idnits 2.17.1 draft-ietf-lisp-18.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 date (December 14, 2011) is 4510 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for intended status: Experimental ---------------------------------------------------------------------------- == Outdated reference: A later version (-10) exists of draft-ietf-lisp-alt-09 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-ms-12 ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-12) exists of draft-ietf-6man-udpzero-04 == Outdated reference: A later version (-09) exists of draft-ietf-lisp-map-versioning-05 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-interworking-02 == Outdated reference: A later version (-10) exists of draft-ietf-karp-design-guide-06 == Outdated reference: A later version (-10) exists of draft-farinacci-lisp-lcaf-06 == Outdated reference: A later version (-12) exists of draft-ietf-lisp-deployment-02 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-mib-02 == Outdated reference: A later version (-16) exists of draft-meyer-lisp-mn-06 -- Unexpected draft version: The latest known version of draft-meyer-loc-id-implications is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 == Outdated reference: A later version (-14) exists of draft-ietf-lisp-multicast-10 == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-08 Summary: 2 errors (**), 0 flaws (~~), 14 warnings (==), 2 comments (--).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: June 16, 2012 D. Lewis 6 cisco Systems 7 December 14, 2011 9 Locator/ID Separation Protocol (LISP) 10 draft-ietf-lisp-18 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 June 16, 2012. 44 Copyright Notice 46 Copyright (c) 2011 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 . . . . . . . . . . . . . . . . . . . . . . . . 13 65 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . . 15 66 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . . 17 67 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 18 68 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 18 69 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . . 20 70 5.4. Dealing with Large Encapsulated Packets . . . . . . . . . 24 71 5.4.1. A Stateless Solution to MTU Handling . . . . . . . . . 24 72 5.4.2. A Stateful Solution to MTU Handling . . . . . . . . . 25 73 5.5. Using Virtualization and Segmentation with LISP . . . . . 25 74 6. EID-to-RLOC Mapping . . . . . . . . . . . . . . . . . . . . . 27 75 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats . . . . . 27 76 6.1.1. LISP Packet Type Allocations . . . . . . . . . . . . . 29 77 6.1.2. Map-Request Message Format . . . . . . . . . . . . . . 29 78 6.1.3. EID-to-RLOC UDP Map-Request Message . . . . . . . . . 32 79 6.1.4. Map-Reply Message Format . . . . . . . . . . . . . . . 33 80 6.1.5. EID-to-RLOC UDP Map-Reply Message . . . . . . . . . . 37 81 6.1.6. Map-Register Message Format . . . . . . . . . . . . . 39 82 6.1.7. Map-Notify Message Format . . . . . . . . . . . . . . 41 83 6.1.8. Encapsulated Control Message Format . . . . . . . . . 42 84 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 44 85 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 46 86 6.3.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . 48 87 6.3.2. RLOC Probing Algorithm . . . . . . . . . . . . . . . . 49 88 6.4. EID Reachability within a LISP Site . . . . . . . . . . . 50 89 6.5. Routing Locator Hashing . . . . . . . . . . . . . . . . . 51 90 6.6. Changing the Contents of EID-to-RLOC Mappings . . . . . . 52 91 6.6.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 53 92 6.6.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 53 93 6.6.3. Database Map Versioning . . . . . . . . . . . . . . . 55 94 7. Router Performance Considerations . . . . . . . . . . . . . . 56 95 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 57 96 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 58 97 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 58 98 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 59 99 8.4. LISP Functionality with Conventional NATs . . . . . . . . 59 100 8.5. Packets Egressing a LISP Site . . . . . . . . . . . . . . 60 101 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 61 102 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 62 103 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 62 104 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 62 105 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 64 106 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 64 107 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 64 108 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 64 109 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 66 110 10.5. LISP Mobile Node Mobility . . . . . . . . . . . . . . . . 66 111 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 68 112 12. Security Considerations . . . . . . . . . . . . . . . . . . . 69 113 13. Network Management Considerations . . . . . . . . . . . . . . 71 114 14. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 72 115 14.1. LISP ACT and Flag Fields . . . . . . . . . . . . . . . . . 72 116 14.2. LISP Address Type Codes . . . . . . . . . . . . . . . . . 72 117 14.3. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . . 72 118 14.4. LISP Key ID Numbers . . . . . . . . . . . . . . . . . . . 73 119 15. Known Open Issues and Areas of Future Work . . . . . . . . . . 74 120 16. References . . . . . . . . . . . . . . . . . . . . . . . . . . 76 121 16.1. Normative References . . . . . . . . . . . . . . . . . . . 76 122 16.2. Informative References . . . . . . . . . . . . . . . . . . 77 123 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 81 124 Appendix B. Document Change Log . . . . . . . . . . . . . . . . . 82 125 B.1. Changes to draft-ietf-lisp-18.txt . . . . . . . . . . . . 82 126 B.2. Changes to draft-ietf-lisp-17.txt . . . . . . . . . . . . 82 127 B.3. Changes to draft-ietf-lisp-16.txt . . . . . . . . . . . . 82 128 B.4. Changes to draft-ietf-lisp-15.txt . . . . . . . . . . . . 82 129 B.5. Changes to draft-ietf-lisp-14.txt . . . . . . . . . . . . 82 130 B.6. Changes to draft-ietf-lisp-13.txt . . . . . . . . . . . . 83 131 B.7. Changes to draft-ietf-lisp-12.txt . . . . . . . . . . . . 84 132 B.8. Changes to draft-ietf-lisp-11.txt . . . . . . . . . . . . 85 133 B.9. Changes to draft-ietf-lisp-10.txt . . . . . . . . . . . . 86 134 B.10. Changes to draft-ietf-lisp-09.txt . . . . . . . . . . . . 86 135 B.11. Changes to draft-ietf-lisp-08.txt . . . . . . . . . . . . 87 136 B.12. Changes to draft-ietf-lisp-07.txt . . . . . . . . . . . . 88 137 B.13. Changes to draft-ietf-lisp-06.txt . . . . . . . . . . . . 90 138 B.14. Changes to draft-ietf-lisp-05.txt . . . . . . . . . . . . 91 139 B.15. Changes to draft-ietf-lisp-04.txt . . . . . . . . . . . . 92 140 B.16. Changes to draft-ietf-lisp-03.txt . . . . . . . . . . . . 93 141 B.17. Changes to draft-ietf-lisp-02.txt . . . . . . . . . . . . 94 142 B.18. Changes to draft-ietf-lisp-01.txt . . . . . . . . . . . . 