular tunnel router is attached. 595 o The ITR(s) and ETR(s) are directly connected to the source and 596 destination, respectively, but the source and destination can be 597 located anywhere in LISP site. 599 o Map-Requests can be sent on the underlying routing system 600 topology, to a mapping database system, or directly over an 601 alternative topology [ALT]. A Map-Request is sent for an external 602 destination when the destination is not found in the forwarding 603 table or matches a default route. 605 o Map-Replies are sent on the underlying routing system topology. 607 Client host1.abc.example.com wants to communicate with server 608 host2.xyz.example.com: 610 1. host1.abc.example.com wants to open a TCP connection to 611 host2.xyz.example.com. It does a DNS lookup on 612 host2.xyz.example.com. An A/AAAA record is returned. This 613 address is the destination EID. The locally-assigned address of 614 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 615 packet is built and forwarded through the LISP site as a normal 616 IP packet until it reaches a LISP ITR. 618 2. The LISP ITR must be able to map the destination EID to an RLOC 619 of one of the ETRs at the destination site. The specific method 620 used to do this is not described in this example. See [ALT] or 621 [CONS] for possible solutions. 623 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be 624 rate-limited. 626 4. When an alternate mapping system is not in use, the Map-Request 627 packet is routed through the underlying routing system. 628 Otherwise, the Map-Request packet is routed on an alternate 629 logical topology, for example the [ALT] database mapping system. 630 In either case, when the Map-Request arrives at one of the ETRs 631 at the destination site, it will process the packet as a control 632 message. 634 5. The ETR looks at the destination EID of the Map-Request and 635 matches it against the prefixes in the ETR's configured EID-to- 636 RLOC mapping database. This is the list of EID-prefixes the ETR 637 is supporting for the site it resides in. If there is no match, 638 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 639 returned to the ITR. 641 6. The ITR receives the Map-Reply message, parses the message (to 642 check for format validity) and stores the mapping information 643 from the packet. This information is stored in the ITR's EID-to- 644 RLOC mapping cache. Note that the map cache is an on-demand 645 cache. An ITR will manage its map cache in such a way that 646 optimizes for its resource constraints. 648 7. Subsequent packets from host1.abc.example.com to 649 host2.xyz.example.com will have a LISP header prepended by the 650 ITR using the appropriate RLOC as the LISP header destination 651 address learned from the ETR. Note the packet MAY be sent to a 652 different ETR than the one which returned the Map-Reply due to 653 the source site's hashing policy or the destination site's 654 locator-set policy. 656 8. The ETR receives these packets directly (since the destination 657 address is one of its assigned IP addresses), checks the validity 658 of the addresses, strips the LISP header, and forwards packets to 659 the attached destination host. 661 In order to defer the need for a mapping lookup in the reverse 662 direction, an ETR MAY create a cache entry that maps the source EID 663 (inner header source IP address) to the source RLOC (outer header 664 source IP address) in a received LISP packet. Such a cache entry is 665 termed a "gleaned" mapping and only contains a single RLOC for the 666 EID in question. More complete information about additional RLOCs 667 SHOULD be verified by sending a LISP Map-Request for that EID. Both 668 ITR and the ETR may also influence the decision the other makes in 669 selecting an RLOC. See Section 6 for more details. 671 5. LISP Encapsulation Details 673 Since additional tunnel headers are prepended, the packet becomes 674 larger and can exceed the MTU of any link traversed from the ITR to 675 the ETR. It is RECOMMENDED in IPv4 that packets do not get 676 fragmented as they are encapsulated by the ITR. Instead, the packet 677 is dropped and an ICMP Too Big message is returned to the source. 679 This specification RECOMMENDS that implementations provide support 680 for one of the proposed fragmentation and reassembly schemes. Two 681 existing schemes are detailed in Section 5.4. 683 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 684 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 685 header is in IPv4 packet format and the other header is in IPv6 686 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 687 is in IPv6 packet format and the other header is in IPv4 packet 688 format). The next sub-sections illustrate packet formats for the 689 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6) but all 4 690 combinations MUST be supported. 692 5.1. LISP IPv4-in-IPv4 Header Format 694 0 1 2 3 695 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 1 696 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 697 / |Version| IHL |Type of Service| Total Length | 698 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 699 | | Identification |Flags| Fragment Offset | 700 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 701 OH | Time to Live | Protocol = 17 | Header Checksum | 702 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 703 | | Source Routing Locator | 704 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 705 \ | Destination Routing Locator | 706 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 707 / | Source Port = xxxx | Dest Port = 4341 | 708 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 709 \ | UDP Length | UDP Checksum | 710 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 711 L |N|L|E|V|I|flags| Nonce/Map-Version | 712 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 713 S / | Instance ID/Locator Status Bits | 714 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 715 / |Version| IHL |Type of Service| Total Length | 716 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 717 | | Identification |Flags| Fragment Offset | 718 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 719 IH | Time to Live | Protocol | Header Checksum | 720 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 721 | | Source EID | 722 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 723 \ | Destination EID | 724 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 726 5.2. LISP IPv6-in-IPv6 Header Format 728 0 1 2 3 729 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 1 730 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 731 / |Version| Traffic Class | Flow Label | 732 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 734 | | Payload Length | Next Header=17| Hop Limit | 735 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 736 | | 737 O + + 738 u | | 739 t + Source Routing Locator + 740 e | | 741 r + + 742 | | 743 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 744 d | | 745 r + + 746 | | 747 ^ + Destination Routing Locator + 748 | | | 749 \ + + 750 \ | | 751 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 752 / | Source Port = xxxx | Dest Port = 4341 | 753 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 754 \ | UDP Length | UDP Checksum | 755 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 756 L |N|L|E|V|I|flags| Nonce/Map-Version | 757 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 758 S / | Instance ID/Locator Status Bits | 759 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 760 / |Version| Traffic Class | Flow Label | 761 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 762 / | Payload Length | Next Header | Hop Limit | 763 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 764 | | 765 I + + 766 n | | 767 n + Source EID + 768 e | | 769 r + + 770 | | 771 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 772 d | | 773 r + + 774 | | 775 ^ + Destination EID + 776 \ | | 777 \ + + 778 \ | | 779 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 781 5.3. Tunnel Header Field Descriptions 783 Inner Header (IH): The inner header is the header on the datagram 784 received from the originating host. The source and destination IP 785 addresses are EIDs, [RFC0791], [RFC2460]. 787 Outer Header: (OH) The outer header is a new header prepended by an 788 ITR. The address fields contain RLOCs obtained from the ingress 789 router's EID-to-RLOC cache. The IP protocol number is "UDP (17)" 790 from [RFC0768]. The setting of the DF bit Flags field is 791 according to rules in Section 5.4.1 and Section 5.4.2. 793 UDP Header: The UDP header contains an ITR selected source port when 794 encapsulating a packet. See Section 6.5 for details on the hash 795 algorithm used to select a source port based on the 5-tuple of the 796 inner header. The destination port MUST be set to the well-known 797 IANA assigned port value 4341. 799 UDP Checksum: The UDP checksum field SHOULD be transmitted as zero 800 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 801 [UDP-TUNNELS] [UDP-ZERO]. When a packet with a zero UDP checksum 802 is received by an ETR, the ETR MUST accept the packet for 803 decapsulation. When an ITR transmits a non-zero value for the UDP 804 checksum, it MUST send a correctly computed value in this field. 805 When an ETR receives a packet with a non-zero UDP checksum, it MAY 806 choose to verify the checksum value. If it chooses to perform 807 such verification, and the verification fails, the packet MUST be 808 silently dropped. If the ETR chooses not to perform the 809 verification, or performs the verification successfully, the 810 packet MUST be accepted for decapsulation. The handling of UDP 811 checksums for all tunneling protocols, including LISP, is under 812 active discussion within the IETF. When that discussion 813 concludes, any necessary changes will be made to align LISP with 814 the outcome of the broader discussion. 816 UDP Length: The UDP length field is set for an IPv4 encapsulated 817 packet to be the sum of the inner header IPv4 Total Length plus 818 the UDP and LISP header lengths. For an IPv6 encapsulated packet, 819 the UDP length field is the sum of the inner header IPv6 Payload 820 Length, the size of the IPv6 header (40 octets), and the size of 821 the UDP and LISP headers. 823 N: The N bit is the nonce-present bit. When this bit is set to 1, 824 the low-order 24-bits of the first 32-bits of the LISP header 825 contains a Nonce. See Section 6.3.1 for details. Both N and V 826 bits MUST NOT be set in the same packet. If they are, a 827 decapsulating ETR MUST treat the "Nonce/Map-Version" field as 828 having a Nonce value present. 830 L: The L bit is the Locator Status Bits field enabled bit. When this 831 bit is set to 1, the Locator Status Bits in the second 32-bits of 832 the LISP header are in use. 834 x 1 x x 0 x x x 835 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 836 |N|L|E|V|I|flags| Nonce/Map-Version | 837 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 838 | Locator Status Bits | 839 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 841 E: The E bit is the echo-nonce-request bit. This bit MUST be ignored 842 and has no meaning when the N bit is set to 0. When the N bit is 843 set to 1 and this bit is set to 1, means an ITR is requesting for 844 the nonce value in the Nonce field to be echoed back in LISP 845 encapsulated packets when the ITR is also an ETR. See 846 Section 6.3.1 for details. 848 V: The V bit is the Map-Version present bit. When this bit is set to 849 1, the N bit MUST be 0. Refer to Section 6.6.3 for more details. 850 This bit indicates that the LISP header is encoded in this case 851 as: 853 0 x 0 1 x x x x 854 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 855 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version | 856 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 857 | Instance ID/Locator Status Bits | 858 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 860 I: The I bit is the Instance ID bit. See Section 5.5 for more 861 details. When this bit is set to 1, the Locator Status Bits field 862 is reduced to 8-bits and the high-order 24-bits are used as an 863 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 864 are transmitted as zero and ignored on receipt. The format of the 865 LISP header would look like in this case: 867 x x x x 1 x x x 868 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 869 |N|L|E|V|I|flags| Nonce/Map-Version | 870 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 871 | Instance ID | LSBs | 872 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 874 flags: The flags field is a 3-bit field is reserved for future flag 875 use. It MUST be set to 0 on transmit and MUST be ignored on 876 receipt. 878 LISP Nonce: The LISP nonce field is a 24-bit value that is randomly 879 generated by an ITR when the N-bit is set to 1. Nonce generation 880 algorithms are an implementation matter but are required to 881 generate different nonces when sending to different destinations. 882 However, the same nonce can be used for a period of time to the 883 same destination. The nonce is also used when the E-bit is set to 884 request the nonce value to be echoed by the other side when 885 packets are returned. When the E-bit is clear but the N-bit is 886 set, a remote ITR is either echoing a previously requested echo- 887 nonce or providing a random nonce. See Section 6.3.1 for more 888 details. 890 LISP Locator Status Bits (LSBs): When the L-bit is also set, the 891 locator status bits field in the LISP header is set by an ITR to 892 indicate to an ETR the up/down status of the Locators in the 893 source site. Each RLOC in a Map-Reply is assigned an ordinal 894 value from 0 to n-1 (when there are n RLOCs in a mapping entry). 895 The Locator Status Bits are numbered from 0 to n-1 from the least 896 significant bit of field. The field is 32-bits when the I-bit is 897 set to 0 and is 8 bits when the I-bit is set to 1. When a Locator 898 Status Bit is set to 1, the ITR is indicating to the ETR the RLOC 899 associated with the bit ordinal has up status. See Section 6.3 900 for details on how an ITR can determine the status of the ETRs at 901 the same site. When a site has multiple EID-prefixes which result 902 in multiple mappings (where each could have a different locator- 903 set), the Locator Status Bits setting in an encapsulated packet 904 MUST reflect the mapping for the EID-prefix that the inner-header 905 source EID address matches. If the LSB for an anycast locator is 906 set to 1, then there is at least one RLOC with that address the 907 ETR is considered 'up'. 909 When doing ITR/PITR encapsulation: 911 o The outer header Time to Live field (or Hop Limit field, in case 912 of IPv6) SHOULD be copied from the inner header Time to Live 913 field. 915 o The outer header Type of Service field (or the Traffic Class 916 field, in the case of IPv6) SHOULD be copied from the inner header 917 Type of Service field (with one exception, see below). 919 When doing ETR/PETR decapsulation: 921 o The inner header Time to Live field (or Hop Limit field, in case 922 of IPv6) SHOULD be copied from the outer header Time to Live 923 field, when the Time to Live field of the outer header is less 924 than the Time to Live of the inner header. Failing to perform 925 this check can cause the Time to Live of the inner header to 926 increment across encapsulation/decapsulation cycle. This check is 927 also performed when doing initial encapsulation when a packet 928 comes to an ITR or PITR destined for a LISP site. 930 o The inner header Type of Service field (or the Traffic Class 931 field, in the case of IPv6) SHOULD be copied from the outer header 932 Type of Service field (with one exception, see below). 934 Note if an ETR/PETR is also an ITR/PITR and choose to reencapsulate 935 after decapsulating, the net effect of this is that the new outer 936 header will carry the same Time to Live as the old outer header minus 937 1. 939 Copying the TTL serves two purposes: first, it preserves the distance 940 the host intended the packet to travel; second, and more importantly, 941 it provides for suppression of looping packets in the event there is 942 a loop of concatenated tunnels due to misconfiguration. See 943 Section 9.3 for TTL exception handling for traceroute packets. 945 The ECN field occupies bits 6 and 7 of both the IPv4 Type of Service 946 field and the IPv6 Traffic Class field [RFC3168]. The ECN field 947 requires special treatment in order to avoid discarding indications 948 of congestion [RFC3168]. ITR encapsulation MUST copy the 2-bit ECN 949 field from the inner header to the outer header. Re-encapsulation 950 MUST copy the 2-bit ECN field from the stripped outer header to the 951 new outer header. If the ECN field contains a congestion indication 952 codepoint (the value is '11', the Congestion Experienced (CE) 953 codepoint), then ETR decapsulation MUST copy the 2-bit ECN field from 954 the stripped outer header to the surviving inner header that is used 955 to forward the packet beyond the ETR. These requirements preserve 956 Congestion Experienced (CE) indications when a packet that uses ECN 957 traverses a LISP tunnel and becomes marked with a CE indication due 958 to congestion between the tunnel endpoints. 960 5.4. Dealing with Large Encapsulated Packets 962 This section proposes two mechanisms to deal with packets that exceed 963 the path MTU between the ITR and ETR. 965 It is left to the implementor to decide if the stateless or stateful 966 mechanism should be implemented. Both or neither can be used since 967 it is a local decision in the ITR regarding how to deal with MTU 968 issues, and sites can interoperate with differing mechanisms. 970 Both stateless and stateful mechanisms also apply to Reencapsulating 971 and Recursive Tunneling. So any actions below referring to an ITR 972 also apply to an TE-ITR. 974 5.4.1. A Stateless Solution to MTU Handling 976 An ITR stateless solution to handle MTU issues is described as 977 follows: 979 1. Define H to be the size, in octets, of the outer header an ITR 980 prepends to a packet. This includes the UDP and LISP header 981 lengths. 983 2. Define L to be the size, in octets, of the maximum sized packet 984 an ITR can send to an ETR without the need for the ITR or any 985 intermediate routers to fragment the packet. 987 3. Define an architectural constant S for the maximum size of a 988 packet, in octets, an ITR must receive so the effective MTU can 989 be met. That is, S = L - H. 991 When an ITR receives a packet from a site-facing interface and adds H 992 octets worth of encapsulation to yield a packet size greater than L 993 octets, it resolves the MTU issue by first splitting the original 994 packet into 2 equal-sized fragments. A LISP header is then prepended 995 to each fragment. The size of the encapsulated fragments is then 996 (S/2 + H), which is less than the ITR's estimate of the path MTU 997 between the ITR and its correspondent ETR. 999 When an ETR receives encapsulated fragments, it treats them as two 1000 individually encapsulated packets. It strips the LISP headers then 1001 forwards each fragment to the destination host of the destination 1002 site. The two fragments are reassembled at the destination host into 1003 the single IP datagram that was originated by the source host. Note 1004 that reassembly can happen at the ETR if the encapsulated packet was 1005 fragmented at or after the ITR. 1007 This behavior is performed by the ITR when the source host originates 1008 a packet with the DF field of the IP header is set to 0. When the DF 1009 field of the IP header is set to 1, or the packet is an IPv6 packet 1010 originated by the source host, the ITR will drop the packet when the 1011 size is greater than L, and sends an ICMP Too Big message to the 1012 source with a value of S, where S is (L - H). 