94 143 B.19. Changes to draft-ietf-lisp-00.txt . . . . . . . . . . . . 94 145 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 95 147 1. Requirements Notation 149 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 150 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 151 document are to be interpreted as described in [RFC2119]. 153 2. Introduction 155 This document describes the Locator/Identifier Separation Protocol 156 (LISP), which provides a set of functions for routers to exchange 157 information used to map from non globally routeable Endpoint 158 Identifiers (EIDs) to routeable Routing Locators (RLOCs). It also 159 defines a mechanism for these LISP routers to encapsulate IP packets 160 addressed with EIDs for transmission across the< Internet that uses 161 RLOCs for routing and forwarding. 163 Creation of LISP was initially motivated by discussions during the 164 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 165 October, 2006 (see [RFC4984]). A key conclusion of the workshop was 166 that the Internet routing and addressing system was not scaling well 167 in the face of the explosive growth of new sites; one reason for this 168 poor scaling is the increasing number of multi-homed and other sites 169 that cannot be addressed as part of topologically- or provider-based 170 aggregated prefixes. Additional work that more completely described 171 the problem statement may be found in [RADIR]. 173 A basic observation, made many years ago in early networking research 174 such as that documented in [CHIAPPA] and [RFC4984], is that using a 175 single address field for both identifying a device and for 176 determining where it is topologically located in the network requires 177 optimization along two conflicting axes: for routing to be efficient, 178 the address must be assigned topologically; for collections of 179 devices to be easily and effectively managed, without the need for 180 renumbering in response to topological change (such as that caused by 181 adding or removing attachment points to the network or by mobility 182 events), the address must explicitly not be tied to the topology. 184 The approach that LISP takes to solving the routing scalability 185 problem is to replace IP addresses with two new types of numbers: 186 Routing Locators (RLOCs), which are topologically assigned to network 187 attachment points (and are therefore amenable to aggregation) and 188 used for routing and forwarding of packets through the network; and 189 Endpoint Identifiers (EIDs), which are assigned independently from 190 the network topology, are used for numbering devices, and are 191 aggregated along administrative boundaries. LISP then defines 192 functions for mapping between the two numbering spaces and for 193 encapsulating traffic originated by devices using non-routeable EIDs 194 for transport across a network infrastructure that routes and 195 forwards using RLOCs. Both RLOCs and EIDs are syntactically- 196 identical to IP addresses; it is the semantics of how they are used 197 that differs. 199 This document describes the protocol that implements these functions. 200 The database which stores the mappings between EIDs and RLOCs is 201 explicitly a separate "module" to facilitate experimentation with a 202 variety of approaches. One database design that is being developed 203 and prototyped as part of the LISP working group work is [ALT]. 204 Others that have been described but not implemented include [CONS], 205 [EMACS], [NERD]. Finally, [LISP-MS], documents a general-purpose 206 service interface for accessing a mapping database; this interface is 207 intended to make the mapping database modular so that different 208 approaches can be tried without the need to modify installed LISP 209 capable devices in LISP sites. 211 This experimental specification has areas that require additional 212 experience and measurement. Results of such work may lead to 213 modifications and enhancements of protocol mechanisms defined in this 214 document. See Section 15 for specific, known issues that are in need 215 of further work during development, implementation, and prototype 216 deployment. 218 3. Definition of Terms 220 Provider Independent (PI) Addresses: PI addresses are an address 221 block assigned from a pool where blocks are not associated with 222 any particular location in the network (e.g. from a particular 223 service provider), and is therefore not topologically aggregatable 224 in the routing system. 226 Provider Assigned (PA) Addresses: PA addresses are an an address 227 block assigned to a site by each service provider to which a site 228 connects. Typically, each block is sub-block of a service 229 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 230 is aggregated into the larger block before being advertised into 231 the global Internet. Traditionally, IP multihoming has been 232 implemented by each multi-homed site acquiring its own, globally- 233 visible prefix. LISP uses only topologically-assigned and 234 aggregatable address blocks for RLOCs, eliminating this 235 demonstrably non-scalable practice. 237 Routing Locator (RLOC): A RLOC is an IPv4 or IPv6 address of an 238 egress tunnel router (ETR). A RLOC is the output of an EID-to- 239 RLOC mapping lookup. An EID maps to one or more RLOCs. 240 Typically, RLOCs are numbered from topologically-aggregatable 241 blocks that are assigned to a site at each point to which it 242 attaches to the global Internet; where the topology is defined by 243 the connectivity of provider networks, RLOCs can be thought of as 244 PA addresses. Multiple RLOCs can be assigned to the same ETR 245 device or to multiple ETR devices at a site. 247 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 248 IPv6) value used in the source and destination address fields of 249 the first (most inner) LISP header of a packet. The host obtains 250 a destination EID the same way it obtains an destination address 251 today, for example through a Domain Name System (DNS) [RFC1034] 252 lookup or Session Invitation Protocol (SIP) [RFC3261] exchange. 253 The source EID is obtained via existing mechanisms used to set a 254 host's "local" IP address. An EID is allocated to a host from an 255 EID-prefix block associated with the site where the host is 256 located. An EID can be used by a host to refer to other hosts. 257 EIDs MUST NOT be used as LISP RLOCs. Note that EID blocks MAY be 258 assigned in a hierarchical manner, independent of the network 259 topology, to facilitate scaling of the mapping database. In 260 addition, an EID block assigned to a site may have site-local 261 structure (subnetting) for routing within the site; this structure 262 is not visible to the global routing system. In theory, the bit 263 string that represents an EID for one device can represent an RLOC 264 for a different device. As the architecture is realized, if a 265 given bit string is both an RLOC and an EID, it must refer to the 266 same entity in both cases. When used in discussions with other 267 Locator/ID separation proposals, a LISP EID will be called a 268 "LEID". Throughout this document, any references to "EID" refers 269 to an LEID. 271 EID-prefix: An EID-prefix is a power-of-two block of EIDs which are 272 allocated to a site by an address allocation authority. EID- 273 prefixes are associated with a set of RLOC addresses which make up 274 a "database mapping". EID-prefix allocations can be broken up 275 into smaller blocks when an RLOC set is to be associated with the 276 larger EID-prefix block. A globally routed address block (whether 277 PI or PA) is not inherently an EID-prefix. A globally routed 278 address block MAY be used by its assignee as an EID block. The 279 converse is not supported. That is, a site which receives an 280 explicitly allocated EID-prefix may not use that EID-prefix as a 281 globally routed prefix. This would require coordination and 282 cooperation with the entities managing the mapping infrastructure. 283 Once this has been done, that block could be removed from the 284 globally routed IP system, if other suitable transition and access 285 mechanisms are in place. Discussion of such transition and access 286 mechanisms can be found in [INTERWORK] and [LISP-DEPLOY]. 288 End-system: An end-system is an IPv4 or IPv6 device that originates 289 packets with a single IPv4 or IPv6 header. The end-system 290 supplies an EID value for the destination address field of the IP 291 header when communicating globally (i.e. outside of its routing 292 domain). An end-system can be a host computer, a switch or router 293 device, or any network appliance. 295 Ingress Tunnel Router (ITR): An ITR is a router which accepts an IP 296 packet with a single IP header (more precisely, an IP packet that 297 does not contain a LISP header). The router treats this "inner" 298 IP destination address as an EID and performs an EID-to-RLOC 299 mapping lookup. The router then prepends an "outer" IP header 300 with one of its globally-routable RLOCs in the source address 301 field and the result of the mapping lookup in the destination 302 address field. Note that this destination RLOC MAY be an 303 intermediate, proxy device that has better knowledge of the EID- 304 to-RLOC mapping closer to the destination EID. In general, an ITR 305 receives IP packets from site end-systems on one side and sends 306 LISP-encapsulated IP packets toward the Internet on the other 307 side. 309 Specifically, when a service provider prepends a LISP header for 310 Traffic Engineering purposes, the router that does this is also 311 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 312 on the outer destination address (the originating ITR's supplied 313 RLOC) or the inner destination address (the originating hosts 314 supplied EID). 316 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 317 network that prepends an additional LISP header for Traffic 318 Engineering purposes. 320 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 321 packet where the destination address in the "outer" IP header is 322 one of its own RLOCs. The router strips the "outer" header and 323 forwards the packet based on the next IP header found. In 324 general, an ETR receives LISP-encapsulated IP packets from the 325 Internet on one side and sends decapsulated IP packets to site 326 end-systems on the other side. ETR functionality does not have to 327 be limited to a router device. A server host can be the endpoint 328 of a LISP tunnel as well. 330 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 331 network that strips an outer LISP header for Traffic Engineering 332 purposes. 334 xTR: A xTR is a reference to an ITR or ETR when direction of data 335 flow is not part of the context description. xTR refers to the 336 router that is the tunnel endpoint. Used synonymously with the 337 term "Tunnel Router". For example, "An xTR can be located at the 338 Customer Edge (CE) router", meaning both ITR and ETR functionality 339 is at the CE router. 341 LISP Router: A LISP router is a router that performs the functions 342 of any or all of ITR, ETR, PITR, or PETR. 344 EID-to-RLOC Cache: The EID-to-RLOC cache is a short-lived, on- 345 demand table in an ITR that stores, tracks, and is responsible for 346 timing-out and otherwise validating EID-to-RLOC mappings. This 347 cache is distinct from the full "database" of EID-to-RLOC 348 mappings, it is dynamic, local to the ITR(s), and relatively small 349 while the database is distributed, relatively static, and much 350 more global in scope. 352 EID-to-RLOC Database: The EID-to-RLOC database is a global 353 distributed database that contains all known EID-prefix to RLOC 354 mappings. Each potential ETR typically contains a small piece of 355 the database: the EID-to-RLOC mappings for the EID prefixes 356 "behind" the router. These map to one of the router's own, 357 globally-visible, IP addresses. The same database mapping entries 358 MUST be configured on all ETRs for a given site. In a steady 359 state the EID-prefixes for the site and the locator-set for each 360 EID-prefix MUST be the same on all ETRs. Procedures to enforce 361 and/or verify this are outside the scope of this document. Note 362 that there MAY be transient conditions when the EID-prefix for the 363 site and locator-set for each EID-prefix may not be the same on 364 all ETRs. This has no negative implications. 366 Recursive Tunneling: Recursive tunneling occurs when a packet has 367 more than one LISP IP header. Additional layers of tunneling MAY 368 be employed to implement traffic engineering or other re-routing 369 as needed. When this is done, an additional "outer" LISP header 370 is added and the original RLOCs are preserved in the "inner" 371 header. Any references to tunnels in this specification refers to 372 dynamic encapsulating tunnels and they are never statically 373 configured. 375 Reencapsulating Tunnels: Reencapsulating tunneling occurs when an 376 ETR removes a LISP header, then acts as an ITR to prepend another 377 LISP header. Doing this allows a packet to be re-routed by the 378 re-encapsulating router without adding the overhead of additional 379 tunnel headers. Any references to tunnels in this specification 380 refers to dynamic encapsulating tunnels and they are never 381 statically configured. When using multiple mapping database 382 systems, care must be taken to not create reencapsulation loops. 384 LISP Header: a term used in this document to refer to the outer 385 IPv4 or IPv6 header, a UDP header, and a LISP-specific 8-byte 386 header that follows the UDP header, an ITR prepends or an ETR 387 strips. 389 Address Family Identifier (AFI): a term used to describe an address 390 encoding in a packet. An address family currently pertains to an 391 IPv4 or IPv6 address. See [AFI]/[AFI-REGISTRY] and [RFC3232] for 392 details. An AFI value of 0 used in this specification indicates 393 an unspecified encoded address where the length of the address is 394 0 bytes following the 16-bit AFI value of 0. 