1014 When the outer header encapsulation uses an IPv4 header, an 1015 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 1016 can be avoided. An implementation MAY set the DF bit in such headers 1017 to 0 if it has good reason to believe there are unresolvable path MTU 1018 issues between the sending ITR and the receiving ETR. 1020 This specification RECOMMENDS that L be defined as 1500. 1022 5.4.2. A Stateful Solution to MTU Handling 1024 An ITR stateful solution to handle MTU issues is described as follows 1025 and was first introduced in [OPENLISP]: 1027 1. The ITR will keep state of the effective MTU for each locator per 1028 mapping cache entry. The effective MTU is what the core network 1029 can deliver along the path between ITR and ETR. 1031 2. When an IPv6 encapsulated packet or an IPv4 encapsulated packet 1032 with DF bit set to 1, exceeds what the core network can deliver, 1033 one of the intermediate routers on the path will send an ICMP Too 1034 Big message to the ITR. The ITR will parse the ICMP message to 1035 determine which locator is affected by the effective MTU change 1036 and then record the new effective MTU value in the mapping cache 1037 entry. 1039 3. When a packet is received by the ITR from a source inside of the 1040 site and the size of the packet is greater than the effective MTU 1041 stored with the mapping cache entry associated with the 1042 destination EID the packet is for, the ITR will send an ICMP Too 1043 Big message back to the source. The packet size advertised by 1044 the ITR in the ICMP Too Big message is the effective MTU minus 1045 the LISP encapsulation length. 1047 Even though this mechanism is stateful, it has advantages over the 1048 stateless IP fragmentation mechanism, by not involving the 1049 destination host with reassembly of ITR fragmented packets. 1051 5.5. Using Virtualization and Segmentation with LISP 1053 When multiple organizations inside of a LISP site are using private 1054 addresses [RFC1918] as EID-prefixes, their address spaces MUST remain 1055 segregated due to possible address duplication. An Instance ID in 1056 the address encoding can aid in making the entire AFI based address 1057 unique. See IANA Considerations Section 14.2 for details for 1058 possible address encodings. 1060 An Instance ID can be carried in a LISP encapsulated packet. An ITR 1061 that prepends a LISP header, will copy a 24-bit value, used by the 1062 LISP router to uniquely identify the address space. The value is 1063 copied to the Instance ID field of the LISP header and the I-bit is 1064 set to 1. 1066 When an ETR decapsulates a packet, the Instance ID from the LISP 1067 header is used as a table identifier to locate the forwarding table 1068 to use for the inner destination EID lookup. 1070 For example, a 802.1Q VLAN tag or VPN identifier could be used as a 1071 24-bit Instance ID. 1073 6. EID-to-RLOC Mapping 1075 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats 1077 The following UDP packet formats are used by the LISP control-plane. 1079 0 1 2 3 1080 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 1081 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1082 |Version| IHL |Type of Service| Total Length | 1083 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1084 | Identification |Flags| Fragment Offset | 1085 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1086 | Time to Live | Protocol = 17 | Header Checksum | 1087 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1088 | Source Routing Locator | 1089 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1090 | Destination Routing Locator | 1091 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1092 / | Source Port | Dest Port | 1093 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1094 \ | UDP Length | UDP Checksum | 1095 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1096 | | 1097 | LISP Message | 1098 | | 1099 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1101 0 1 2 3 1102 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 1103 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1104 |Version| Traffic Class | Flow Label | 1105 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1106 | Payload Length | Next Header=17| Hop Limit | 1107 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1108 | | 1109 + + 1110 | | 1111 + Source Routing Locator + 1112 | | 1113 + + 1114 | | 1115 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1116 | | 1117 + + 1118 | | 1119 + Destination Routing Locator + 1120 | | 1121 + + 1122 | | 1123 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1124 / | Source Port | Dest Port | 1125 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1126 \ | UDP Length | UDP Checksum | 1127 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1128 | | 1129 | LISP Message | 1130 | | 1131 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1133 The LISP UDP-based messages are the Map-Request and Map-Reply 1134 messages. When a UDP Map-Request is sent, the UDP source port is 1135 chosen by the sender and the destination UDP port number is set to 1136 4342. When a UDP Map-Reply is sent, the source UDP port number is 1137 set to 4342 and the destination UDP port number is copied from the 1138 source port of either the Map-Request or the invoking data packet. 1139 Implementations MUST be prepared to accept packets when either the 1140 source port or destination UDP port is set to 4342 due to NATs 1141 changing port number values. 1143 The UDP Length field will reflect the length of the UDP header and 1144 the LISP Message payload. 1146 The UDP Checksum is computed and set to non-zero for Map-Request, 1147 Map-Reply, Map-Register and ECM control messages. It MUST be checked 1148 on receipt and if the checksum fails, the packet MUST be dropped. 1150 The format of control messages includes the UDP header so the 1151 checksum and length fields can be used to protect and delimit message 1152 boundaries. 1154 6.1.1. LISP Packet Type Allocations 1156 This section will be the authoritative source for allocating LISP 1157 Type values and for defining LISP control message formats. Current 1158 allocations are: 1160 Reserved: 0 b'0000' 1161 LISP Map-Request: 1 b'0001' 1162 LISP Map-Reply: 2 b'0010' 1163 LISP Map-Register: 3 b'0011' 1164 LISP Map-Notify: 4 b'0100' 1165 LISP Encapsulated Control Message: 8 b'1000' 1167 6.1.2. Map-Request Message Format 1169 0 1 2 3 1170 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 1171 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1172 |Type=1 |A|M|P|S|p|s| Reserved | IRC | Record Count | 1173 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1174 | Nonce . . . | 1175 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1176 | . . . Nonce | 1177 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1178 | Source-EID-AFI | Source EID Address ... | 1179 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1180 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... | 1181 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1182 | ... | 1183 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1184 | ITR-RLOC-AFI n | ITR-RLOC Address n ... | 1185 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1186 / | Reserved | EID mask-len | EID-prefix-AFI | 1187 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1188 \ | EID-prefix ... | 1189 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1190 | Map-Reply Record ... | 1191 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1193 Packet field descriptions: 1195 Type: 1 (Map-Request) 1197 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 1198 Requests sent by an ITR. Set to 1 when an ITR wants the 1199 destination site to return the Map-Reply rather than the mapping 1200 database system. 1202 M: This is the map-data-present bit, when set, it indicates a Map- 1203 Reply Record segment is included in the Map-Request. 1205 P: This is the probe-bit which indicates that a Map-Request SHOULD be 1206 treated as a locator reachability probe. The receiver SHOULD 1207 respond with a Map-Reply with the probe-bit set, indicating the 1208 Map-Reply is a locator reachability probe reply, with the nonce 1209 copied from the Map-Request. See Section 6.3.2 for more details. 1211 S: This is the Solicit-Map-Request (SMR) bit. See Section 6.6.2 for 1212 details. 1214 p: This is the PITR bit. This bit is set to 1 when a PITR sends a 1215 Map-Request. 1217 s: This is the SMR-invoked bit. This bit is set to 1 when an xTR is 1218 sending a Map-Request in response to a received SMR-based Map- 1219 Request. 1221 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1222 receipt. 1224 IRC: This 5-bit field is the ITR-RLOC Count which encodes the 1225 additional number of (ITR-RLOC-AFI, ITR-RLOC Address) fields 1226 present in this message. At least one (ITR-RLOC-AFI, ITR-RLOC- 1227 Address) pair MUST be encoded. Multiple ITR-RLOC Address fields 1228 are used so a Map-Replier can select which destination address to 1229 use for a Map-Reply. The IRC value ranges from 0 to 31. For a 1230 value of 0, there is 1 ITR-RLOC address encoded, and for a value 1231 of 1, there are 2 ITR-RLOC addresses encoded and so on up to 31 1232 which encodes a total of 32 ITR-RLOC addresses. 1234 Record Count: The number of records in this Map-Request message. A 1235 record is comprised of the portion of the packet that is labeled 1236 'Rec' above and occurs the number of times equal to Record Count. 1237 For this version of the protocol, a receiver MUST accept and 1238 process Map-Requests that contain one or more records, but a 1239 sender MUST only send Map-Requests containing one record. Support 1240 for requesting multiple EIDs in a single Map-Request message will 1241 be specified in a future version of the protocol. 1243 Nonce: An 8-octet random value created by the sender of the Map- 1244 Request. This nonce will be returned in the Map-Reply. The 1245 security of the LISP mapping protocol depends critically on the 1246 strength of the nonce in the Map-Request message. The nonce 1247 SHOULD be generated by a properly seeded pseudo-random (or strong 1248 random) source. See [RFC4086] for advice on generating security- 1249 sensitive random data. 1251 Source-EID-AFI: Address family of the "Source EID Address" field. 1253 Source EID Address: This is the EID of the source host which 1254 originated the packet which is caused the Map-Request. When Map- 1255 Requests are used for refreshing a map-cache entry or for RLOC- 1256 probing, an AFI value 0 is used and this field is of zero length. 1258 ITR-RLOC-AFI: Address family of the "ITR-RLOC Address" field that 1259 follows this field. 1261 ITR-RLOC Address: Used to give the ETR the option of selecting the 1262 destination address from any address family for the Map-Reply 1263 message. This address MUST be a routable RLOC address of the 1264 sender of the Map-Request message. 1266 EID mask-len: Mask length for EID prefix. 1268 EID-prefix-AFI: Address family of EID-prefix according to [AFI] 1270 EID-prefix: 4 octets if an IPv4 address-family, 16 octets if an IPv6 1271 address-family. When a Map-Request is sent by an ITR because a 1272 data packet is received for a destination where there is no 1273 mapping entry, the EID-prefix is set to the destination IP address 1274 of the data packet. And the 'EID mask-len' is set to 32 or 128 1275 for IPv4 or IPv6, respectively. When an xTR wants to query a site 1276 about the status of a mapping it already has cached, the EID- 1277 prefix used in the Map-Request has the same mask-length as the 1278 EID-prefix returned from the site when it sent a Map-Reply 1279 message. 1281 Map-Reply Record: When the M bit is set, this field is the size of a 1282 single "Record" in the Map-Reply format. This Map-Reply record 1283 contains the EID-to-RLOC mapping entry associated with the Source 1284 EID. This allows the ETR which will receive this Map-Request to 1285 cache the data if it chooses to do so. 1287 6.1.3. EID-to-RLOC UDP Map-Request Message 1289 A Map-Request is sent from an ITR when it needs a mapping for an EID, 1290 wants to test an RLOC for reachability, or wants to refresh a mapping 1291 before TTL expiration. For the initial case, the destination IP 1292 address used for the Map-Request is the data packet's destination 1293 address (i.e. the destination-EID) which had a mapping cache lookup 1294 failure. For the latter two cases, the destination IP address used 1295 for the Map-Request is one of the RLOC addresses from the locator-set 1296 of the map cache entry. The source address is either an IPv4 or IPv6 1297 RLOC address depending if the Map-Request is using an IPv4 versus 1298 IPv6 header, respectively. In all cases, the UDP source port number 1299 for the Map-Request message is an ITR/PITR selected 16-bit value and 1300 the UDP destination port number is set to the well-known destination 1301 port number 4342. A successful Map-Reply, which is one that has a 1302 nonce that matches an outstanding Map-Request nonce, will update the 1303 cached set of RLOCs associated with the EID prefix range. 1305 One or more Map-Request (ITR-RLOC-AFI, ITR-RLOC-Address) fields MUST 1306 be filled in by the ITR. The number of fields (minus 1) encoded MUST 1307 be placed in the IRC field. The ITR MAY include all locally 1308 configured locators in this list or just provide one locator address 1309 from each address family it supports. If the ITR erroneously 1310 provides no ITR-RLOC addresses, the Map-Replier MUST drop the Map- 1311 Request. 1313 Map-Requests can also be LISP encapsulated using UDP destination port 1314 4342 with a LISP type value set to "Encapsulated Control Message", 1315 when sent from an ITR to a Map-Resolver. Likewise, Map-Requests are 1316 LISP encapsulated the same way from a Map-Server to an ETR. Details 1317 on encapsulated Map-Requests and Map-Resolvers can be found in 1318 [LISP-MS]. 1320 Map-Requests MUST be rate-limited. It is RECOMMENDED that a Map- 1321 Request for the same EID-prefix be sent no more than once per second. 1323 An ITR that is configured with mapping database information (i.e. it 1324 is also an ETR) MAY optionally include those mappings in a Map- 1325 Request. When an ETR configured to accept and verify such 1326 "piggybacked" mapping data receives such a Map-Request and it does 1327 not have this mapping in the map-cache, it MAY originate a "verifying 1328 Map-Request", addressed to the map-requesting ITR and the ETR MAY add 1329 a map-cache entry. If the ETR has a map-cache entry that matches the 1330 "piggybacked" EID and the RLOC is in the locator-set for the entry, 1331 then it may send the "verifying Map-Request" directly to the 1332 originating Map-Request source. If the RLOC is not in the locator- 1333 set, then the ETR MUST send the "verifying Map-Request" to the 1334 "piggybacked" EID. Doing this forces the "verifying Map-Request" to 1335 go through the mapping database system to reach the authoritative 1336 source of information about that EID, guarding against RLOC-spoofing 1337 in in the "piggybacked" mapping data. 1339 6.1.4. Map-Reply Message Format 1341 0 1 2 3 1342 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 1343 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1344 |Type=2 |P|E|S| Reserved | Record Count | 1345 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1346 | Nonce . . . | 1347 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1348 | . . . Nonce | 1349 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1350 | | Record TTL | 1351 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1352 R | Locator Count | EID mask-len | ACT |A| Reserved | 1353 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1354 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1355 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1356 r | EID-prefix | 1357 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1358 | /| Priority | Weight | M Priority | M Weight | 1359 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1360 | o | Unused Flags |L|p|R| Loc-AFI | 1361 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1362 | \| Locator | 1363 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1365 Packet field descriptions: 1367 Type: 2 (Map-Reply) 1369 P: This is the probe-bit which indicates that the Map-Reply is in 1370 response to a locator reachability probe Map-Request. The nonce 1371 field MUST contain a copy of the nonce value from the original 1372 Map-Request. See Section 6.3.2 for more details. 1374 E: Indicates that the ETR which sends this Map-Reply message is 1375 advertising that the site is enabled for the Echo-Nonce locator 1376 reachability algorithm. See Section 6.3.1 for more details. 1378 S: This is the Security bit. When set to 1 the following 1379 authentication information will be appended to the end of the Map- 1380 Reply. The detailed format of the Authentication Data Content is 1381 for further study. 1383 0 1 2 3 1384 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 1385 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1386 | AD Type | Authentication Data Content . . . | 1387 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1389 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1390 receipt. 1392 Record Count: The number of records in this reply message. A record 1393 is comprised of that portion of the packet labeled 'Record' above 1394 and occurs the number of times equal to Record count. 1396 Nonce: A 24-bit value set in a Data-Probe packet or a 64-bit value 1397 from the Map-Request is echoed in this Nonce field of the Map- 1398 Reply. When a 24-bit value is supplied, it resides in the low- 1399 order 64 bits of the nonce field. 1401 Record TTL: The time in minutes the recipient of the Map-Reply will 1402 store the mapping. If the TTL is 0, the entry SHOULD be removed 1403 from the cache immediately. If the value is 0xffffffff, the 1404 recipient can decide locally how long to store the mapping. 1406 Locator Count: The number of Locator entries. A locator entry 1407 comprises what is labeled above as 'Loc'. The locator count can 1408 be 0 indicating there are no locators for the EID-prefix. 1410 EID mask-len: Mask length for EID prefix. 1412 ACT: This 3-bit field describes negative Map-Reply actions. In any 1413 other message type, these bits are set to 0 and ignored on 1414 receipt. These bits are used only when the 'Locator Count' field 1415 is set to 0. The action bits are encoded only in Map-Reply 1416 messages. The actions defined are used by an ITR or PITR when a 1417 destination EID matches a negative mapping cache entry. 1418 Unassigned values should cause a map-cache entry to be created 1419 and, when packets match this negative cache entry, they will be 1420 dropped. The current assigned values are: 1422 (0) No-Action: The map-cache is kept alive and no packet 1423 encapsulation occurs. 1425 (1) Natively-Forward: The packet is not encapsulated or dropped 1426 but natively forwarded. 1428 (2) Send-Map-Request: The packet invokes sending a Map-Request. 1430 (3) Drop: A packet that matches this map-cache entry is dropped. 1431 An ICMP Unreachable message SHOULD be sent. 1433 A: The Authoritative bit, when sent is always set to 1 by an ETR. 1434 When a Map-Server is proxy Map-Replying [LISP-MS] for a LISP site, 1435 the Authoritative bit is set to 0. This indicates to requesting 1436 ITRs that the Map-Reply was not originated by a LISP node managed 1437 at the site that owns the EID-prefix. 1439 Map-Version Number: When this 12-bit value is non-zero the Map-Reply 1440 sender is informing the ITR what the version number is for the 1441 EID-record contained in the Map-Reply. The ETR can allocate this 1442 number internally but MUST coordinate this value with other ETRs 1443 for the site. When this value is 0, there is no versioning 1444 information conveyed. The Map-Version Number can be included in 1445 Map-Request and Map-Register messages. See Section 6.6.3 for more 1446 details. 1448 EID-prefix-AFI: Address family of EID-prefix according to [AFI]. 1450 EID-prefix: 4 octets if an IPv4 address-family, 16 octets if an IPv6 1451 address-family. 1453 Priority: each RLOC is assigned a unicast priority. Lower values 1454 are more preferable. When multiple RLOCs have the same priority, 1455 they MAY be used in a load-split fashion. A value of 255 means 1456 the RLOC MUST NOT be used for unicast forwarding. 1458 Weight: when priorities are the same for multiple RLOCs, the weight 1459 indicates how to balance unicast traffic between them. Weight is 1460 encoded as a relative weight of total unicast packets that match 1461 the mapping entry. For example if there are 4 locators in a 1462 locator set, where the weights assigned are 30, 20, 20, and 10, 1463 the first locator will get 37.5% of the traffic, the 2nd and 3rd 1464 locators will get 25% of traffic and the 4th locator will get 1465 12.5% of the traffic. If all weights for a locator-set are equal, 1466 receiver of the Map-Reply will decide how to load-split traffic. 1467 See Section 6.5 for a suggested hash algorithm to distribute load 1468 across locators with same priority and equal weight values. 1470 M Priority: each RLOC is assigned a multicast priority used by an 1471 ETR in a receiver multicast site to select an ITR in a source 1472 multicast site for building multicast distribution trees. A value 1473 of 255 means the RLOC MUST NOT be used for joining a multicast 1474 distribution tree. For more details, see [MLISP]. 1476 M Weight: when priorities are the same for multiple RLOCs, the 1477 weight indicates how to balance building multicast distribution 1478 trees across multiple ITRs. The weight is encoded as a relative 1479 weight (similar to the unicast Weights) of total number of trees 1480 built to the source site identified by the EID-prefix. If all 1481 weights for a locator-set are equal, the receiver of the Map-Reply 1482 will decide how to distribute multicast state across ITRs. For 1483 more details, see [MLISP]. 1485 Unused Flags: set to 0 when sending and ignored on receipt. 1487 L: when this bit is set, the locator is flagged as a local locator to 1488 the ETR that is sending the Map-Reply. When a Map-Server is doing 1489 proxy Map-Replying [LISP-MS] for a LISP site, the L bit is set to 1490 0 for all locators in this locator-set. 1492 p: when this bit is set, an ETR informs the RLOC-probing ITR that the 1493 locator address, for which this bit is set, is the one being RLOC- 1494 probed and MAY be different from the source address of the Map- 1495 Reply. An ITR that RLOC-probes a particular locator, MUST use 1496 this locator for retrieving the data structure used to store the 1497 fact that the locator is reachable. The "p" bit is set for a 1498 single locator in the same locator set. If an implementation sets 1499 more than one "p" bit erroneously, the receiver of the Map-Reply 1500 MUST select the first locator. The "p" bit MUST NOT be set for 1501 locator-set records sent in Map-Request and Map-Register messages. 1503 R: set when the sender of a Map-Reply has a route to the locator in 1504 the locator data record. This receiver may find this useful to 1505 know if the locator is up but not necessarily reachable from the 1506 receiver's point of view. See also Section 6.4 for another way 1507 the R-bit may be used. 1509 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1510 assigned to an ETR. Note that the destination RLOC address MAY be 1511 an anycast address. A source RLOC can be an anycast address as 1512 well. The source or destination RLOC MUST NOT be the broadcast 1513 address (255.255.255.255 or any subnet broadcast address known to 1514 the router), and MUST NOT be a link-local multicast address. The 1515 source RLOC MUST NOT be a multicast address. The destination RLOC 1516 SHOULD be a multicast address if it is being mapped from a 1517 multicast destination EID. 1519 6.1.5. EID-to-RLOC UDP Map-Reply Message 1521 A Map-Reply returns an EID-prefix with a prefix length that is less 1522 than or equal to the EID being requested. The EID being requested is 1523 either from the destination field of an IP header of a Data-Probe or 1524 the EID record of a Map-Request. The RLOCs in the Map-Reply are 1525 globally-routable IP addresses of all ETRs for the LISP site. Each 1526 RLOC conveys status reachability but does not convey path 1527 reachability from a requesters perspective. Separate testing of path 1528 reachability is required, See Section 6.3 for details. 1530 Note that a Map-Reply may contain different EID-prefix granularity 1531 (prefix + length) than the Map-Request which triggers it. This might 1532 occur if a Map-Request were for a prefix that had been returned by an 1533 earlier Map-Reply. In such a case, the requester updates its cache 1534 with the new prefix information and granularity. For example, a 1535 requester with two cached EID-prefixes that are covered by a Map- 1536 Reply containing one, less-specific prefix, replaces the entry with 1537 the less-specific EID-prefix. Note that the reverse, replacement of 1538 one less-specific prefix with multiple more-specific prefixes, can 1539 also occur but not by removing the less-specific prefix rather by 1540 adding the more-specific prefixes which during a lookup will override 1541 the less-specific prefix. 1543 When an ETR is configured with overlapping EID-prefixes, a Map- 1544 Request with an EID that longest matches any EID-prefix MUST be 1545 returned in a single Map-Reply message. For instance, if an ETR had 1546 database mapping entries for EID-prefixes: 1548 10.0.0.0/8 1549 10.1.0.0/16 1550 10.1.1.0/24 1551 10.1.2.0/24 1553 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record 1554 count of 1 to be returned with a mapping record EID-prefix of 1555 10.1.1.0/24. 1557 A Map-Request for EID 10.1.5.5, would cause a Map-Reply with a record 1558 count of 3 to be returned with mapping records for EID-prefixes 1559 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24. 1561 Note that not all overlapping EID-prefixes need to be returned, only 1562 the more specifics (note in the second example above 10.0.0.0/8 was 1563 not returned for requesting EID 10.1.5.5) entries for the matching 1564 EID-prefix of the requesting EID. When more than one EID-prefix is 1565 returned, all SHOULD use the same Time-to-Live value so they can all 1566 time out at the same time. When a more specific EID-prefix is 1567 received later, its Time-to-Live value in the Map-Reply record can be 1568 stored even when other less specifics exist. When a less specific 1569 EID-prefix is received later, its map-cache expiration time SHOULD be 1570 set to the minimum expiration time of any more specific EID-prefix in 1571 the map-cache. This is done so the integrity of the EID-prefix set 1572 is wholly maintained so no more-specific entries are removed from the 1573 map-cache while keeping less-specific entries. 1575 Map-Replies SHOULD be sent for an EID-prefix no more often than once 1576 per second to the same requesting router. For scalability, it is 1577 expected that aggregation of EID addresses into EID-prefixes will 1578 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1579 prefix range thereby reducing the number of Map-Request messages. 1581 Map-Reply records can have an empty locator-set. A negative Map- 1582 Reply is a Map-Reply with an empty locator-set. Negative Map-Replies 1583 convey special actions by the sender to the ITR or PITR which have 1584 solicited the Map-Reply. There are two primary applications for 1585 Negative Map-Replies. The first is for a Map-Resolver to instruct an 1586 ITR or PITR when a destination is for a LISP site versus a non-LISP 1587 site. And the other is to source quench Map-Requests which are sent 1588 for non-allocated EIDs. 1590 For each Map-Reply record, the list of locators in a locator-set MUST 1591 appear in the same order for each ETR that originates a Map-Reply 1592 message. The locator-set MUST be sorted in order of ascending IP 1593 address where an IPv4 locator address is considered numerically 'less 1594 than' an IPv6 locator address. 1596 When sending a Map-Reply message, the destination address is copied 1597 from the one of the ITR-RLOC fields from the Map-Request. The ETR 1598 can choose a locator address from one of the address families it 1599 supports. For Data-Probes, the destination address of the Map-Reply 1600 is copied from the source address of the Data-Probe message which is 1601 invoking the reply. The source address of the Map-Reply is one of 1602 the local IP addresses chosen to allow uRPF checks to succeed in the 1603 upstream service provider. The destination port of a Map-Reply 1604 message is copied from the source port of the Map-Request or Data- 1605 Probe and the source port of the Map-Reply message is set to the 1606 well-known UDP port 4342. 1608 6.1.5.1. Traffic Redirection with Coarse EID-Prefixes 1610 When an ETR is misconfigured or compromised, it could return coarse 1611 EID-prefixes in Map-Reply messages it sends. The EID-prefix could 1612 cover EID-prefixes which are allocated to other sites redirecting 1613 their traffic to the locators of the compromised site. 1615 To solve this problem, there are two basic solutions that could be 1616 used. The first is to have Map-Servers proxy-map-reply on behalf of 1617 ETRs so their registered EID-prefixes are the ones returned in Map- 1618 Replies. Since the interaction between an ETR and Map-Server is 1619 secured with shared-keys, it is easier for an ETR to detect 1620 misbehavior. The second solution is to have ITRs and PITRs cache 1621 EID-prefixes with mask-lengths that are greater than or equal to a 1622 configured prefix length. This limits the damage to a specific width 1623 of any EID-prefix advertised, but needs to be coordinated with the 1624 allocation of site prefixes. These solutions can be used 1625 independently or at the same time. 1627 At the time of this writing, other approaches are being considered 1628 and researched. 1630 6.1.6. Map-Register Message Format 1632 The usage details of the Map-Register message can be found in 1633 specification [LISP-MS]. This section solely defines the message 1634 format. 1636 The message is sent in UDP with a destination UDP port of 4342 and a 1637 randomly selected UDP source port number. 1639 The Map-Register message format is: 1641 0 1 2 3 1642 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 1643 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1644 |Type=3 |P| Reserved |M| Record Count | 1645 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1646 | Nonce . . . | 1647 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1648 | . . . Nonce | 1649 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1650 | Key ID | Authentication Data Length | 1651 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1652 ~ Authentication Data ~ 1653 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1654 | | Record TTL | 1655 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1656 R | Locator Count | EID mask-len | ACT |A| Reserved | 1657 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1658 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1659 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1660 r | EID-prefix | 1661 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1662 | /| Priority | Weight | M Priority | M Weight | 1663 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1664 | o | Unused Flags |L|p|R| Loc-AFI | 1665 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1666 | \| Locator | 1667 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1669 Packet field descriptions: 1671 Type: 3 (Map-Register) 1673 P: This is the proxy-map-reply bit, when set to 1 an ETR sends a Map- 1674 Register message requesting for the Map-Server to proxy Map-Reply. 1675 The Map-Server will send non-authoritative Map-Replies on behalf 1676 of the ETR. Details on this usage can be found in [LISP-MS]. 1678 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1679 receipt. 1681 M: This is the want-map-notify bit, when set to 1 an ETR is 1682 requesting for a Map-Notify message to be returned in response to 1683 sending a Map-Register message. The Map-Notify message sent by a 1684 Map-Server is used to an acknowledge receipt of a Map-Register 1685 message. 1687 Record Count: The number of records in this Map-Register message. A 1688 record is comprised of that portion of the packet labeled 'Record' 1689 above and occurs the number of times equal to Record count. 1691 Nonce: This 8-octet Nonce field is set to 0 in Map-Register 1692 messages. Since the Map-Register message is authenticated, the 1693 nonce field is not currently used for any security function but 1694 may be in the future as part of an anti-replay solution. 1696 Key ID: A configured ID to find the configured Message 1697 Authentication Code (MAC) algorithm and key value used for the 1698 authentication function. See Section 14.4 for codepoint 1699 assignments. 1701 Authentication Data Length: The length in octets of the 1702 Authentication Data field that follows this field. The length of 1703 the Authentication Data field is dependent on the Message 1704 Authentication Code (MAC) algorithm used. The length field allows 1705 a device that doesn't know the MAC algorithm to correctly parse 1706 the packet. 1708 Authentication Data: The message digest used from the output of the 1709 Message Authentication Code (MAC) algorithm. The entire Map- 1710 Register payload is authenticated with this field preset to 0. 1711 After the MAC is computed, it is placed in this field. 1712 Implementations of this specification MUST include support for 1713 HMAC-SHA-1-96 [RFC2404] and support for HMAC-SHA-256-128 [RFC6234] 1714 is RECOMMENDED. 1716 The definition of the rest of the Map-Register can be found in the 1717 Map-Reply section. 1719 6.1.7. Map-Notify Message Format 1721 The usage details of the Map-Notify message can be found in 1722 specification [LISP-MS]. This section solely defines the message 1723 format. 1725 The message is sent inside a UDP packet with source and destination 1726 UDP ports equal to 4342. 1728 The Map-Notify message format is: 1730 0 1 2 3 1731 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 1732 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1733 |Type=4 | Reserved | Record Count | 1734 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1735 | Nonce . . . | 1736 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1737 | . . . Nonce | 1738 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1739 | Key ID | Authentication Data Length | 1740 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1741 ~ Authentication Data ~ 1742 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1743 | | Record TTL | 1744 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1745 R | Locator Count | EID mask-len | ACT |A| Reserved | 1746 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1747 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1748 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1749 r | EID-prefix | 1750 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1751 | /| Priority | Weight | M Priority | M Weight | 1752 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1753 | o | Unused Flags |L|p|R| Loc-AFI | 1754 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1755 | \| Locator | 1756 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1758 Packet field descriptions: 1760 Type: 4 (Map-Notify) 1762 The Map-Notify message has the same contents as a Map-Register 1763 message. See Map-Register section for field descriptions. 1765 6.1.8. Encapsulated Control Message Format 1767 An Encapsulated Control Message (ECM) is used to encapsulate control 1768 packets sent between xTRs and the mapping database system described 1769 in [LISP-MS]. 1771 0 1 2 3 1772 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 1773 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1774 / | IPv4 or IPv6 Header | 1775 OH | (uses RLOC addresses) | 1776 \ | | 1777 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1778 / | Source Port = xxxx | Dest Port = 4342 | 1779 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1780 \ | UDP Length | UDP Checksum | 1781 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1782 LH |Type=8 |S| Reserved | 1783 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1784 / | IPv4 or IPv6 Header | 1785 IH | (uses RLOC or EID addresses) | 1786 \ | | 1787 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1788 / | Source Port = xxxx | Dest Port = yyyy | 1789 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1790 \ | UDP Length | UDP Checksum | 1791 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1792 LCM | LISP Control Message | 1793 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1795 Packet header descriptions: 1797 OH: The outer IPv4 or IPv6 header which uses RLOC addresses in the 1798 source and destination header address fields. 1800 UDP: The outer UDP header with destination port 4342. The source 1801 port is randomly allocated. The checksum field MUST be non-zero. 1803 LH: Type 8 is defined to be a "LISP Encapsulated Control Message" 1804 and what follows is either an IPv4 or IPv6 header as encoded by 1805 the first 4 bits after the reserved field. 1807 S: This is the Security bit. When set to 1 the field following the 1808 Reserved field will have the following format. The detailed 1809 format of the Authentication Data Content is for further study. 1811 0 1 2 3 1812 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 1813 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1814 | AD Type | Authentication Data Content . . . | 1815 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1817 IH: The inner IPv4 or IPv6 header which can use either RLOC or EID 1818 addresses in the header address fields. When a Map-Request is 1819 encapsulated in this packet format the destination address in this 1820 header is an EID. 1822 UDP: The inner UDP header where the port assignments depends on the 1823 control packet being encapsulated. When the control packet is a 1824 Map-Request or Map-Register, the source port is ITR/PITR selected 1825 and the destination port is 4342. When the control packet is a 1826 Map-Reply, the source port is 4342 and the destination port is 1827 assigned from the source port of the invoking Map-Request. Port 1828 number 4341 MUST NOT be assigned to either port. The checksum 1829 field MUST be non-zero. 1831 LCM: The format is one of the control message formats described in 1832 this section. At this time, only Map-Request messages are allowed 1833 to be encapsulated. And in the future, PIM Join-Prune messages 1834 [MLISP] might be allowed. Encapsulating other types of LISP 1835 control messages are for further study. When Map-Requests are 1836 sent for RLOC-probing purposes (i.e the probe-bit is set), they 1837 MUST NOT be sent inside Encapsulated Control Messages. 