396 Negative Mapping Entry: A negative mapping entry, also known as a 397 negative cache entry, is an EID-to-RLOC entry where an EID-prefix 398 is advertised or stored with no RLOCs. That is, the locator-set 399 for the EID-to-RLOC entry is empty or has an encoded locator count 400 of 0. This type of entry could be used to describe a prefix from 401 a non-LISP site, which is explicitly not in the mapping database. 402 There are a set of well defined actions that are encoded in a 403 Negative Map-Reply. 405 Data Probe: A data-probe is a LISP-encapsulated data packet where 406 the inner header destination address equals the outer header 407 destination address used to trigger a Map-Reply by a decapsulating 408 ETR. In addition, the original packet is decapsulated and 409 delivered to the destination host if the destination EID is in the 410 EID-prefix range configured on the ETR. Otherwise, the packet is 411 discarded. A Data Probe is used in some of the mapping database 412 designs to "probe" or request a Map-Reply from an ETR; in other 413 cases, Map-Requests are used. See each mapping database design 414 for details. 416 Proxy ITR (PITR): A PITR is also known as a PTR is defined and 417 described in [INTERWORK], a PITR acts like an ITR but does so on 418 behalf of non-LISP sites which send packets to destinations at 419 LISP sites. 421 Proxy ETR (PETR): A PETR is defined and described in [INTERWORK], a 422 PETR acts like an ETR but does so on behalf of LISP sites which 423 send packets to destinations at non-LISP sites. 425 Route-returnability: is an assumption that the underlying routing 426 system will deliver packets to the destination. When combined 427 with a nonce that is provided by a sender and returned by a 428 receiver, this limits off-path data insertion. 430 LISP site: is a set of routers in an edge network that are under a 431 single technical administration. LISP routers which reside in the 432 edge network are the demarcation points to separate the edge 433 network from the core network. 435 Client-side: a term used in this document to indicate a connection 436 initiation attempt by an EID. The ITR(s) at the LISP site are the 437 first to get involved in obtaining database map cache entries by 438 sending Map-Request messages. 440 Server-side: a term used in this document to indicate a connection 441 initiation attempt is being accepted for a destination EID. The 442 ETR(s) at the destination LISP site are the first to send Map- 443 Replies to the source site initiating the connection. The ETR(s) 444 at this destination site can obtain mappings by gleaning 445 information from Map-Requests, Data-Probes, or encapsulated 446 packets. 448 Locator Status Bits (LSBs): Locator status bits are present in the 449 LISP header. They are used by ITRs to inform ETRs about the up/ 450 down status of all ETRs at the local site. These bits are used as 451 a hint to convey up/down router status and not path reachability 452 status. The LSBs can be verified by use of one of the Locator 453 Reachability Algorithms described in Section 6.3. 455 4. Basic Overview 457 One key concept of LISP is that end-systems (hosts) operate the same 458 way they do today. The IP addresses that hosts use for tracking 459 sockets, connections, and for sending and receiving packets do not 460 change. In LISP terminology, these IP addresses are called Endpoint 461 Identifiers (EIDs). 463 Routers continue to forward packets based on IP destination 464 addresses. When a packet is LISP encapsulated, these addresses are 465 referred to as Routing Locators (RLOCs). Most routers along a path 466 between two hosts will not change; they continue to perform routing/ 467 forwarding lookups on the destination addresses. For routers between 468 the source host and the ITR as well as routers from the ETR to the 469 destination host, the destination address is an EID. For the routers 470 between the ITR and the ETR, the destination address is an RLOC. 472 Another key LISP concept is the "Tunnel Router". A tunnel router 473 prepends LISP headers on host-originated packets and strips them 474 prior to final delivery to their destination. The IP addresses in 475 this "outer header" are RLOCs. During end-to-end packet exchange 476 between two Internet hosts, an ITR prepends a new LISP header to each 477 packet and an egress tunnel router strips the new header. The ITR 478 performs EID-to-RLOC lookups to determine the routing path to the 479 ETR, which has the RLOC as one of its IP addresses. 481 Some basic rules governing LISP are: 483 o End-systems (hosts) only send to addresses which are EIDs. They 484 don't know addresses are EIDs versus RLOCs but assume packets get 485 to destinations, which in turn, LISP routers deliver packets to 486 the destination the end-system has specified. The procedure a 487 host uses to send IP packets does not change. 489 o EIDs are always IP addresses assigned to hosts. 491 o LISP routers mostly deal with Routing Locator addresses. See 492 details later in Section 4.1 to clarify what is meant by "mostly". 494 o RLOCs are always IP addresses assigned to routers; preferably, 495 topologically-oriented addresses from provider CIDR (Classless 496 Inter-Domain Routing) blocks. 498 o When a router originates packets it may use as a source address 499 either an EID or RLOC. When acting as a host (e.g. when 500 terminating a transport session such as SSH, TELNET, or SNMP), it 501 may use an EID that is explicitly assigned for that purpose. An 502 EID that identifies the router as a host MUST NOT be used as an 503 RLOC; an EID is only routable within the scope of a site. A 504 typical BGP configuration might demonstrate this "hybrid" EID/RLOC 505 usage where a router could use its "host-like" EID to terminate 506 iBGP sessions to other routers in a site while at the same time 507 using RLOCs to terminate eBGP sessions to routers outside the 508 site. 510 o Packets with EIDs in them are not expected to be delivered end-to- 511 end in the absence of an EID-to-RLOC mapping operation. They are 512 expected to be used locally for intra-site communication or to be 513 encapsulated for inter-site communication. 515 o EID prefixes are likely to be hierarchically assigned in a manner 516 which is optimized for administrative convenience and to 517 facilitate scaling of the EID-to-RLOC mapping database. The 518 hierarchy is based on a address allocation hierarchy which is 519 independent of the network topology. 521 o EIDs may also be structured (subnetted) in a manner suitable for 522 local routing within an autonomous system. 524 An additional LISP header MAY be prepended to packets by a TE-ITR 525 when re-routing of the path for a packet is desired. A potential 526 use-case for this would be an ISP router that needs to perform 527 traffic engineering for packets flowing through its network. In such 528 a situation, termed Recursive Tunneling, an ISP transit acts as an 529 additional ingress tunnel router and the RLOC it uses for the new 530 prepended header would be either a TE-ETR within the ISP (along 531 intra-ISP traffic engineered path) or a TE-ETR within another ISP (an 532 inter-ISP traffic engineered path, where an agreement to build such a 533 path exists). 