1839 6.2. Routing Locator Selection 1841 Both client-side and server-side may need control over the selection 1842 of RLOCs for conversations between them. This control is achieved by 1843 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1844 messages. Alternatively, RLOC information MAY be gleaned from 1845 received tunneled packets or EID-to-RLOC Map-Request messages. 1847 The following enumerates different scenarios for choosing RLOCs and 1848 the controls that are available: 1850 o Server-side returns one RLOC. Client-side can only use one RLOC. 1851 Server-side has complete control of the selection. 1853 o Server-side returns a list of RLOC where a subset of the list has 1854 the same best priority. Client can only use the subset list 1855 according to the weighting assigned by the server-side. In this 1856 case, the server-side controls both the subset list and load- 1857 splitting across its members. The client-side can use RLOCs 1858 outside of the subset list if it determines that the subset list 1859 is unreachable (unless RLOCs are set to a Priority of 255). Some 1860 sharing of control exists: the server-side determines the 1861 destination RLOC list and load distribution while the client-side 1862 has the option of using alternatives to this list if RLOCs in the 1863 list are unreachable. 1865 o Server-side sets weight of 0 for the RLOC subset list. In this 1866 case, the client-side can choose how the traffic load is spread 1867 across the subset list. Control is shared by the server-side 1868 determining the list and the client determining load distribution. 1869 Again, the client can use alternative RLOCs if the server-provided 1870 list of RLOCs are unreachable. 1872 o Either side (more likely on the server-side ETR) decides not to 1873 send a Map-Request. For example, if the server-side ETR does not 1874 send Map-Requests, it gleans RLOCs from the client-side ITR, 1875 giving the client-side ITR responsibility for bidirectional RLOC 1876 reachability and preferability. Server-side ETR gleaning of the 1877 client-side ITR RLOC is done by caching the inner header source 1878 EID and the outer header source RLOC of received packets. The 1879 client-side ITR controls how traffic is returned and can alternate 1880 using an outer header source RLOC, which then can be added to the 1881 list the server-side ETR uses to return traffic. Since no 1882 Priority or Weights are provided using this method, the server- 1883 side ETR MUST assume each client-side ITR RLOC uses the same best 1884 Priority with a Weight of zero. In addition, since EID-prefix 1885 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1886 on tunnel routers can grow to be very large. 1888 o A "gleaned" map-cache entry, one learned from the source RLOC of a 1889 received encapsulated packet, is only stored and used for a few 1890 seconds, pending verification. Verification is performed by 1891 sending a Map-Request to the source EID (the inner header IP 1892 source address) of the received encapsulated packet. A reply to 1893 this "verifying Map-Request" is used to fully populate the map- 1894 cache entry for the "gleaned" EID and is stored and used for the 1895 time indicated from the TTL field of a received Map-Reply. When a 1896 verified map-cache entry is stored, data gleaning no longer occurs 1897 for subsequent packets which have a source EID that matches the 1898 EID-prefix of the verified entry. 1900 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1901 reachable when the R-bit for the locator record is set to 1. When 1902 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1903 RLOC. Neither the information contained in a Map-Reply or that 1904 stored in the mapping database system provides reachability 1905 information for RLOCs. Note that reachability is not part of the 1906 mapping system and is determined using one or more of the Routing 1907 Locator Reachability Algorithms described in the next section. 1909 6.3. Routing Locator Reachability 1911 Several mechanisms for determining RLOC reachability are currently 1912 defined: 1914 1. An ETR may examine the Locator Status Bits in the LISP header of 1915 an encapsulated data packet received from an ITR. If the ETR is 1916 also acting as an ITR and has traffic to return to the original 1917 ITR site, it can use this status information to help select an 1918 RLOC. 1920 2. An ITR may receive an ICMP Network or ICMP Host Unreachable 1921 message for an RLOC it is using. This indicates that the RLOC is 1922 likely down. Note, trusting ICMP messages may not be desirable 1923 but neither is ignoring them completely. Implementations are 1924 encouraged to follow current best practices in treating these 1925 conditions. 1927 3. An ITR which participates in the global routing system can 1928 determine that an RLOC is down if no BGP RIB route exists that 1929 matches the RLOC IP address. 1931 4. An ITR may receive an ICMP Port Unreachable message from a 1932 destination host. This occurs if an ITR attempts to use 1933 interworking [INTERWORK] and LISP-encapsulated data is sent to a 1934 non-LISP-capable site. 1936 5. An ITR may receive a Map-Reply from an ETR in response to a 1937 previously sent Map-Request. The RLOC source of the Map-Reply is 1938 likely up since the ETR was able to send the Map-Reply to the 1939 ITR. 1941 6. When an ETR receives an encapsulated packet from an ITR, the 1942 source RLOC from the outer header of the packet is likely up. 1944 7. An ITR/ETR pair can use the Locator Reachability Algorithms 1945 described in this section, namely Echo-Noncing or RLOC-Probing. 1947 When determining Locator up/down reachability by examining the 1948 Locator Status Bits from the LISP encapsulated data packet, an ETR 1949 will receive up to date status from an encapsulating ITR about 1950 reachability for all ETRs at the site. CE-based ITRs at the source 1951 site can determine reachability relative to each other using the site 1952 IGP as follows: 1954 o Under normal circumstances, each ITR will advertise a default 1955 route into the site IGP. 1957 o If an ITR fails or if the upstream link to its PE fails, its 1958 default route will either time-out or be withdrawn. 1960 Each ITR can thus observe the presence or lack of a default route 1961 originated by the others to determine the Locator Status Bits it sets 1962 for them. 1964 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1965 Locator Status Bits in a LISP encapsulated packet are numbered from 0 1966 to n-1 starting with the least significant bit. For example, if an 1967 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1968 value 2), then all ITRs at the site will clear the 3rd least 1969 significant bit (xxxx x0xx) of the Locator Status Bits field for the 1970 packets they encapsulate. 1972 When an ETR decapsulates a packet, it will check for any change in 1973 the Locator Status Bits field. When a bit goes from 1 to 0, the ETR 1974 if acting also as an ITR, will refrain from encapsulating packets to 1975 an RLOC that is indicated as down. It will only resume using that 1976 RLOC if the corresponding Locator Status Bit returns to a value of 1. 1977 Locator Status Bits are associated with a locator-set per EID-prefix. 1978 Therefore, when a locator becomes unreachable, the Locator Status Bit 1979 that corresponds to that locator's position in the list returned by 1980 the last Map-Reply will be set to zero for that particular EID- 1981 prefix. 1983 When ITRs at the site are not deployed in CE routers, the IGP can 1984 still be used to determine the reachability of Locators provided they 1985 are injected into the IGP. This is typically done when a /32 address 1986 is configured on a loopback interface. 1988 When ITRs receive ICMP Network or Host Unreachable messages as a 1989 method to determine unreachability, they will refrain from using 1990 Locators which are described in Locator lists of Map-Replies. 1991 However, using this approach is unreliable because many network 1992 operators turn off generation of ICMP Unreachable messages. 1994 If an ITR does receive an ICMP Network or Host Unreachable message, 1995 it MAY originate its own ICMP Unreachable message destined for the 1996 host that originated the data packet the ITR encapsulated. 1998 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1999 locator address from a locator-set in a mapping entry matches a 2000 prefix. If it does not find one and BGP is running in the Default 2001 Free Zone (DFZ), it can decide to not use the locator even though the 2002 Locator Status Bits indicate the locator is up. In this case, the 2003 path from the ITR to the ETR that is assigned the locator is not 2004 available. More details are in [LOC-ID-ARCH]. 2006 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 2007 Reply is returned, reachability of the Locator has been determined. 2008 Obviously, sending such probes increases the number of control 2009 messages originated by tunnel routers for active flows, so Locators 2010 are assumed to be reachable when they are advertised. 2012 This assumption does create a dependency: Locator unreachability is 2013 detected by the receipt of ICMP Host Unreachable messages. When an 2014 Locator has been determined to be unreachable, it is not used for 2015 active traffic; this is the same as if it were listed in a Map-Reply 2016 with priority 255. 2018 The ITR can test the reachability of the unreachable Locator by 2019 sending periodic Requests. Both Requests and Replies MUST be rate- 2020 limited. Locator reachability testing is never done with data 2021 packets since that increases the risk of packet loss for end-to-end 2022 sessions. 2024 When an ETR decapsulates a packet, it knows that it is reachable from 2025 the encapsulating ITR because that is how the packet arrived. In 2026 most cases, the ETR can also reach the ITR but cannot assume this to 2027 be true due to the possibility of path asymmetry. In the presence of 2028 unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD NOT 2029 use the lack of return traffic as an indication that the ETR is 2030 unreachable. Instead, it MUST use an alternate mechanisms to 2031 determine reachability. 2033 6.3.1. Echo Nonce Algorithm 2035 When data flows bidirectionally between locators from different 2036 sites, a data-plane mechanism called "nonce echoing" can be used to 2037 determine reachability between an ITR and ETR. When an ITR wants to 2038 solicit a nonce echo, it sets the N and E bits and places a 24-bit 2039 nonce [RFC4086] in the LISP header of the next encapsulated data 2040 packet. 2042 When this packet is received by the ETR, the encapsulated packet is 2043 forwarded as normal. When the ETR next sends a data packet to the 2044 ITR, it includes the nonce received earlier with the N bit set and E 2045 bit cleared. The ITR sees this "echoed nonce" and knows the path to 2046 and from the ETR is up. 2048 The ITR will set the E-bit and N-bit for every packet it sends while 2049 in echo-nonce-request state. The time the ITR waits to process the 2050 echoed nonce before it determines the path is unreachable is variable 2051 and a choice left for the implementation. 2053 If the ITR is receiving packets from the ETR but does not see the 2054 nonce echoed while being in echo-nonce-request state, then the path 2055 to the ETR is unreachable. This decision may be overridden by other 2056 locator reachability algorithms. Once the ITR determines the path to 2057 the ETR is down it can switch to another locator for that EID-prefix. 2059 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 2060 be implementing both ITR and ETR functionality for the echo nonce 2061 mechanism to operate. 2063 The ITR and ETR may both go into echo-nonce-request state at the same 2064 time. The number of packets sent or the time during which echo nonce 2065 requests are sent is an implementation specific setting. However, 2066 when an ITR is in echo-nonce-request state, it can echo the ETR's 2067 nonce in the next set of packets that it encapsulates and then 2068 subsequently, continue sending echo-nonce-request packets. 2070 This mechanism does not completely solve the forward path 2071 reachability problem as traffic may be unidirectional. That is, the 2072 ETR receiving traffic at a site may not be the same device as an ITR 2073 which transmits traffic from that site or the site to site traffic is 2074 unidirectional so there is no ITR returning traffic. 2076 The echo-nonce algorithm is bilateral. That is, if one side sets the 2077 E-bit and the other side is not enabled for echo-noncing, then the 2078 echoing of the nonce does not occur and the requesting side may 2079 regard the locator unreachable erroneously. An ITR SHOULD only set 2080 the E-bit in a encapsulated data packet when it knows the ETR is 2081 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 2082 Reply message. 2084 Note that other locator reachability mechanisms are being researched 2085 and can be used to compliment or even override the Echo Nonce 2086 Algorithm. See next section for an example of control-plane probing. 2088 6.3.2. RLOC Probing Algorithm 2090 RLOC Probing is a method that an ITR or PITR can use to determine the 2091 reachability status of one or more locators that it has cached in a 2092 map-cache entry. The probe-bit of the Map-Request and Map-Reply 2093 messages are used for RLOC Probing. 2095 RLOC probing is done in the control-plane on a timer basis where an 2096 ITR or PITR will originate a Map-Request destined to a locator 2097 address from one of its own locator addresses. A Map-Request used as 2098 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or on 2099 the ALT like one would when soliciting mapping data. The EID record 2100 encoded in the Map-Request is the EID-prefix of the map-cache entry 2101 cached by the ITR or PITR. The ITR may include a mapping data record 2102 for its own database mapping information which contains the local 2103 EID-prefixes and RLOCs for its site. RLOC-probes are sent 2104 periodically using a jittered timer interval. 2106 When an ETR receives a Map-Request message with the probe-bit set, it 2107 returns a Map-Reply with the probe-bit set. The source address of 2108 the Map-Reply is set according to the procedure described in 2109 Section 6.1.5. The Map-Reply SHOULD contain mapping data for the 2110 EID-prefix contained in the Map-Request. This provides the 2111 opportunity for the ITR or PITR, which sent the RLOC-probe to get 2112 mapping updates if there were changes to the ETR's database mapping 2113 entries. 2115 There are advantages and disadvantages of RLOC Probing. The greatest 2116 benefit of RLOC Probing is that it can handle many failure scenarios 2117 allowing the ITR to determine when the path to a specific locator is 2118 reachable or has become unreachable, thus providing a robust 2119 mechanism for switching to using another locator from the cached 2120 locator. RLOC Probing can also provide rough RTT estimates between a 2121 pair of locators which can be useful for network management purposes 2122 as well as for selecting low delay paths. The major disadvantage of 2123 RLOC Probing is in the number of control messages required and the 2124 amount of bandwidth used to obtain those benefits, especially if the 2125 requirement for failure detection times are very small. 2127 Continued research and testing will attempt to characterize the 2128 tradeoffs of failure detection times versus message overhead. 2130 6.4. EID Reachability within a LISP Site 2132 A site may be multihomed using two or more ETRs. The hosts and 2133 infrastructure within a site will be addressed using one or more EID 2134 prefixes that are mapped to the RLOCs of the relevant ETRs in the 2135 mapping system. One possible failure mode is for an ETR to lose 2136 reachability to one or more of the EID prefixes within its own site. 2137 When this occurs when the ETR sends Map-Replies, it can clear the 2138 R-bit associated with its own locator. And when the ETR is also an 2139 ITR, it can clear its locator-status-bit in the encapsulation data 2140 header. 2142 It is recognized there are no simple solutions to the site 2143 partitioning problem because it is hard to know which part of the 2144 EID-prefix range is partitioned. And which locators can reach any 2145 sub-ranges of the EID-prefixes. This problem is under investigation 2146 with the expectation that experiments will tell us more. Note, this 2147 is not a new problem introduced by the LISP architecture. The 2148 problem exists today when a multi-homed site uses BGP to advertise 2149 its reachability upstream. 2151 6.5. Routing Locator Hashing 2153 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 2154 a requesting ITR, the locator-set for the EID-prefix may contain 2155 different priority values for each locator address. When more than 2156 one best priority locator exists, the ITR can decide how to load 2157 share traffic against the corresponding locators. 2159 The following hash algorithm may be used by an ITR to select a 2160 locator for a packet destined to an EID for the EID-to-RLOC mapping: 2162 1. Either a source and destination address hash can be used or the 2163 traditional 5-tuple hash which includes the source and 2164 destination addresses, source and destination TCP, UDP, or SCTP 2165 port numbers and the IP protocol number field or IPv6 next- 2166 protocol fields of a packet a host originates from within a LISP 2167 site. When a packet is not a TCP, UDP, or SCTP packet, the 2168 source and destination addresses only from the header are used to 2169 compute the hash. 2171 2. Take the hash value and divide it by the number of locators 2172 stored in the locator-set for the EID-to-RLOC mapping. 2174 3. The remainder will yield a value of 0 to "number of locators 2175 minus 1". Use the remainder to select the locator in the 2176 locator-set. 2178 Note that when a packet is LISP encapsulated, the source port number 2179 in the outer UDP header needs to be set. Selecting a hashed value 2180 allows core routers which are attached to Link Aggregation Groups 2181 (LAGs) to load-split the encapsulated packets across member links of 2182 such LAGs. Otherwise, core routers would see a single flow, since 2183 packets have a source address of the ITR, for packets which are 2184 originated by different EIDs at the source site. A suggested setting 2185 for the source port number computed by an ITR is a 5-tuple hash 2186 function on the inner header, as described above. 2188 Many core router implementations use a 5-tuple hash to decide how to 2189 balance packet load across members of a LAG. The 5-tuple hash 2190 includes the source and destination addresses of the packet and the 2191 source and destination ports when the protocol number in the packet 2192 is TCP or UDP. For this reason, UDP encoding is used for LISP 2193 encapsulation. 2195 6.6. Changing the Contents of EID-to-RLOC Mappings 2197 Since the LISP architecture uses a caching scheme to retrieve and 2198 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 2199 date mapping is to re-request the mapping. However, the ITRs do not 2200 know when the mappings change and the ETRs do not keep track of which 2201 ITRs requested its mappings. For scalability reasons, we want to 2202 maintain this approach but need to provide a way for ETRs change 2203 their mappings and inform the sites that are currently communicating 2204 with the ETR site using such mappings. 2206 When adding a new locator record in lexicographic order to the end of 2207 a locator-set, it is easy to update mappings. We assume new mappings 2208 will maintain the same locator ordering as the old mapping but just 2209 have new locators appended to the end of the list. So some ITRs can 2210 have a new mapping while other ITRs have only an old mapping that is 2211 used until they time out. When an ITR has only an old mapping but 2212 detects bits set in the loc-status-bits that correspond to locators 2213 beyond the list it has cached, it simply ignores them. However, this 2214 can only happen for locator addresses that are lexicographically 2215 greater than the locator addresses in the existing locator-set. 2217 When a locator record is inserted in the middle of a locator-set, to 2218 maintain lexicographic order, the SMR procedure in Section 6.6.2 is 2219 used to inform ITRs and PITRs of the new locator-status-bit mappings. 2221 When a locator record is removed from a locator-set, ITRs that have 2222 the mapping cached will not use the removed locator because the xTRs 2223 will set the loc-status-bit to 0. So even if the locator is in the 2224 list, it will not be used. For new mapping requests, the xTRs can 2225 set the locator AFI to 0 (indicating an unspecified address), as well 2226 as setting the corresponding loc-status-bit to 0. This forces ITRs 2227 with old or new mappings to avoid using the removed locator. 2229 If many changes occur to a mapping over a long period of time, one 2230 will find empty record slots in the middle of the locator-set and new 2231 records appended to the locator-set. At some point, it would be 2232 useful to compact the locator-set so the loc-status-bit settings can 2233 be efficiently packed. 2235 We propose here three approaches for locator-set compaction, one 2236 operational and two protocol mechanisms. The operational approach 2237 uses a clock sweep method. The protocol approaches use the concept 2238 of Solicit-Map-Requests and Map-Versioning. 2240 6.6.1. Clock Sweep 2242 The clock sweep approach uses planning in advance and the use of 2243 count-down TTLs to time out mappings that have already been cached. 2244 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 2245 there is a 24 hour window to time out old mappings. The following 2246 clock sweep procedure is used: 2248 1. 24 hours before a mapping change is to take effect, a network 2249 administrator configures the ETRs at a site to start the clock 2250 sweep window. 2252 2. During the clock sweep window, ETRs continue to send Map-Reply 2253 messages with the current (unchanged) mapping records. The TTL 2254 for these mappings is set to 1 hour. 2256 3. 24 hours later, all previous cache entries will have timed out, 2257 and any active cache entries will time out within 1 hour. During 2258 this 1 hour window the ETRs continue to send Map-Reply messages 2259 with the current (unchanged) mapping records with the TTL set to 2260 1 minute. 2262 4. At the end of the 1 hour window, the ETRs will send Map-Reply 2263 messages with the new (changed) mapping records. So any active 2264 caches can get the new mapping contents right away if not cached, 2265 or in 1 minute if they had the mapping cached. The new mappings 2266 are cached with a time to live equal to the TTL in the Map-Reply. 2268 6.6.2. Solicit-Map-Request (SMR) 2270 Soliciting a Map-Request is a selective way for ETRs, at the site 2271 where mappings change, to control the rate they receive requests for 2272 Map-Reply messages. SMRs are also used to tell remote ITRs to update 2273 the mappings they have cached. 2275 Since the ETRs don't keep track of remote ITRs that have cached their 2276 mappings, they do not know which ITRs need to have their mappings 2277 updated. As a result, an ETR will solicit Map-Requests (called an 2278 SMR message) from those sites to which it has been sending 2279 encapsulated data to for the last minute. In particular, an ETR will 2280 send an SMR an ITR to which it has recently sent encapsulated data. 2282 An SMR message is simply a bit set in a Map-Request message. An ITR 2283 or PITR will send a Map-Request when they receive an SMR message. 2284 Both the SMR sender and the Map-Request responder MUST rate-limited 2285 these messages. Rate-limiting can be implemented as a global rate- 2286 limiter or one rate-limiter per SMR destination. 2288 The following procedure shows how a SMR exchange occurs when a site 2289 is doing locator-set compaction for an EID-to-RLOC mapping: 2291 1. When the database mappings in an ETR change, the ETRs at the site 2292 begin to send Map-Requests with the SMR bit set for each locator 2293 in each map-cache entry the ETR caches. 2295 2. A remote ITR which receives the SMR message will schedule sending 2296 a Map-Request message to the source locator address of the SMR 2297 message or to the mapping database system. A newly allocated 2298 random nonce is selected and the EID-prefix used is the one 2299 copied from the SMR message. If the source locator is the only 2300 locator in the cached locator-set, the remote ITR SHOULD send a 2301 Map-Request to the database mapping system just in case the 2302 single locator has changed and may no longer be reachable to 2303 accept the Map-Request. 2305 3. The remote ITR MUST rate-limit the Map-Request until it gets a 2306 Map-Reply while continuing to use the cached mapping. When Map 2307 Versioning is used, described in Section 6.6.3, an SMR sender can 2308 detect if an ITR is using the most up to date database mapping. 2310 4. The ETRs at the site with the changed mapping will reply to the 2311 Map-Request with a Map-Reply message that has a nonce from the 2312 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate 2313 limited. This is important to avoid Map-Reply implosion. 2315 5. The ETRs, at the site with the changed mapping, record the fact 2316 that the site that sent the Map-Request has received the new 2317 mapping data in the mapping cache entry for the remote site so 2318 the loc-status-bits are reflective of the new mapping for packets 2319 going to the remote site. The ETR then stops sending SMR 2320 messages. 2322 Experimentation is in progress to determine the appropriate rate- 2323 limit parameters. 2325 For security reasons an ITR MUST NOT process unsolicited Map-Replies. 2326 To avoid map-cache entry corruption by a third-party, a sender of an 2327 SMR-based Map-Request MUST be verified. If an ITR receives an SMR- 2328 based Map-Request and the source is not in the locator-set for the 2329 stored map-cache entry, then the responding Map-Request MUST be sent 2330 with an EID destination to the mapping database system. Since the 2331 mapping database system is more secure to reach an authoritative ETR, 2332 it will deliver the Map-Request to the authoritative source of the 2333 mapping data. 2335 When an ITR receives an SMR-based Map-Request for which it does not 2336 have a cached mapping for the EID in the SMR message, it MAY not send 2337 a SMR-invoked Map-Request. This scenario can occur when an ETR sends 2338 SMR messages to all locators in the locator-set it has stored in its 2339 map-cache but the remote ITRs that receive the SMR may not be sending 2340 packets to the site. There is no point in updating the ITRs until 2341 they need to send, in which case, they will send Map-Requests to 2342 obtain a map-cache entry. 2344 6.6.3. Database Map Versioning 2346 When there is unidirectional packet flow between an ITR and ETR, and 2347 the EID-to-RLOC mappings change on the ETR, it needs to inform the 2348 ITR so encapsulation can stop to a removed locator and start to a new 2349 locator in the locator-set. 2351 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 2352 number. This is known as the Destination Map-Version Number. ITRs 2353 include the Destination Map-Version Number in packets they 2354 encapsulate to the site. When an ETR decapsulates a packet and 2355 detects the Destination Map-Version Number is less than the current 2356 version for its mapping, the SMR procedure described in Section 6.6.2 2357 occurs. 2359 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 2360 Version number. This is known as the Source Map-Version Number. 2361 When an ETR decapsulates a packet and detects the Source Map-Version 2362 Number is greater than the last Map-Version Number sent in a Map- 2363 Reply from the ITR's site, the ETR will send a Map-Request to one of 2364 the ETRs for the source site. 2366 A Map-Version Number is used as a sequence number per EID-prefix. So 2367 values that are greater, are considered to be more recent. A value 2368 of 0 for the Source Map-Version Number or the Destination Map-Version 2369 Number conveys no versioning information and an ITR does no 2370 comparison with previously received Map-Version Numbers. 2372 A Map-Version Number can be included in Map-Register messages as 2373 well. This is a good way for the Map-Server can assure that all ETRs 2374 for a site registering to it will be Map-Version number synchronized. 2376 See [VERSIONING] for a more detailed analysis and description of 2377 Database Map Versioning. 2379 7. Router Performance Considerations 2381 LISP is designed to be very hardware-based forwarding friendly. A 2382 few implementation techniques can be used to incrementally implement 2383 LISP: 2385 o When a tunnel encapsulated packet is received by an ETR, the outer 2386 destination address may not be the address of the router. This 2387 makes it challenging for the control plane to get packets from the 2388 hardware. This may be mitigated by creating special FIB entries 2389 for the EID-prefixes of EIDs served by the ETR (those for which 2390 the router provides an RLOC translation). These FIB entries are 2391 marked with a flag indicating that control plane processing should 2392 be performed. The forwarding logic of testing for particular IP 2393 protocol number value is not necessary. There are a few proven 2394 cases where no changes to existing deployed hardware were needed 2395 to support the LISP data-plane. 2397 o On an ITR, prepending a new IP header consists of adding more 2398 octets to a MAC rewrite string and prepending the string as part 2399 of the outgoing encapsulation procedure. Routers that support GRE 2400 tunneling [RFC2784] or 6to4 tunneling [RFC3056] may already 2401 support this action. 2403 o A packet's source address or interface the packet was received on 2404 can be used to select a VRF (Virtual Routing/Forwarding). The 2405 VRF's routing table can be used to find EID-to-RLOC mappings. 2407 For performance issues related to map-cache management, see section 2408 Section 12. 2410 8. Deployment Scenarios 2412 This section will explore how and where ITRs and ETRs can be deployed 2413 and will discuss the pros and cons of each deployment scenario. For 2414 a more detailed deployment recommendation, refer to [LISP-DEPLOY]. 2416 There are two basic deployment trade-offs to consider: centralized 2417 versus distributed caches and flat, recursive, or re-encapsulating 2418 tunneling. When deciding on centralized versus distributed caching, 2419 the following issues should be considered: 2421 o Are the tunnel routers spread out so that the caches are spread 2422 across all the memories of each router? A centralized cache is 2423 when an ITR keeps a cache for all the EIDs it is encapsulating to. 2424 The packet takes a direct path to the destination locator. A 2425 distributed cache is when an ITR needs help from other re- 2426 encapsulating routers because it does not store all the cache 2427 entries for the EIDs is it encapsulating to. So the packet takes 2428 a path through re-encapsulating routers that have a different set 2429 of cache entries. 2431 o Should management "touch points" be minimized by choosing few 2432 tunnel routers, just enough for redundancy? 2434 o In general, using more ITRs doesn't increase management load, 2435 since caches are built and stored dynamically. On the other hand, 2436 more ETRs does require more management since EID-prefix-to-RLOC 2437 mappings need to be explicitly configured. 2439 When deciding on flat, recursive, or re-encapsulation tunneling, the 2440 following issues should be considered: 2442 o Flat tunneling implements a single tunnel between source site and 2443 destination site. This generally offers better paths between 2444 sources and destinations with a single tunnel path. 2446 o Recursive tunneling is when tunneled traffic is again further 2447 encapsulated in another tunnel, either to implement VPNs or to 2448 perform Traffic Engineering. When doing VPN-based tunneling, the 2449 site has some control since the site is prepending a new tunnel 2450 header. In the case of TE-based tunneling, the site may have 2451 control if it is prepending a new tunnel header, but if the site's 2452 ISP is doing the TE, then the site has no control. Recursive 2453 tunneling generally will result in suboptimal paths but at the 2454 benefit of steering traffic to resource available parts of the 2455 network. 2457 o The technique of re-encapsulation ensures that packets only 2458 require one tunnel header. So if a packet needs to be rerouted, 2459 it is first decapsulated by the ETR and then re-encapsulated with 2460 a new tunnel header using a new RLOC. 2462 The next sub-sections will survey where tunnel routers can reside in 2463 the network. 2465 8.1. First-hop/Last-hop Tunnel Routers 2467 By locating tunnel routers close to hosts, the EID-prefix set is at 2468 the granularity of an IP subnet. So at the expense of more EID- 2469 prefix-to-RLOC sets for the site, the caches in each tunnel router 2470 can remain relatively small. But caches always depend on the number 2471 of non-aggregated EID destination flows active through these tunnel 2472 routers. 2474 With more tunnel routers doing encapsulation, the increase in control 2475 traffic grows as well: since the EID-granularity is greater, more 2476 Map-Requests and Map-Replies are traveling between more routers. 2478 The advantage of placing the caches and databases at these stub 2479 routers is that the products deployed in this part of the network 2480 have better price-memory ratios then their core router counterparts. 2481 Memory is typically less expensive in these devices and fewer routes 2482 are stored (only IGP routes). These devices tend to have excess 2483 capacity, both for forwarding and routing state. 2485 LISP functionality can also be deployed in edge switches. These 2486 devices generally have layer-2 ports facing hosts and layer-3 ports 2487 facing the Internet. Spare capacity is also often available in these 2488 devices as well. 2490 8.2. Border/Edge Tunnel Routers 2492 Using customer-edge (CE) routers for tunnel endpoints allows the EID 2493 space associated with a site to be reachable via a small set of RLOCs 2494 assigned to the CE routers for that site. This is the default 2495 behavior envisioned in the rest of this specification. 2497 This offers the opposite benefit of the first-hop/last-hop tunnel 2498 router scenario: the number of mapping entries and network management 2499 touch points are reduced, allowing better scaling. 2501 One disadvantage is that less of the network's resources are used to 2502 reach host endpoints thereby centralizing the point-of-failure domain 2503 and creating network choke points at the CE router. 2505 Note that more than one CE router at a site can be configured with 2506 the same IP address. In this case an RLOC is an anycast address. 2507 This allows resilience between the CE routers. That is, if a CE 2508 router fails, traffic is automatically routed to the other routers 2509 using the same anycast address. However, this comes with the 2510 disadvantage where the site cannot control the entrance point when 2511 the anycast route is advertised out from all border routers. Another 2512 disadvantage of using anycast locators is the limited advertisement 2513 scope of /32 (or /128 for IPv6) routes. 2515 8.3. ISP Provider-Edge (PE) Tunnel Routers 2517 Use of ISP PE routers as tunnel endpoint routers is not the typical 2518 deployment scenario envisioned in the specification. This section 2519 attempts to capture some of reasoning behind this preference of 2520 implementing LISP on CE routers. 2522 Use of ISP PE routers as tunnel endpoint routers gives an ISP, rather 2523 than a site, control over the location of the egress tunnel 2524 endpoints. That is, the ISP can decide if the tunnel endpoints are 2525 in the destination site (in either CE routers or last-hop routers 2526 within a site) or at other PE edges. The advantage of this case is 2527 that two tunnel headers can be avoided. By having the PE be the 2528 first router on the path to encapsulate, it can choose a TE path 2529 first, and the ETR can decapsulate and re-encapsulate for a tunnel to 2530 the destination end site. 2532 An obvious disadvantage is that the end site has no control over 2533 where its packets flow or the RLOCs used. Other disadvantages 2534 include the difficulty in synchronizing path liveness updates between 2535 CE and PE routers. 2537 As mentioned in earlier sections a combination of these scenarios is 2538 possible at the expense of extra packet header overhead, if both site 2539 and provider want control, then recursive or re-encapsulating tunnels 2540 are used. 2542 8.4. LISP Functionality with Conventional NATs 2544 LISP routers can be deployed behind Network Address Translator (NAT) 2545 devices to provide the same set of packet services hosts have today 2546 when they are addressed out of private address space. 2548 It is important to note that a locator address in any LISP control 2549 message MUST be a globally routable address and therefore SHOULD NOT 2550 contain [RFC1918] addresses. If a LISP router is configured with 2551 private addresses, they MUST be used only in the outer IP header so 2552 the NAT device can translate properly. Otherwise, EID addresses MUST 2553 be translated before encapsulation is performed. Both NAT 2554 translation and LISP encapsulation functions could be co-located in 2555 the same device. 2557 More details on LISP address translation can be found in [INTERWORK]. 2559 8.5. Packets Egressing a LISP Site 2561 When a LISP site is using two ITRs for redundancy, the failure of one 2562 ITR will likely shift outbound traffic to the second. This second 2563 ITR's cache may not not be populated with the same EID-to-RLOC 2564 mapping entries as the first. If this second ITR does not have these 2565 mappings, traffic will be dropped while the mappings are retrieved 2566 from the mapping system. The retrieval of these messages may 2567 increase the load of requests being sent into the mapping system. 2568 Deployment and experimentation will determine whether this issue 2569 requires more attention. 2571 9. Traceroute Considerations 2573 When a source host in a LISP site initiates a traceroute to a 2574 destination host in another LISP site, it is highly desirable for it 2575 to see the entire path. Since packets are encapsulated from ITR to 2576 ETR, the hop across the tunnel could be viewed as a single hop. 2577 However, LISP traceroute will provide the entire path so the user can 2578 see 3 distinct segments of the path from a source LISP host to a 2579 destination LISP host: 2581 Segment 1 (in source LISP site based on EIDs): 2583 source-host ---> first-hop ... next-hop ---> ITR 2585 Segment 2 (in the core network based on RLOCs): 2587 ITR ---> next-hop ... next-hop ---> ETR 2589 Segment 3 (in the destination LISP site based on EIDs): 2591 ETR ---> next-hop ... last-hop ---> destination-host 2593 For segment 1 of the path, ICMP Time Exceeded messages are returned 2594 in the normal manner as they are today. The ITR performs a TTL 2595 decrement and test for 0 before encapsulating. So the ITR hop is 2596 seen by the traceroute source has an EID address (the address of 2597 site-facing interface). 