535 In order to avoid excessive packet overhead as well as possible 536 encapsulation loops, this document mandates that a maximum of two 537 LISP headers can be prepended to a packet. For initial LISP 538 deployments, it is assumed two headers is sufficient, where the first 539 prepended header is used at a site for Location/Identity separation 540 and second prepended header is used inside a service provider for 541 Traffic Engineering purposes. 543 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 544 For example, the ITR for a particular end-to-end packet exchange 545 might be the first-hop or default router within a site for the source 546 host. Similarly, the egress tunnel router might be the last-hop 547 router directly-connected to the destination host. Another example, 548 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 549 could be the site's border router at the service provider attachment 550 point. Mixing and matching of site-operated, ISP-operated, and other 551 tunnel routers is allowed for maximum flexibility. See Section 8 for 552 more details. 554 4.1. Packet Flow Sequence 556 This section provides an example of the unicast packet flow with the 557 following conditions: 559 o Source host "host1.abc.example.com" is sending a packet to 560 "host2.xyz.example.com", exactly what host1 would do if the site 561 was not using LISP. 563 o Each site is multi-homed, so each tunnel router has an address 564 (RLOC) assigned from the service provider address block for each 565 provider to which that particular tunnel router is attached. 567 o The ITR(s) and ETR(s) are directly connected to the source and 568 destination, respectively, but the source and destination can be 569 located anywhere in LISP site. 571 o Map-Requests can be sent on the underlying routing system 572 topology, to a mapping database system, or directly over an 573 alternative topology [ALT]. 575 o Map-Replies are sent on the underlying routing system topology. 577 Client host1.abc.example.com wants to communicate with server 578 host2.xyz.example.com: 580 1. host1.abc.example.com wants to open a TCP connection to 581 host2.xyz.example.com. It does a DNS lookup on 582 host2.xyz.example.com. An A/AAAA record is returned. This 583 address is the destination EID. The locally-assigned address of 584 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 585 packet is built and forwarded through the LISP site as a normal 586 IP packet until it reaches a LISP ITR. 588 2. The LISP ITR must be able to map the destination EID to an RLOC 589 of one of the ETRs at the destination site. The specific method 590 used to do this is not described in this example. See [ALT] or 591 [CONS] for possible solutions. 593 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be 594 rate-limited. 596 4. When an alternate mapping system is not in use, the Map-Request 597 packet is routed through the underlying routing system. 598 Otherwise, the Map-Request packet is routed on an alternate 599 logical topology, for example the [ALT] database mapping system. 600 In either case, when the Map-Request arrives at one of the ETRs 601 at the destination site, it will process the packet as a control 602 message. 604 5. The ETR looks at the destination EID of the Map-Request and 605 matches it against the prefixes in the ETR's configured EID-to- 606 RLOC mapping database. This is the list of EID-prefixes the ETR 607 is supporting for the site it resides in. If there is no match, 608 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 609 returned to the ITR. 611 6. The ITR receives the Map-Reply message, parses the message (to 612 check for format validity) and stores the mapping information 613 from the packet. This information is stored in the ITR's EID-to- 614 RLOC mapping cache. Note that the map cache is an on-demand 615 cache. An ITR will manage its map cache in such a way that 616 optimizes for its resource constraints. 618 7. Subsequent packets from host1.abc.example.com to 619 host2.xyz.example.com will have a LISP header prepended by the 620 ITR using the appropriate RLOC as the LISP header destination 621 address learned from the ETR. Note the packet MAY be sent to a 622 different ETR than the one which returned the Map-Reply due to 623 the source site's hashing policy or the destination site's 624 locator-set policy. 626 8. The ETR receives these packets directly (since the destination 627 address is one of its assigned IP addresses), checks the validity 628 of the addresses, strips the LISP header, and forwards packets to 629 the attached destination host. 631 In order to defer the need for a mapping lookup in the reverse 632 direction, an ETR MAY create a cache entry that maps the source EID 633 (inner header source IP address) to the source RLOC (outer header 634 source IP address) in a received LISP packet. Such a cache entry is 635 termed a "gleaned" mapping and only contains a single RLOC for the 636 EID in question. More complete information about additional RLOCs 637 SHOULD be verified by sending a LISP Map-Request for that EID. Both 638 ITR and the ETR may also influence the decision the other makes in 639 selecting an RLOC. See Section 6 for more details. 641 5. LISP Encapsulation Details 643 Since additional tunnel headers are prepended, the packet becomes 644 larger and can exceed the MTU of any link traversed from the ITR to 645 the ETR. It is RECOMMENDED in IPv4 that packets do not get 646 fragmented as they are encapsulated by the ITR. Instead, the packet 647 is dropped and an ICMP Too Big message is returned to the source. 649 This specification RECOMMENDS that implementations provide support 650 for one of the proposed fragmentation and reassembly schemes. Two 651 existing schemes are detailed in Section 5.4. 653 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 654 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 655 header is in IPv4 packet format and the other header is in IPv6 656 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 657 is in IPv6 packet format and the other header is in IPv4 packet 658 format). The next sub-sections illustrate packet formats for the 659 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6) but all 4 660 combinations MUST be supported. 