2599 For segment 2 of the path, ICMP Time Exceeded messages are returned 2600 to the ITR because the TTL decrement to 0 is done on the outer 2601 header, so the destination of the ICMP messages are to the ITR RLOC 2602 address, the source RLOC address of the encapsulated traceroute 2603 packet. The ITR looks inside of the ICMP payload to inspect the 2604 traceroute source so it can return the ICMP message to the address of 2605 the traceroute client as well as retaining the core router IP address 2606 in the ICMP message. This is so the traceroute client can display 2607 the core router address (the RLOC address) in the traceroute output. 2608 The ETR returns its RLOC address and responds to the TTL decrement to 2609 0 like the previous core routers did. 2611 For segment 3, the next-hop router downstream from the ETR will be 2612 decrementing the TTL for the packet that was encapsulated, sent into 2613 the core, decapsulated by the ETR, and forwarded because it isn't the 2614 final destination. If the TTL is decremented to 0, any router on the 2615 path to the destination of the traceroute, including the next-hop 2616 router or destination, will send an ICMP Time Exceeded message to the 2617 source EID of the traceroute client. The ICMP message will be 2618 encapsulated by the local ITR and sent back to the ETR in the 2619 originated traceroute source site, where the packet will be delivered 2620 to the host. 2622 9.1. IPv6 Traceroute 2624 IPv6 traceroute follows the procedure described above since the 2625 entire traceroute data packet is included in ICMP Time Exceeded 2626 message payload. Therefore, only the ITR needs to pay special 2627 attention for forwarding ICMP messages back to the traceroute source. 2629 9.2. IPv4 Traceroute 2631 For IPv4 traceroute, we cannot follow the above procedure since IPv4 2632 ICMP Time Exceeded messages only include the invoking IP header and 8 2633 octets that follow the IP header. Therefore, when a core router 2634 sends an IPv4 Time Exceeded message to an ITR, all the ITR has in the 2635 ICMP payload is the encapsulated header it prepended followed by a 2636 UDP header. The original invoking IP header, and therefore the 2637 identity of the traceroute source is lost. 2639 The solution we propose to solve this problem is to cache traceroute 2640 IPv4 headers in the ITR and to match them up with corresponding IPv4 2641 Time Exceeded messages received from core routers and the ETR. The 2642 ITR will use a circular buffer for caching the IPv4 and UDP headers 2643 of traceroute packets. It will select a 16-bit number as a key to 2644 find them later when the IPv4 Time Exceeded messages are received. 2645 When an ITR encapsulates an IPv4 traceroute packet, it will use the 2646 16-bit number as the UDP source port in the encapsulating header. 2647 When the ICMP Time Exceeded message is returned to the ITR, the UDP 2648 header of the encapsulating header is present in the ICMP payload 2649 thereby allowing the ITR to find the cached headers for the 2650 traceroute source. The ITR puts the cached headers in the payload 2651 and sends the ICMP Time Exceeded message to the traceroute source 2652 retaining the source address of the original ICMP Time Exceeded 2653 message (a core router or the ETR of the site of the traceroute 2654 destination). 2656 The signature of a traceroute packet comes in two forms. The first 2657 form is encoded as a UDP message where the destination port is 2658 inspected for a range of values. The second form is encoded as an 2659 ICMP message where the IP identification field is inspected for a 2660 well-known value. 2662 9.3. Traceroute using Mixed Locators 2664 When either an IPv4 traceroute or IPv6 traceroute is originated and 2665 the ITR encapsulates it in the other address family header, you 2666 cannot get all 3 segments of the traceroute. Segment 2 of the 2667 traceroute can not be conveyed to the traceroute source since it is 2668 expecting addresses from intermediate hops in the same address format 2669 for the type of traceroute it originated. Therefore, in this case, 2670 segment 2 will make the tunnel look like one hop. All the ITR has to 2671 do to make this work is to not copy the inner TTL to the outer, 2672 encapsulating header's TTL when a traceroute packet is encapsulated 2673 using an RLOC from a different address family. This will cause no 2674 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2676 10. Mobility Considerations 2678 There are several kinds of mobility of which only some might be of 2679 concern to LISP. Essentially they are as follows. 2681 10.1. Site Mobility 2683 A site wishes to change its attachment points to the Internet, and 2684 its LISP Tunnel Routers will have new RLOCs when it changes upstream 2685 providers. Changes in EID-RLOC mappings for sites are expected to be 2686 handled by configuration, outside of the LISP protocol. 2688 10.2. Slow Endpoint Mobility 2690 An individual endpoint wishes to move, but is not concerned about 2691 maintaining session continuity. Renumbering is involved. LISP can 2692 help with the issues surrounding renumbering [RFC4192] [LISA96] by 2693 decoupling the address space used by a site from the address spaces 2694 used by its ISPs. [RFC4984] 2696 10.3. Fast Endpoint Mobility 2698 Fast endpoint mobility occurs when an endpoint moves relatively 2699 rapidly, changing its IP layer network attachment point. Maintenance 2700 of session continuity is a goal. This is where the Mobile IPv4 2701 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used, 2702 and primarily where interactions with LISP need to be explored. 2704 The problem is that as an endpoint moves, it may require changes to 2705 the mapping between its EID and a set of RLOCs for its new network 2706 location. When this is added to the overhead of mobile IP binding 2707 updates, some packets might be delayed or dropped. 2709 In IPv4 mobility, when an endpoint is away from home, packets to it 2710 are encapsulated and forwarded via a home agent which resides in the 2711 home area the endpoint's address belongs to. The home agent will 2712 encapsulate and forward packets either directly to the endpoint or to 2713 a foreign agent which resides where the endpoint has moved to. 2714 Packets from the endpoint may be sent directly to the correspondent 2715 node, may be sent via the foreign agent, or may be reverse-tunneled 2716 back to the home agent for delivery to the mobile node. As the 2717 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 2718 mappings are required for communication between the mobile node and 2719 the home agent, whether via foreign agent or not. As a mobile 2720 endpoint changes networks, up to three LISP mapping changes may be 2721 required: 2723 o The mobile node moves from an old location to a new visited 2724 network location and notifies its home agent that it has done so. 2725 The Mobile IPv4 control packets the mobile node sends pass through 2726 one of the new visited network's ITRs, which needs an EID-RLOC 2727 mapping for the home agent. 2729 o The home agent might not have the EID-RLOC mappings for the mobile 2730 node's "care-of" address or its foreign agent in the new visited 2731 network, in which case it will need to acquire them. 2733 o When packets are sent directly to the correspondent node, it may 2734 be that no traffic has been sent from the new visited network to 2735 the correspondent node's network, and the new visited network's 2736 ITR will need to obtain an EID-RLOC mapping for the correspondent 2737 node's site. 2739 In addition, if the IPv4 endpoint is sending packets from the new 2740 visited network using its original EID, then LISP will need to 2741 perform a route-returnability check on the new EID-RLOC mapping for 2742 that EID. 2744 In IPv6 mobility, packets can flow directly between the mobile node 2745 and the correspondent node in either direction. The mobile node uses 2746 its "care-of" address (EID). In this case, the route-returnability 2747 check would not be needed but one more LISP mapping lookup may be 2748 required instead: 2750 o As above, three mapping changes may be needed for the mobile node 2751 to communicate with its home agent and to send packets to the 2752 correspondent node. 2754 o In addition, another mapping will be needed in the correspondent 2755 node's ITR, in order for the correspondent node to send packets to 2756 the mobile node's "care-of" address (EID) at the new network 2757 location. 2759 When both endpoints are mobile the number of potential mapping 2760 lookups increases accordingly. 2762 As a mobile node moves there are not only mobility state changes in 2763 the mobile node, correspondent node, and home agent, but also state 2764 changes in the ITRs and ETRs for at least some EID-prefixes. 2766 The goal is to support rapid adaptation, with little delay or packet 2767 loss for the entire system. Also IP mobility can be modified to 2768 require fewer mapping changes. In order to increase overall system 2769 performance, there may be a need to reduce the optimization of one 2770 area in order to place fewer demands on another. 2772 In LISP, one possibility is to "glean" information. When a packet 2773 arrives, the ETR could examine the EID-RLOC mapping and use that 2774 mapping for all outgoing traffic to that EID. It can do this after 2775 performing a route-returnability check, to ensure that the new 2776 network location does have a internal route to that endpoint. 2777 However, this does not cover the case where an ITR (the node assigned 2778 the RLOC) at the mobile-node location has been compromised. 2780 Mobile IP packet exchange is designed for an environment in which all 2781 routing information is disseminated before packets can be forwarded. 2782 In order to allow the Internet to grow to support expected future 2783 use, we are moving to an environment where some information may have 2784 to be obtained after packets are in flight. Modifications to IP 2785 mobility should be considered in order to optimize the behavior of 2786 the overall system. Anything which decreases the number of new EID- 2787 RLOC mappings needed when a node moves, or maintains the validity of 2788 an EID-RLOC mapping for a longer time, is useful. 2790 10.4. Fast Network Mobility 2792 In addition to endpoints, a network can be mobile, possibly changing 2793 xTRs. A "network" can be as small as a single router and as large as 2794 a whole site. This is different from site mobility in that it is 2795 fast and possibly short-lived, but different from endpoint mobility 2796 in that a whole prefix is changing RLOCs. However, the mechanisms 2797 are the same and there is no new overhead in LISP. A map request for 2798 any endpoint will return a binding for the entire mobile prefix. 2800 If mobile networks become a more common occurrence, it may be useful 2801 to revisit the design of the mapping service and allow for dynamic 2802 updates of the database. 2804 The issue of interactions between mobility and LISP needs to be 2805 explored further. Specific improvements to the entire system will 2806 depend on the details of mapping mechanisms. Mapping mechanisms 2807 should be evaluated on how well they support session continuity for 2808 mobile nodes. 2810 10.5. LISP Mobile Node Mobility 2812 A mobile device can use the LISP infrastructure to achieve mobility 2813 by implementing the LISP encapsulation and decapsulation functions 2814 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 2815 node" can use topologically-independent EID IP addresses that are not 2816 advertised into and do not impose a cost on the global routing 2817 system. These EIDs are maintained at the edges of the mapping system 2818 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 2819 only the correspondents of the LISP mobile node. 2821 Refer to the LISP Mobility Architecture specification [LISP-MN] for 2822 more details. 2824 11. Multicast Considerations 2826 A multicast group address, as defined in the original Internet 2827 architecture is an identifier of a grouping of topologically 2828 independent receiver host locations. The address encoding itself 2829 does not determine the location of the receiver(s). The multicast 2830 routing protocol, and the network-based state the protocol creates, 2831 determines where the receivers are located. 2833 In the context of LISP, a multicast group address is both an EID and 2834 a Routing Locator. Therefore, no specific semantic or action needs 2835 to be taken for a destination address, as it would appear in an IP 2836 header. Therefore, a group address that appears in an inner IP 2837 header built by a source host will be used as the destination EID. 2838 The outer IP header (the destination Routing Locator address), 2839 prepended by a LISP router, will use the same group address as the 2840 destination Routing Locator. 2842 Having said that, only the source EID and source Routing Locator 2843 needs to be dealt with. Therefore, an ITR merely needs to put its 2844 own IP address in the source Routing Locator field when prepending 2845 the outer IP header. This source Routing Locator address, like any 2846 other Routing Locator address MUST be globally routable. 2848 Therefore, an EID-to-RLOC mapping does not need to be performed by an 2849 ITR when a received data packet is a multicast data packet or when 2850 processing a source-specific Join (either by IGMPv3 or PIM). But the 2851 source Routing Locator is decided by the multicast routing protocol 2852 in a receiver site. That is, an EID to Routing Locator translation 2853 is done at control-time. 2855 Another approach is to have the ITR not encapsulate a multicast 2856 packet and allow the host built packet to flow into the core even if 2857 the source address is allocated out of the EID namespace. If the 2858 RPF-Vector TLV [RFC5496] is used by PIM in the core, then core 2859 routers can RPF to the ITR (the Locator address which is injected 2860 into core routing) rather than the host source address (the EID 2861 address which is not injected into core routing). 2863 To avoid any EID-based multicast state in the network core, the first 2864 approach is chosen for LISP-Multicast. Details for LISP-Multicast 2865 and Interworking with non-LISP sites is described in specification 2866 [MLISP]. 2868 12. Security Considerations 2870 It is believed that most of the security mechanisms will be part of 2871 the mapping database service when using control plane procedures for 2872 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 2873 as described in this specification, protection is provided against 2874 ETR spoofing by using Return-Routability (see Section 3) mechanisms 2875 evidenced by the use of a 24-bit Nonce field in the LISP 2876 encapsulation header and a 64-bit Nonce field in the LISP control 2877 message. 2879 The nonce, coupled with the ITR accepting only solicited Map-Replies 2880 provides a basic level of security, in many ways similar to the 2881 security experienced in the current Internet routing system. It is 2882 hard for off-path attackers to launch attacks against these LISP 2883 mechanisms, as they do not have the nonce values. Sending a large 2884 number of packets to accidentally find the right nonce value is 2885 possible, but would already by itself be a denial-of-service attack. 2886 On-path attackers can perform far more serious attacks, but on-path 2887 attackers can launch serious attacks in the current Internet as well, 2888 including eavesdropping, blocking or redirecting traffic. See more 2889 discussion on this topic in Section 6.1.5.1. 2891 LISP does not rely on a PKI or a more heavy weight authentication 2892 system. These systems challenge the scalability of LISP which was a 2893 primary design goal. 2895 DoS attack prevention will depend on implementations rate-limiting 2896 Map-Requests and Map-Replies to the control plane as well as rate- 2897 limiting the number of data-triggered Map-Replies. 2899 An incorrectly implemented or malicious ITR might choose to ignore 2900 the priority and weights provided by the ETR in its Map-Reply. This 2901 traffic steering would be limited to the traffic that is sent by this 2902 ITR's site, and no more severe than if the site initiated a bandwidth 2903 DoS attack on (one of) the ETR's ingress links. The ITR's site would 2904 typically gain no benefit from not respecting the weights, and would 2905 likely to receive better service by abiding by them. 2907 To deal with map-cache exhaustion attempts in an ITR/PITR, the 2908 implementation should consider putting a maximum cap on the number of 2909 entries stored with a reserve list for special or frequently accessed 2910 sites. This should be a configuration policy control set by the 2911 network administrator who manages ITRs and PITRs. When overlapping 2912 EID-prefixes occur across multiple map-cache entries, the integrity 2913 of the set must be wholly maintained. So if a more-specific entry 2914 cannot be added due to reaching the maximum cap, then none of the 2915 less specifics should be stored in the map-cache. 2917 Given that the ITR/PITR maintains a cache of EID-to-RLOC mappings, 2918 cache sizing and maintenance is an issue to be kept in mind during 2919 implementation. It is a good idea to have instrumentation in place 2920 to detect thrashing of the cache. Implementation experimentation 2921 will be used to determine which cache management strategies work 2922 best. In general, it is difficult to defend against cache trashing 2923 attacks. It should be noted that an undersized cache in an ITR/PITR 2924 not only causes adverse affect on the site or region they support, 2925 but may also cause increased Map-Request load on the mapping system. 2927 "Piggybacked" mapping data discussed in Section 6.1.3 specifies how 2928 to handle such mappings and includes the possibility for an ETR to 2929 temporarily accept such a mapping before verification when running in 2930 "trusted" environments. In such cases, there is a potential threat 2931 that a fake mapping could be inserted (even if only for a short 2932 period) into a map-cache. As noted in Section 6.1.3, an ETR MUST be 2933 specifically configured to run in such a mode and might usefully only 2934 consider some specific ITRs as also running in that same trusted 2935 environment. 2937 There is a security risk implicit in the fact that ETRs generate the 2938 EID prefix to which they are responding. An ETR can claim a shorter 2939 prefix than it is actually responsible for. Various mechanisms to 2940 ameliorate or resolve this issue will be examined in the future, 2941 [LISP-SEC]. 2943 Spoofing of inner header addresses of LISP encapsulated packets is 2944 possible like with any tunneling mechanism. ITRs MUST verify the 2945 source address of a packet to be an EID that belongs to the site's 2946 EID-prefix range prior to encapsulation. An ETR must only 2947 decapsulate and forward datagrams with an inner header destination 2948 that matches one of its EID-prefix ranges. If, upon receipt and 2949 decapsulation, the destination EID of a datagram does not match one 2950 of the ETR's configured EID-prefixes, the ETR MUST drop the datagram. 2951 If a LISP encapsulated packet arrives at an ETR, it SHOULD compare 2952 the inner header source EID address and the outer header source RLOC 2953 address with the mapping that exists in the mapping database. Then 2954 when spoofing attacks occur, the outer header source RLOC address can 2955 be used to trace back the attack to the source site, using existing 2956 operational tools. 2958 This experimental specification does not address automated key 2959 management (AKM). BCP 107 provides guidance in this area. In 2960 addition, at the time of this writing, substantial work is being 2961 undertaken to improve security of the routing system [KARP], [RPKI], 2962 [BGP-SEC], [LISP-SEC]. Future work on LISP should address BCP-107 as 2963 well as other open security considerations, which may require changes 2964 to this specification. 2966 13. Network Management Considerations 2968 Considerations for Network Management tools exist so the LISP 2969 protocol suite can be operationally managed. The mechanisms can be 2970 found in [LISP-MIB] and [LISP-LIG]. 2972 14. IANA Considerations 2974 This section provides guidance to the Internet Assigned Numbers 2975 Authority (IANA) regarding registration of values related to the LISP 2976 specification, in accordance with BCP 26 and RFC 5226 [RFC5226]. 2978 There are four name spaces in LISP that require registration: 2980 o LISP IANA registry allocations should not be made for purposes 2981 unrelated to LISP routing or transport protocols. 2983 o The following policies are used here with the meanings defined in 2984 BCP 26: "Specification Required", "IETF Review", "Experimental 2985 Use", "First Come First Served". 2987 14.1. LISP ACT and Flag Fields 2989 New ACT values (Section 6.1.4) can be allocated through IETF review 2990 or IESG approval. Four values have already been allocated by this 2991 specification (Section 6.1.4). 2993 In addition, the LISP protocol has a number of flag and reserved 2994 fields, such as the LISP header flags field (Section 5.3). New bits 2995 for flags can be taken into use from these fields through IETF review 2996 or IESG approval, but these need not be managed by IANA. 2998 14.2. LISP Address Type Codes 3000 LISP Address [LCAF] type codes have a range from 0 to 255. New type 3001 codes MUST be allocated consecutively starting at 0. Type Codes 0 - 3002 127 are to be assigned by IETF review or IESG approval. 3004 Type Codes 128 - 255 are available on a First Come First Served 3005 policy. 3007 This registry, initially empty, is constructed for future-use 3008 experimental work of LCAF values. See [LCAF] for details for other 3009 possible unapproved address encodings. The unapproved LCAF encodings 3010 are an area for further study and experimentation. 3012 14.3. LISP UDP Port Numbers 3014 The IANA registry has allocated UDP port numbers 4341 and 4342 for 3015 LISP data-plane and control-plane operation, respectively. 3017 14.4. LISP Key ID Numbers 3019 The following Key ID values are defined by this specification as used 3020 in any packet type that references a Key ID field: 3022 Name Number Defined in 3023 ----------------------------------------------- 3024 None 0 n/a 3025 HMAC-SHA-1-96 1 [RFC2404] 3026 HMAC-SHA-256-128 2 [RFC6234] 3028 15. Known Open Issues and Areas of Future Work 3030 As an experimental specification, this work is, by definition, 3031 incomplete. Specific areas where additional experience and work are 3032 needed include: 3034 o At present, only [ALT] is defined for implementing a database of 3035 EID-to-RLOC mapping information. Additional research on other 3036 mapping database systems is strongly encouraged. 3038 o Failure and recovery of LISP site partitioning (see Section 6.4), 3039 in the presence of redundant configuration (see Section 8.5) needs 3040 further research and experimentation. 3042 o The characteristics of map-cache management under exceptional 3043 conditions, such as denial-of-service attacks are not fully 3044 understood. Further experience is needed to determine whether 3045 current caching methods are practical or in need of further 3046 development. In particular, the performance, scaling and security 3047 characteristics of the map-cache will be discovered as part of 3048 this experiment. Performance metrics to be observed are packet 3049 reordering associated with the LISP data probe and loss of the 3050 first packet in a flow associated with map-caching. The impact of 3051 these upon TCP will be observed. See Section 12 for additional 3052 thoughts and considerations. 3054 o Preliminary work has been done to ensure that sites employing LISP 3055 can interconnect with the rest of the Internet. This work is 3056 documented in [INTERWORK], but further experimentation and 3057 experience is needed. 3059 o At present, no mechanism for automated key management for message 3060 authentication is defined. Addressing automated key management is 3061 necessary before this specification could be developed into a 3062 standards track RFC. See Section 12 for further details regarding 3063 security considerations. 3065 o In order to maintain security and stability, Internet Protocols 3066 typically isolate the control and data planes. Therefore, user 3067 activity cannot cause control plane state to be created or 3068 destroyed. LISP does not maintain this separation. The degree to 3069 which the loss of separation impacts security and stability is a 3070 topic for experimental observation. 3072 o LISP allows for different mapping database systems to be used. 3073 While only one [ALT] is currently well-defined, each mapping 3074 database will likely have some impact on the security of the EID- 3075 to-RLOC mappings. How each mapping database system's security 3076 properties impact on LISP overall is for further study. 3078 o An examination of the implications of LISP on Internet traffic, 3079 applications, routers, and security is needed. This will help to 3080 understand the consequences for network stability, routing 3081 protocol function, routing scalability, migration and backward 3082 compatibility, and implementation scalability (as influenced by 3083 additional protocol components, additional state, and additional 3084 processing for encapsulation, decapsulation, liveness). 3086 o Experiments need to verify that LISP produces no significant 3087 change in the behavior of protocols run between end-systems over a 3088 LISP infrastructure versus being run directly between those same 3089 end-systems. 3091 o Experiments need to verify that the issues raised in the Critique 3092 section of [RFC6115] are either insignificant or have been 3093 addressed by updates to the LISP protocol. 3095 Other LISP documents may also include open issues and areas for 3096 future work. 3098 16. References 3100 16.1. Normative References 3102 [ALT] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "LISP 3103 Alternative Topology (LISP-ALT)", 3104 draft-ietf-lisp-alt-09.txt (work in progress). 3106 [LISP-MS] Farinacci, D. and V. Fuller, "LISP Map Server", 3107 draft-ietf-lisp-ms-12.txt (work in progress). 3109 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 3110 August 1980. 3112 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3113 September 1981. 3115 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 3116 E. Lear, "Address Allocation for Private Internets", 3117 BCP 5, RFC 1918, February 1996. 3119 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3120 Requirement Levels", BCP 14, RFC 2119, March 1997. 3122 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 3123 ESP and AH", RFC 2404, November 1998. 3125 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3126 (IPv6) Specification", RFC 2460, December 1998. 3128 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3129 of Explicit Congestion Notification (ECN) to IP", 3130 RFC 3168, September 2001. 3132 [RFC3232] Reynolds, J., "Assigned Numbers: RFC 1700 is Replaced by 3133 an On-line Database", RFC 3232, January 2002. 3135 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 3136 Requirements for Security", BCP 106, RFC 4086, June 2005. 3138 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 3139 (CIDR): The Internet Address Assignment and Aggregation 3140 Plan", BCP 122, RFC 4632, August 2006. 3142 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 3143 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 3144 May 2008. 3146 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 3147 Forwarding (RPF) Vector TLV", RFC 5496, March 2009. 3149 [RFC5944] Perkins, C., "IP Mobility Support for IPv4, Revised", 3150 RFC 5944, November 2010. 3152 [RFC6115] Li, T., "Recommendation for a Routing Architecture", 3153 RFC 6115, February 2011. 3155 [RFC6234] Eastlake, D. and T. Hansen, "US Secure Hash Algorithms 3156 (SHA and SHA-based HMAC and HKDF)", RFC 6234, May 2011. 3158 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 3159 in IPv6", RFC 6275, July 2011. 3161 [UDP-TUNNELS] 3162 Eubanks, M. and P. Chimento, "UDP Checksums for Tunneled 3163 Packets", draft-eubanks-chimento-6man-01.txt (work in 3164 progress), October 2010. 3166 [UDP-ZERO] 3167 Fairhurst, G. and M. Westerland, "IPv6 UDP Checksum 3168 Considerations", draft-ietf-6man-udpzero-04.txt (work in 3169 progress), October 2011. 3171 [VERSIONING] 3172 Iannone, L., Saucez, D., and O. Bonaventure, "LISP Mapping 3173 Versioning", draft-ietf-lisp-map-versioning-05.txt (work 3174 in progress). 3176 16.2. Informative References 3178 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3179 NUMBERS 3180 http://www.iana.org/assignments/address-family-numbers. 3182 [AFI-REGISTRY] 3183 IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3184 NUMBER registry http://www.iana.org/assignments/ 3185 address-family-numbers/ 3186 address-family-numbers.xml#address-family-numbers-1. 3188 [BGP-SEC] Lepinski, M., "An Overview of BGPSEC", 3189 draft-lepinski-bgpsec-overview-00.txt (work in progress), 3190 March 2011. 3192 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 3193 Enhancement to the Internet Architecture", Internet- 3194 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt. 3196 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 3197 Content distribution Overlay Network Service for LISP", 3198 draft-meyer-lisp-cons-04.txt (work in progress). 3200 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 3201 Mappings Multicast Across Cooperating Systems for LISP", 3202 draft-curran-lisp-emacs-00.txt (work in progress). 3204 [INTERWORK] 3205 Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 3206 "Interworking LISP with IPv4 and IPv6", 3207 draft-ietf-lisp-interworking-02.txt (work in progress). 3209 [KARP] Lebovitz, G. and M. Bhatia, "Keying and Authentication for 3210 Routing Protocols (KARP)Design Guidelines", 3211 draft-ietf-karp-design-guide-06.txt (work in progress), 3212 October 2011. 3214 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 3215 Address Format", draft-farinacci-lisp-lcaf-06.txt (work in 3216 progress). 3218 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 3219 "Renumbering: Threat or Menace?", Usenix . 3221 [LISP-DEPLOY] 3222 Jakab, L., Coras, F., Domingo-Pascual, J., and D. Lewis, 3223 "LISP Network Element Deployment Considerations", 3224 draft-ietf-lisp-deployment-02.txt (work in progress). 3226 [LISP-LIG] 3227 Farinacci, D. and D. Meyer, "LISP Internet Groper (LIG)", 3228 draft-ietf-lisp-lig-06.txt (work in progress). 3230 [LISP-MAIN] 3231 Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 3232 "Locator/ID Separation Protocol (LISP)", 3233 draft-farinacci-lisp-12.txt (work in progress). 3235 [LISP-MIB] 3236 Schudel, G., Jain, A., and V. Moreno, "LISP MIB", 3237 draft-ietf-lisp-mib-02.txt (work in progress). 3239 [LISP-MN] Farinacci, D., Fuller, V., Lewis, D., and D. Meyer, "LISP 3240 Mobility Architecture", draft-meyer-lisp-mn-06.txt (work 3241 in progress). 3243 [LISP-SEC] 3244 Maino, F., Ermagon, V., Cabellos, A., Sausez, D., and O. 3245 Bonaventure, "LISP-Security (LISP-SEC)", 3246 draft-ietf-lisp-sec-00.txt (work in progress). 3248 [LOC-ID-ARCH] 3249 Meyer, D. and D. Lewis, "Architectural Implications of 3250 Locator/ID Separation", 3251 draft-meyer-loc-id-implications-02.txt (work in progress). 3253 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 3254 "LISP for Multicast Environments", 3255 draft-ietf-lisp-multicast-10.txt (work in progress). 3257 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 3258 draft-lear-lisp-nerd-08.txt (work in progress). 3260 [OPENLISP] 3261 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 3262 Report", draft-iannone-openlisp-implementation-01.txt 3263 (work in progress). 3265 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 3266 draft-narten-radir-problem-statement-05.txt (work in 3267 progress). 3269 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 3270 STD 13, RFC 1034, November 1987. 3272 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3273 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3274 March 2000. 3276 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 3277 via IPv4 Clouds", RFC 3056, February 2001. 3279 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 3280 A., Peterson, J., Sparks, R., Handley, M., and E. 3281 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 3282 June 2002. 3284 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 3285 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 3286 September 2005. 3288 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 3289 Optimization for Mobile IPv6", RFC 4866, May 2007. 3291 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 3292 Workshop on Routing and Addressing", RFC 4984, 3293 September 2007. 3295 [RPKI] Lepinski, M., "An Infrastructure to Support Secure 3296 Internet Routing", draft-ietf-sidr-arch-13.txt (work in 3297 progress), February 2011. 3299 Appendix A. Acknowledgments 3301 An initial thank you goes to Dave Oran for planting the seeds for the 3302 initial ideas for LISP. His consultation continues to provide value 3303 to the LISP authors. 3305 A special and appreciative thank you goes to Noel Chiappa for 3306 providing architectural impetus over the past decades on separation 3307 of location and identity, as well as detailed review of the LISP 3308 architecture and documents, coupled with enthusiasm for making LISP a 3309 practical and incremental transition for the Internet. 3311 The authors would like to gratefully acknowledge many people who have 3312 contributed discussion and ideas to the making of this proposal. 3313 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 3314 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 3315 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 3316 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 3317 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 3318 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 3319 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 3320 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 3321 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 3322 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 3323 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 3324 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 3325 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 3326 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 3327 Ermagan, Albert Cabellos, Fabio Maino, Victor Moreno, Chris White, 3328 Clarence Filsfils, and Alia Atlas. 3330 This work originated in the Routing Research Group (RRG) of the IRTF. 3331 The individual submission [LISP-MAIN] was converted into this IETF 3332 LISP working group draft. 3334 The LISP working group would like to give a special thanks to Jari 3335 Arkko, the Internet Area AD at the time the set of LISP documents 3336 were being prepared for IESG last call, for his meticulous review and 3337 detail commentary on the 7 working group last call drafts progressing 3338 toward experimental RFCs. 3340 Appendix B. Document Change Log 3342 B.1. Changes to draft-ietf-lisp-22.txt 3344 o Posted February 2012 to reflect final DISCUSS comments from Adrian 3345 Farrel. 3347 B.2. Changes to draft-ietf-lisp-21.txt 3349 o Posted February 2012 to reflect DISCUSS comments from Adrian 3350 Farrel, Stewart Bryant, and Wesley Eddy. 3352 B.3. Changes to draft-ietf-lisp-20.txt 3354 o Posted January 2012 for resolution to Adrian Farrel's security 3355 comments as well as additions to the end of section 2, Elwyn 3356 Davies Gen-Art comments, and Ralph Droms' IANA and EID definition 3357 comments. 3359 B.4. Changes to draft-ietf-lisp-19.txt 3361 o Posted January 2012 for Stephen Farrell's comment resolution. 3363 B.5. Changes to draft-ietf-lisp-18.txt 3365 o Posted December 2011 after reflecting comments from IANA. 3367 o Create reference to sections 5.4.1 and 5.4.2 about DF bit setting 3368 from section 5.3. 3370 o Inserted two references for Route-Returnability and on-path 3371 attacks in Security Considerations section. 3373 B.6. Changes to draft-ietf-lisp-17.txt 3375 o Posted December 2011 after IETF last call comments. 3377 o Make Map-Notify port assignment be 4342 in both source and 3378 destination ports. This change was agreed on and put in [LISP-MS] 3379 but was not updated in this spec. 3381 B.7. Changes to draft-ietf-lisp-16.txt 3383 o Posted October 2011 after AD review by Jari. 3385 B.8. Changes to draft-ietf-lisp-15.txt 3387 o Posted July 2011. Fixing IDnits errors. 3389 o Change description on how to select a source address for RLOC- 3390 probe Map-Replies to refer to the "EID-to-RLOC Map-Reply Message" 3391 section. 3393 B.9. Changes to draft-ietf-lisp-14.txt 3395 o Post working group last call and pre-IESG last call review. 3397 o Indicate that an ICMP Unreachable message should be sent when a 3398 packet matches a drop-based negative map-cache entry. 3400 o Indicate how a map-cache set of overlapping EID-prefixes must 3401 maintain integrity when the map-cache maximum cap is reached. 3403 o Add Joel's description for the definition of an EID, that the bit 3404 string value can be an RLOC for another device in abstract but the 3405 architecture allows it to be an EID of one device and the same 3406 value as an RLOC for another device. 3408 o In the "Tunnel Encapsulation Details" section, indicate that 4 3409 combinations of encapsulation are supported. 3411 o Add what ETR should do for a Data-Probe when received for a 3412 destination EID outside of its EID-prefix range. This was added 3413 in the Data Probe definition section. 3415 o Added text indicating that more-specific EID-prefixes must not be 3416 removed when less-specific entries stay in the map-cache. This is 3417 to preserve the integrity of the EID-prefix set. 3419 o Add clarifying text in the Security Considerations section about 3420 how an ETR must not decapsulate and forward a packet that is not 3421 for its configured EID-prefix range. 3423 B.10. Changes to draft-ietf-lisp-13.txt 3425 o Posted June 2011 to complete working group last call. 3427 o Tracker item 87. Put Yakov suggested wording in the EID-prefix 3428 definition section to reference [INTERWORK] and [LISP-DEPLOY] 3429 about discussion on transition and access mechanisms. 3431 o Change "ITRs" to "ETRs" in the Locator Status Bit definition 3432 section and data packet description section per Damien's comment. 3434 o Remove the normative reference to [LISP-SEC] when describing the 3435 S-bit in the ECM and Map-Reply headers. 3437 o Tracker item 54. Added text from John Scudder in the "Packets 3438 Egressing a LISP Site" section. 3440 o Add sentence to the "Reencapsulating Tunnel" definition about how 3441 reencapsulation loops can occur when not coordinating among 3442 multiple mapping database systems. 