662 5.1. LISP IPv4-in-IPv4 Header Format 664 0 1 2 3 665 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 1 666 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 667 / |Version| IHL |Type of Service| Total Length | 668 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 669 | | Identification |Flags| Fragment Offset | 670 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 671 OH | Time to Live | Protocol = 17 | Header Checksum | 672 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 673 | | Source Routing Locator | 674 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 675 \ | Destination Routing Locator | 676 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 677 / | Source Port = xxxx | Dest Port = 4341 | 678 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 679 \ | UDP Length | UDP Checksum | 680 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 681 L |N|L|E|V|I|flags| Nonce/Map-Version | 682 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 683 S / | Instance ID/Locator Status Bits | 684 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 685 / |Version| IHL |Type of Service| Total Length | 686 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 687 | | Identification |Flags| Fragment Offset | 688 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 689 IH | Time to Live | Protocol | Header Checksum | 690 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 691 | | Source EID | 692 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 693 \ | Destination EID | 694 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 696 5.2. LISP IPv6-in-IPv6 Header Format 698 0 1 2 3 699 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 1 700 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 701 / |Version| Traffic Class | Flow Label | 702 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 704 | | Payload Length | Next Header=17| Hop Limit | 705 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 706 | | 707 O + + 708 u | | 709 t + Source Routing Locator + 710 e | | 711 r + + 712 | | 713 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 714 d | | 715 r + + 716 | | 717 ^ + Destination Routing Locator + 718 | | | 719 \ + + 720 \ | | 721 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 722 / | Source Port = xxxx | Dest Port = 4341 | 723 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 724 \ | UDP Length | UDP Checksum | 725 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 726 L |N|L|E|V|I|flags| Nonce/Map-Version | 727 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 728 S / | Instance ID/Locator Status Bits | 729 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 730 / |Version| Traffic Class | Flow Label | 731 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 732 / | Payload Length | Next Header | Hop Limit | 733 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 734 | | 735 I + + 736 n | | 737 n + Source EID + 738 e | | 739 r + + 740 | | 741 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 742 d | | 743 r + + 744 | | 745 ^ + Destination EID + 746 \ | | 747 \ + + 748 \ | | 749 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 751 5.3. Tunnel Header Field Descriptions 753 Inner Header: The inner header is the header on the datagram 754 received from the originating host. The source and destination IP 755 addresses are EIDs, [RFC0791], [RFC2460]. 757 Outer Header: The outer header is a new header prepended by an ITR. 758 The address fields contain RLOCs obtained from the ingress 759 router's EID-to-RLOC cache. The IP protocol number is "UDP (17)" 760 from [RFC0768]. The setting of the DF bit Flags field is 761 according to rules in Section 5.4.1 and Section 5.4.2. 763 UDP Header: The UDP header contains an ITR selected source port when 764 encapsulating a packet. See Section 6.5 for details on the hash 765 algorithm used to select a source port based on the 5-tuple of the 766 inner header. The destination port MUST be set to the well-known 767 IANA assigned port value 4341. 769 UDP Checksum: The UDP checksum field SHOULD be transmitted as zero 770 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 771 [UDP-TUNNELS] [UDP-ZERO]. When a packet with a zero UDP checksum 772 is received by an ETR, the ETR MUST accept the packet for 773 decapsulation. When an ITR transmits a non-zero value for the UDP 774 checksum, it MUST send a correctly computed value in this field. 775 When an ETR receives a packet with a non-zero UDP checksum, it MAY 776 choose to verify the checksum value. If it chooses to perform 777 such verification, and the verification fails, the packet MUST be 778 silently dropped. If the ETR chooses not to perform the 779 verification, or performs the verification successfully, the 780 packet MUST be accepted for decapsulation. The handling of UDP 781 checksums for all tunneling protocols, including LISP, is under 782 active discussion within the IETF. When that discussion 783 concludes, any necessary changes will be made to align LISP with 784 the outcome of the broader discussion. 786 UDP Length: The UDP length field is set for an IPv4 encapsulated 787 packet to be the sum of the inner header IPv4 Total Length plus 788 the UDP and LISP header lengths. For an IPv6 encapsulated packet, 789 the UDP length field is the sum of the inner header IPv6 Payload 790 Length, the size of the IPv6 header (40 bytes), and the size of 791 the UDP and LISP headers. 793 N: The N bit is the nonce-present bit. When this bit is set to 1, 794 the low-order 24-bits of the first 32-bits of the LISP header 795 contains a Nonce. See Section 6.3.1 for details. Both N and V 796 bits MUST NOT be set in the same packet. If they are, a 797 decapsulating ETR MUST treat the "Nonce/Map-Version" field as 798 having a Nonce value present. 800 L: The L bit is the Locator Status Bits field enabled bit. When this 801 bit is set to 1, the Locator Status Bits in the second 32-bits of 802 the LISP header are in use. 804 x 1 x x 0 x x x 805 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 806 |N|L|E|V|I|flags| Nonce/Map-Version | 807 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 808 | Locator Status Bits | 809 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 811 E: The E bit is the echo-nonce-request bit. This bit MUST be ignored 812 and has no meaning when the N bit is set to 0. When the N bit is 813 set to 1 and this bit is set to 1, means an ITR is requesting for 814 the nonce value in the Nonce field to be echoed back in LISP 815 encapsulated packets when the ITR is also an ETR. See 816 Section 6.3.1 for details. 818 V: The V bit is the Map-Version present bit. When this bit is set to 819 1, the N bit MUST be 0. Refer to Section 6.6.3 for more details. 820 This bit indicates that the LISP header is encoded in this case 821 as: 823 0 x 0 1 x x x x 824 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 825 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version | 826 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 827 | Instance ID/Locator Status Bits | 828 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 830 I: The I bit is the Instance ID bit. See Section 5.5 for more 831 details. When this bit is set to 1, the Locator Status Bits field 832 is reduced to 8-bits and the high-order 24-bits are used as an 833 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 834 are transmitted as zero and ignored on receipt. The format of the 835 LISP header would look like in this case: 837 x x x x 1 x x x 838 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 839 |N|L|E|V|I|flags| Nonce/Map-Version | 840 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 841 | Instance ID | LSBs | 842 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 844 flags: The flags field is a 3-bit field is reserved for future flag 845 use. It MUST be set to 0 on transmit and MUST be ignored on 846 receipt. 