3444 o Remove "In theory" from a sentence in the Security Considerations 3445 section. 3447 o Remove Security Area Statement title and reword section with 3448 Eliot's provided text. The text was agreed upon by LISP-WG chairs 3449 and Security ADs. 3451 o Remove word "potential" from the over-claiming paragraph of the 3452 Security Considerations section per Stephen's request. 3454 o Wordsmithing and other editorial comments from Alia. 3456 B.11. Changes to draft-ietf-lisp-12.txt 3458 o Posted April 2011. 3460 o Tracker item 87. Provided rewording how an EID-prefix can be 3461 reused in the definition section of "EID-prefix". 3463 o Tracker item 95. Change "eliminate" to "defer" in section 4.1. 3465 o Tracker item 110. Added that the Mapping Protocol Data field in 3466 the Map-Reply message is only used when needed by the particular 3467 Mapping Database System. 3469 o Tracker item 111. Indicate that if an LSB that is associated with 3470 an anycast address, that there is at least one RLOC that is up. 3472 o Tracker item 108. Make clear the R-bit does not define RLOC path 3473 reachability. 3475 o Tracker item 107. Indicate that weights are relative to each 3476 other versus requiring an addition of up to 100%. 3478 o Tracker item 46. Add a sentence how LISP products should be sized 3479 for the appropriate demand so cache thrashing is avoided. 3481 o Change some references of RFC 5226 to [AFI] per Luigi. 3483 o Per Luigi, make reference to "EID-AFI" consistent to "EID-prefix- 3484 AFI". 3486 o Tracker item 66. Indicate that appending locators to a locator- 3487 set is done when the added locators are lexicographically greater 3488 than the previous ones in the set. 3490 o Tracker item 87. Once again reword the definition of the EID- 3491 prefix to reflect recent comments. 3493 o Tracker item 70. Added text to security section on what the 3494 implications could be if an ITR does not obey priority and weights 3495 from a Map-Reply message. 3497 o Tracker item 54. Added text to the new section titled "Packets 3498 Egressing a LISP Site" to describe the implications when two or 3499 more ITRs exist at a site where only one ITR is used for egress 3500 traffic and when there is a shift of traffic to the others, how 3501 the map-cache will need to be populated in those new egress ITRs. 3503 o Tracker item 33. Make more clear in the Routing Locator Selection 3504 section what an ITR should do when it sees an R-bit of 0 in a 3505 locator-record of a Map-Reply. 3507 o Tracker item 33. Add paragraph to the EID Reachability section 3508 indicating that site partitioning is under investigation. 3510 o Tracker item 58. Added last paragraph of Security Considerations 3511 section about how to protect inner header EID address spoofing 3512 attacks. 3514 o Add suggested Sam text to indicate that all security concerns need 3515 not be addressed for moving document to Experimental RFC status. 3516 Put this in a subsection of the Security Considerations section. 3518 B.12. Changes to draft-ietf-lisp-11.txt 3520 o Posted March 30, 2011. 3522 o Change IANA URL. The URL we had pointed to a general protocol 3523 numbers page. 3525 o Added the "s" bit to the Map-Request to allow SMR-invoked Map- 3526 Requests to be sent to a MN ETR via the map-server. 3528 o Generalize text for the definition of Reencapsuatling tunnels. 3530 o Add paragraph suggested by Joel to explain how implementation 3531 experimentation will be used to determine the proper cache 3532 management techniques. 3534 o Add Yakov provided text for the definition of "EID-to-RLOC 3535 "Database". 3537 o Add reference in Section 8, Deployment Scenarios, to the 3538 draft-jakab-lisp-deploy-02.txt draft. 3540 o Clarify sentence about no hardware changes needed to support LISP 3541 encapsulation. 3543 o Add paragraph about what is the procedure when a locator is 3544 inserted in the middle of a locator-set. 3546 o Add a definition for Locator Status Bits so we can emphasize they 3547 are used as a hint for router up/down status and not path 3548 reachability. 3550 o Change "BGP RIB" to "RIB" per Clarence's comment. 3552 o Fixed complaints by IDnits. 3554 o Add subsection to Security Considerations section indicating how 3555 EID-prefix overclaiming in Map-Replies is for further study and 3556 add a reference to LISP-SEC. 3558 B.13. Changes to draft-ietf-lisp-10.txt 3560 o Posted March 2011. 3562 o Add p-bit to Map-Request so there is documentary reasons to know 3563 when a PITR has sent a Map-Request to an ETR. 3565 o Add Map-Notify message which is used to acknowledge a Map-Register 3566 message sent to a Map-Server. 3568 o Add M-bit to the Map-Register message so an ETR that wants an 3569 acknowledgment for the Map-Register can request one. 3571 o Add S-bit to the ECM and Map-Reply messages to describe security 3572 data that can be present in each message. Then refer to 3573 [LISP-SEC] for expansive details. 3575 o Add Network Management Considerations section and point to the MIB 3576 and LIG drafts. 3578 o Remove the word "simple" per Yakov's comments. 3580 B.14. Changes to draft-ietf-lisp-09.txt 3582 o Posted October 2010. 3584 o Add to IANA Consideration section about the use of LCAF Type 3585 values that accepted and maintained by the IANA registry and not 3586 the LCAF specification. 3588 o Indicate that implementations should be able to receive LISP 3589 control messages when either UDP port is 4342, so they can be 3590 robust in the face of intervening NAT boxes. 3592 o Add paragraph to SMR section to indicate that an ITR does not need 3593 to respond to an SMR-based Map-Request when it has no map-cache 3594 entry for the SMR source's EID-prefix. 3596 B.15. Changes to draft-ietf-lisp-08.txt 3598 o Posted August 2010. 3600 o In section 6.1.6, remove statement about setting TTL to 0 in Map- 3601 Register messages. 3603 o Clarify language in section 6.1.5 about Map-Replying to Data- 3604 Probes or Map-Requests. 3606 o Indicate that outer TTL should only be copied to inner TTL when it 3607 is less than inner TTL. 3609 o Indicate a source-EID for RLOC-probes are encoded with an AFI 3610 value of 0. 3612 o Indicate that SMRs can have a global or per SMR destination rate- 3613 limiter. 3615 o Add clarifications to the SMR procedures. 3617 o Add definitions for "client-side" and 'server-side" terms used in 3618 this specification. 3620 o Clear up language in section 6.4, last paragraph. 3622 o Change ACT of value 0 to "no-action". This is so we can RLOC- 3623 probe a PETR and have it return a Map-Reply with a locator-set of 3624 size 0. The way it is spec'ed the map-cache entry has action 3625 "dropped". Drop-action is set to 3. 3627 o Add statement about normalizing locator weights. 3629 o Clarify R-bit definition in the Map-Reply locator record. 3631 o Add section on EID Reachability within a LISP site. 3633 o Clarify another disadvantage of using anycast locators. 3635 o Reworded Abstract. 3637 o Change section 2.0 Introduction to remove obsolete information 3638 such as the LISP variant definitions. 3640 o Change section 5 title from "Tunneling Details" to "LISP 3641 Encapsulation Details". 3643 o Changes to section 5 to include results of network deployment 3644 experience with MTU. Recommend that implementations use either 3645 the stateful or stateless handling. 3647 o Make clarification wordsmithing to Section 7 and 8. 3649 o Identify that if there is one locator in the locator-set of a map- 3650 cache entry, that an SMR from that locator should be responded to 3651 by sending the the SMR-invoked Map-Request to the database mapping 3652 system rather than to the RLOC itself (which may be unreachable). 3654 o When describing Unicast and Multicast Weights indicate the the 3655 values are relative weights rather than percentages. So it 3656 doesn't imply the sum of all locator weights in the locator-set 3657 need to be 100. 3659 o Do some wordsmithing on copying TTL and TOS fields. 3661 o Numerous wordsmithing changes from Dave Meyer. He fine toothed 3662 combed the spec. 3664 o Removed Section 14 "Prototype Plans and Status". We felt this 3665 type of section is no longer appropriate for a protocol 3666 specification. 3668 o Add clarification text for the IRC description per Damien's 3669 commentary. 3671 o Remove text on copying nonce from SMR to SMR-invoked Map- Request 3672 per Vina's comment about a possible DoS vector. 3674 o Clarify (S/2 + H) in the stateless MTU section. 3676 o Add text to reflect Damien's comment about the description of the 3677 "ITR-RLOC Address" field in the Map-Request. that the list of RLOC 3678 addresses are local addresses of the Map-Requester. 3680 B.16. Changes to draft-ietf-lisp-07.txt 3682 o Posted April 2010. 3684 o Added I-bit to data header so LSB field can also be used as an 3685 Instance ID field. When this occurs, the LSB field is reduced to 3686 8-bits (from 32-bits). 3688 o Added V-bit to the data header so the 24-bit nonce field can also 3689 be used for source and destination version numbers. 3691 o Added Map-Version 12-bit value to the EID-record to be used in all 3692 of Map-Request, Map-Reply, and Map-Register messages. 3694 o Added multiple ITR-RLOC fields to the Map-Request packet so an ETR 3695 can decide what address to select for the destination of a Map- 3696 Reply. 3698 o Added L-bit (Local RLOC bit) and p-bit (Probe-Reply RLOC bit) to 3699 the Locator-Set record of an EID-record for a Map-Reply message. 3700 The L-bit indicates which RLOCs in the locator-set are local to 3701 the sender of the message. The P-bit indicates which RLOC is the 3702 source of a RLOC-probe Reply (Map-Reply) message. 3704 o Add reference to the LISP Canonical Address Format [LCAF] draft. 3706 o Made editorial and clarification changes based on comments from 3707 Dhirendra Trivedi. 3709 o Added wordsmithing comments from Joel Halpern on DF=1 setting. 3711 o Add John Zwiebel clarification to Echo Nonce Algorithm section 3712 6.3.1. 3714 o Add John Zwiebel comment about expanding on proxy-map-reply bit 3715 for Map-Register messages. 3717 o Add NAT section per Ron Bonica comments. 3719 o Fix IDnits issues per Ron Bonica. 3721 o Added section on Virtualization and Segmentation to explain the 3722 use if the Instance ID field in the data header. 3724 o There are too many P-bits, keep their scope to the packet format 3725 description and refer to them by name every where else in the 3726 spec. 3728 o Scanned all occurrences of "should", "should not", "must" and 3729 "must not" and uppercased them. 3731 o John Zwiebel offered text for section 4.1 to modernize the 3732 example. Thanks Z! 3734 o Make it more clear in the definition of "EID-to-RLOC Database" 3735 that all ETRs need to have the same database mapping. This 3736 reflects a comment from John Scudder. 3738 o Add a definition "Route-returnability" to the Definition of Terms 3739 section. 3741 o In section 9.2, add text to describe what the signature of 3742 traceroute packets can look like. 3744 o Removed references to Data Probe for introductory example. Data- 3745 probes are still part of the LISP design but not encouraged. 3747 o Added the definition for "LISP site" to the Definition of Terms" 3748 section. 3750 B.17. Changes to draft-ietf-lisp-06.txt 3752 Editorial based changes: 3754 o Posted December 2009. 3756 o Fix typo for flags in LISP data header. Changed from "4" to "5". 3758 o Add text to indicate that Map-Register messages must contain a 3759 computed UDP checksum. 3761 o Add definitions for PITR and PETR. 3763 o Indicate an AFI value of 0 is an unspecified address. 3765 o Indicate that the TTL field of a Map-Register is not used and set 3766 to 0 by the sender. This change makes this spec consistent with 3768 [LISP-MS]. 3770 o Change "... yield a packet size of L octets" to "... yield a 3771 packet size greater than L octets". 3773 o Clarify section 6.1.5 on what addresses and ports are used in Map- 3774 Reply messages. 3776 o Clarify that LSBs that go beyond the number of locators do not to 3777 be SMRed when the locator addresses are greater lexicographically 3778 than the locator in the existing locator-set. 3780 o Add Gregg, Srini, and Amit to acknowledgment section. 3782 o Clarify in the definition of a LISP header what is following the 3783 UDP header. 3785 o Clarify "verifying Map-Request" text in section 6.1.3. 3787 o Add Xu Xiaohu to the acknowledgment section for introducing the 3788 problem of overlapping EID-prefixes among multiple sites in an RRG 3789 email message. 3791 Design based changes: 3793 o Use stronger language to have the outer IPv4 header set DF=1 so we 3794 can avoid fragment reassembly in an ETR or PETR. This will also 3795 make IPv4 and IPv6 encapsulation have consistent behavior. 3797 o Map-Requests should not be sent in ECM with the Probe bit is set. 3798 These type of Map-Requests are used as RLOC-probes and are sent 3799 directly to locator addresses in the underlying network. 3801 o Add text in section 6.1.5 about returning all EID-prefixes in a 3802 Map-Reply sent by an ETR when there are overlapping EID-prefixes 3803 configure. 3805 o Add text in a new subsection of section 6.1.5 about dealing with 3806 Map-Replies with coarse EID-prefixes. 3808 B.18. Changes to draft-ietf-lisp-05.txt 3810 o Posted September 2009. 3812 o Added this Document Change Log appendix. 3814 o Added section indicating that encapsulated Map-Requests must use 3815 destination UDP port 4342. 3817 o Don't use AH in Map-Registers. Put key-id, auth-length, and auth- 3818 data in Map-Register payload. 3820 o Added Jari to acknowledgment section. 3822 o State the source-EID is set to 0 when using Map-Requests to 3823 refresh or RLOC-probe. 3825 o Make more clear what source-RLOC should be for a Map-Request. 3827 o The LISP-CONS authors thought that the Type definitions for CONS 3828 should be removed from this specification. 3830 o Removed nonce from Map-Register message, it wasn't used so no need 3831 for it. 3833 o Clarify what to do for unspecified Action bits for negative Map- 3834 Replies. Since No Action is a drop, make value 0 Drop. 3836 B.19. Changes to draft-ietf-lisp-04.txt 3838 o Posted September 2009. 3840 o How do deal with record count greater than 1 for a Map-Request. 3841 Damien and Joel comment. Joel suggests: 1) Specify that senders 3842 compliant with the current document will always set the count to 3843 1, and note that the count is included for future extensibility. 3844 2) Specify what a receiver compliant with the draft should do if 3845 it receives a request with a count greater than 1. Presumably, it 3846 should send some error back? 3848 o Add Fred Templin in acknowledgment section. 3850 o Add Margaret and Sam to the acknowledgment section for their great 3851 comments. 3853 o Say more about LAGs in the UDP section per Sam Hartman's comment. 3855 o Sam wants to use MAY instead of SHOULD for ignoring checksums on 3856 ETR. From the mailing list: "You'd need to word it as an ITR MAY 3857 send a zero checksum, an ETR MUST accept a 0 checksum and MAY 3858 ignore the checksum completely. And of course we'd need to 3859 confirm that can actually be implemented. In particular, hardware 3860 that verifies UDP checksums on receive needs to be checked to make 3861 sure it permits 0 checksums." 3863 o Margaret wants a reference to 3864 http://www.ietf.org/id/draft-eubanks-chimento-6man-00.txt. 3866 o Fix description in Map-Request section. Where we describe Map- 3867 Reply Record, change "R-bit" to "M-bit". 3869 o Add the mobility bit to Map-Replies. So PITRs don't probe so 3870 often for MNs but often enough to get mapping updates. 3872 o Indicate SHA1 can be used as well for Map-Registers. 3874 o More Fred comments on MTU handling. 3876 o Isidor comment about spec'ing better periodic Map-Registers. Will 3877 be fixed in draft-ietf-lisp-ms-02.txt. 3879 o Margaret's comment on gleaning: "The current specification does 3880 not make it clear how long gleaned map entries should be retained 3881 in the cache, nor does it make it clear how/ when they will be 3882 validated. The LISP spec should, at the very least, include a 3883 (short) default lifetime for gleaned entries, require that they be 3884 validated within a short period of time, and state that a new 3885 gleaned entry should never overwrite an entry that was obtained 3886 from the mapping system. The security implications of storing 3887 "gleaned" entries should also be explored in detail." 3889 o Add section on RLOC-probing per working group feedback. 3891 o Change "loc-reach-bits" to "loc-status-bits" per comment from 3892 Noel. 3894 o Remove SMR-bit from data-plane. Dino prefers to have it in the 3895 control plane only. 3897 o Change LISP header to allow a "Research Bit" so the Nonce and LSB 3898 fields can be turned off and used for another future purpose. For 3899 Luigi et al versioning convergence. 3901 o Add a N-bit to the data header suggested by Noel. Then the nonce 3902 field could be used when N is not 1. 3904 o Clarify that when E-bit is 0, the nonce field can be an echoed 3905 nonce or a random nonce. Comment from Jesper. 3907 o Indicate when doing data-gleaning that a verifying Map-Request is 3908 sent to the source-EID of the gleaned data packet so we can avoid 3909 map-cache corruption by a 3rd party. Comment from Pedro. 3911 o Indicate that a verifying Map-Request, for accepting mapping data, 3912 should be sent over the ALT (or to the EID). 3914 o Reference IPsec RFC 4302. Comment from Sam and Brian Weis. 3916 o Put E-bit in Map-Reply to tell ITRs that the ETR supports echo- 3917 noncing. Comment by Pedro and Dino. 3919 o Jesper made a comment to loosen the language about requiring the 3920 copy of inner TTL to outer TTL since the text to get mixed-AF 3921 traceroute to work would violate the "MUST" clause. Changed from 3922 MUST to SHOULD in section 5.3. 3924 B.20. Changes to draft-ietf-lisp-03.txt 3926 o Posted July 2009. 3928 o Removed loc-reach-bits longword from control packets per Damien 3929 comment. 3931 o Clarifications in MTU text from Roque. 3933 o Added text to indicate that the locator-set be sorted by locator 3934 address from Isidor. 3936 o Clarification text from John Zwiebel in Echo-Nonce section. 3938 B.21. Changes to draft-ietf-lisp-02.txt 3940 o Posted July 2009. 3942 o Encapsulation packet format change to add E-bit and make loc- 3943 reach-bits 32-bits in length. 3945 o Added Echo-Nonce Algorithm section. 3947 o Clarification how ECN bits are copied. 3949 o Moved S-bit in Map-Request. 3951 o Added P-bit in Map-Request and Map-Reply messages to anticipate 3952 RLOC-Probe Algorithm. 3954 o Added to Mobility section to reference [LISP-MN]. 3956 B.22. Changes to draft-ietf-lisp-01.txt 3958 o Posted 2 days after draft-ietf-lisp-00.txt in May 2009. 3960 o Defined LEID to be a "LISP EID". 3962 o Indicate encapsulation use IPv4 DF=0. 3964 o Added negative Map-Reply messages with drop, native-forward, and 3965 send-map-request actions. 3967 o Added Proxy-Map-Reply bit to Map-Register. 3969 B.23. Changes to draft-ietf-lisp-00.txt 3971 o Posted May 2009. 3973 o Rename of draft-farinacci-lisp-12.txt. 3975 o Acknowledgment to RRG. 3977 Authors' Addresses 3979 Dino Farinacci 3980 cisco Systems 3981 Tasman Drive 3982 San Jose, CA 95134 3983 USA 3985 Email: dino@cisco.com 3987 Vince Fuller 3988 cisco Systems 3989 Tasman Drive 3990 San Jose, CA 95134 3991 USA 3993 Email: vaf@cisco.com 3995 Dave Meyer 3996 cisco Systems 3997 170 Tasman Drive 3998 San Jose, CA 3999 USA 4001 Email: dmm@cisco.com 4003 Darrel Lewis 4004 cisco Systems 4005 170 Tasman Drive 4006 San Jose, CA 4007 USA 4009 Email: darlewis@cisco.com