848 LISP Nonce: The LISP nonce field is a 24-bit value that is randomly 849 generated by an ITR when the N-bit is set to 1. Nonce generation 850 algorithms are an implementation matter but are required to 851 generate different nonces when sending to different destinations. 852 However, the same nonce can be used for a period of time to the 853 same destination. The nonce is also used when the E-bit is set to 854 request the nonce value to be echoed by the other side when 855 packets are returned. When the E-bit is clear but the N-bit is 856 set, a remote ITR is either echoing a previously requested echo- 857 nonce or providing a random nonce. See Section 6.3.1 for more 858 details. 860 LISP Locator Status Bits (LSBs): When the L-bit is also set, the 861 locator status bits field in the LISP header is set by an ITR to 862 indicate to an ETR the up/down status of the Locators in the 863 source site. Each RLOC in a Map-Reply is assigned an ordinal 864 value from 0 to n-1 (when there are n RLOCs in a mapping entry). 865 The Locator Status Bits are numbered from 0 to n-1 from the least 866 significant bit of field. The field is 32-bits when the I-bit is 867 set to 0 and is 8 bits when the I-bit is set to 1. When a Locator 868 Status Bit is set to 1, the ITR is indicating to the ETR the RLOC 869 associated with the bit ordinal has up status. See Section 6.3 870 for details on how an ITR can determine the status of the ETRs at 871 the same site. When a site has multiple EID-prefixes which result 872 in multiple mappings (where each could have a different locator- 873 set), the Locator Status Bits setting in an encapsulated packet 874 MUST reflect the mapping for the EID-prefix that the inner-header 875 source EID address matches. If the LSB for an anycast locator is 876 set to 1, then there is at least one RLOC with that address the 877 ETR is considered 'up'. 879 When doing ITR/PITR encapsulation: 881 o The outer header Time to Live field (or Hop Limit field, in case 882 of IPv6) SHOULD be copied from the inner header Time to Live 883 field. 885 o The outer header Type of Service field (or the Traffic Class 886 field, in the case of IPv6) SHOULD be copied from the inner header 887 Type of Service field (with one caveat, see below). 889 When doing ETR/PETR decapsulation: 891 o The inner header Time to Live field (or Hop Limit field, in case 892 of IPv6) SHOULD be copied from the outer header Time to Live 893 field, when the Time to Live field of the outer header is less 894 than the Time to Live of the inner header. Failing to perform 895 this check can cause the Time to Live of the inner header to 896 increment across encapsulation/decapsulation cycle. This check is 897 also performed when doing initial encapsulation when a packet 898 comes to an ITR or PITR destined for a LISP site. 900 o The inner header Type of Service field (or the Traffic Class 901 field, in the case of IPv6) SHOULD be copied from the outer header 902 Type of Service field (with one caveat, see below). 904 Note if an ETR/PETR is also an ITR/PITR and choose to reencapsulate 905 after decapsulating, the net effect of this is that the new outer 906 header will carry the same Time to Live as the old outer header minus 907 1. 909 Copying the TTL serves two purposes: first, it preserves the distance 910 the host intended the packet to travel; second, and more importantly, 911 it provides for suppression of looping packets in the event there is 912 a loop of concatenated tunnels due to misconfiguration. See 913 Section 9.3 for TTL exception handling for traceroute packets. 915 The ECN field occupies bits 6 and 7 of both the IPv4 Type of Service 916 field and the IPv6 Traffic Class field [RFC3168]. The ECN field 917 requires special treatment in order to avoid discarding indications 918 of congestion [RFC3168]. ITR encapsulation MUST copy the 2-bit ECN 919 field from the inner header to the outer header. Re-encapsulation 920 MUST copy the 2-bit ECN field from the stripped outer header to the 921 new outer header. If the ECN field contains a congestion indication 922 codepoint (the value is '11', the Congestion Experienced (CE) 923 codepoint), then ETR decapsulation MUST copy the 2-bit ECN field from 924 the stripped outer header to the surviving inner header that is used 925 to forward the packet beyond the ETR. These requirements preserve 926 Congestion Experienced (CE) indications when a packet that uses ECN 927 traverses a LISP tunnel and becomes marked with a CE indication due 928 to congestion between the tunnel endpoints. 930 5.4. Dealing with Large Encapsulated Packets 932 This section proposes two mechanisms to deal with packets that exceed 933 the path MTU between the ITR and ETR. 935 It is left to the implementor to decide if the stateless or stateful 936 mechanism should be implemented. Both or neither can be used since 937 it is a local decision in the ITR regarding how to deal with MTU 938 issues, and sites can interoperate with differing mechanisms. 940 Both stateless and stateful mechanisms also apply to Reencapsulating 941 and Recursive Tunneling. So any actions below referring to an ITR 942 also apply to an TE-ITR. 944 5.4.1. A Stateless Solution to MTU Handling 946 An ITR stateless solution to handle MTU issues is described as 947 follows: 949 1. Define an architectural constant S for the maximum size of a 950 packet, in bytes, an ITR would like to receive from a source 951 inside of its site. 953 2. Define L to be the maximum size, in bytes, a packet of size S 954 would be after the ITR prepends the LISP header, UDP header, and 955 outer network layer header of size H. Therefore, S + H = L. 957 When an ITR receives a packet from a site-facing interface and adds H 958 bytes worth of encapsulation to yield a packet size greater than L 959 bytes, it resolves the MTU issue by first splitting the original 960 packet into 2 equal-sized fragments. A LISP header is then prepended 961 to each fragment. The size of the encapsulated fragments is then 962 (S/2 + H), which is less than the ITR's estimate of the path MTU 963 between the ITR and its correspondent ETR. 965 When an ETR receives encapsulated fragments, it treats them as two 966 individually encapsulated packets. It strips the LISP headers then 967 forwards each fragment to the destination host of the destination 968 site. The two fragments are reassembled at the destination host into 969 the single IP datagram that was originated by the source host. 971 This behavior is performed by the ITR when the source host originates 972 a packet with the DF field of the IP header is set to 0. When the DF 973 field of the IP header is set to 1, or the packet is an IPv6 packet 974 originated by the source host, the ITR will drop the packet when the 975 size is greater than L, and sends an ICMP Too Big message to the 976 source with a value of S, where S is (L - H). 978 When the outer header encapsulation uses an IPv4 header, an 979 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 980 can be avoided. An implementation MAY set the DF bit in such headers 981 to 0 if it has good reason to believe there are unresolvable path MTU 982 issues between the sending ITR and the receiving ETR. 984 This specification RECOMMENDS that L be defined as 1500. 986 5.4.2. A Stateful Solution to MTU Handling 988 An ITR stateful solution to handle MTU issues is described as follows 989 and was first introduced in [OPENLISP]: 991 1. The ITR will keep state of the effective MTU for each locator per 992 mapping cache entry. The effective MTU is what the core network 993 can deliver along the path between ITR and ETR. 995 2. When an IPv6 encapsulated packet or an IPv4 encapsulated packet 996 with DF bit set to 1, exceeds what the core network can deliver, 997 one of the intermediate routers on the path will send an ICMP Too 998 Big message to the ITR. The ITR will parse the ICMP message to 999 determine which locator is affected by the effective MTU change 1000 and then record the new effective MTU value in the mapping cache 1001 entry. 1003 3. When a packet is received by the ITR from a source inside of the 1004 site and the size of the packet is greater than the effective MTU 1005 stored with the mapping cache entry associated with the 1006 destination EID the packet is for, the ITR will send an ICMP Too 1007 Big message back to the source. The packet size advertised by 1008 the ITR in the ICMP Too Big message is the effective MTU minus 1009 the LISP encapsulation length. 1011 Even though this mechanism is stateful, it has advantages over the 1012 stateless IP fragmentation mechanism, by not involving the 1013 destination host with reassembly of ITR fragmented packets. 1015 5.5. Using Virtualization and Segmentation with LISP 1017 When multiple organizations inside of a LISP site are using private 1018 addresses [RFC1918] as EID-prefixes, their address spaces MUST remain 1019 segregated due to possible address duplication. An Instance ID in 1020 the address encoding can aid in making the entire AFI based address 1021 unique. See IANA Considerations Section 14.2 for details for 1022 possible address encodings. 1024 An Instance ID can be carried in a LISP encapsulated packet. An ITR 1025 that prepends a LISP header, will copy a 24-bit value, used by the 1026 LISP router to uniquely identify the address space. The value is 1027 copied to the Instance ID field of the LISP header and the I-bit is 1028 set to 1. 1030 When an ETR decapsulates a packet, the Instance ID from the LISP 1031 header is used as a table identifier to locate the forwarding table 1032 to use for the inner destination EID lookup. 1034 For example, a 802.1Q VLAN tag or VPN identifier could be used as a 1035 24-bit Instance ID. 1037 6. EID-to-RLOC Mapping 1039 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats 1041 The following UDP packet formats are used by the LISP control-plane. 1043 0 1 2 3 1044 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 1045 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1046 |Version| IHL |Type of Service| Total Length | 1047 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1048 | Identification |Flags| Fragment Offset | 1049 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1050 | Time to Live | Protocol = 17 | Header Checksum | 1051 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1052 | Source Routing Locator | 1053 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1054 | Destination Routing Locator | 1055 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1056 / | Source Port | Dest Port | 1057 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1058 \ | UDP Length | UDP Checksum | 1059 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1060 | | 1061 | LISP Message | 1062 | | 1063 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1065 0 1 2 3 1066 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 1067 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1068 |Version| Traffic Class | Flow Label | 1069 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1070 | Payload Length | Next Header=17| Hop Limit | 1071 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1072 | | 1073 + + 1074 | | 1075 + Source Routing Locator + 1076 | | 1077 + + 1078 | | 1079 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1080 | | 1081 + + 1082 | | 1083 + Destination Routing Locator + 1084 | | 1085 + + 1086 | | 1087 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1088 / | Source Port | Dest Port | 1089 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1090 \ | UDP Length | UDP Checksum | 1091 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1092 | | 1093 | LISP Message | 1094 | | 1095 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1097 The LISP UDP-based messages are the Map-Request and Map-Reply 1098 messages. When a UDP Map-Request is sent, the UDP source port is 1099 chosen by the sender and the destination UDP port number is set to 1100 4342. When a UDP Map-Reply is sent, the source UDP port number is 1101 set to 4342 and the destination UDP port number is copied from the 1102 source port of either the Map-Request or the invoking data packet. 1103 Implementations MUST be prepared to accept packets when either the 1104 source port or destination UDP port is set to 4342 due to NATs 1105 changing port number values. 1107 The UDP Length field will reflect the length of the UDP header and 1108 the LISP Message payload. 1110 The UDP Checksum is computed and set to non-zero for Map-Request, 1111 Map-Reply, Map-Register and ECM control messages. It MUST be checked 1112 on receipt and if the checksum fails, the packet MUST be dropped. 1114 The format of control messages includes the UDP header so the 1115 checksum and length fields can be used to protect and delimit message 1116 boundaries. 1118 This main LISP specification is the authoritative source for message 1119 format definitions for the Map-Request and Map-Reply messages. 1121 6.1.1. LISP Packet Type Allocations 1123 This section will be the authoritative source for allocating LISP 1124 Type values. Current allocations are: 1126 Reserved: 0 b'0000' 1127 LISP Map-Request: 1 b'0001' 1128 LISP Map-Reply: 2 b'0010' 1129 LISP Map-Register: 3 b'0011' 1130 LISP Map-Notify: 4 b'0100' 1131 LISP Encapsulated Control Message: 8 b'1000' 1133 6.1.2. Map-Request Message Format 1135 0 1 2 3 1136 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 1137 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1138 |Type=1 |A|M|P|S|p|s| Reserved | IRC | Record Count | 1139 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1140 | Nonce . . . | 1141 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1142 | . . . Nonce | 1143 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1144 | Source-EID-AFI | Source EID Address ... | 1145 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1146 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... | 1147 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1148 | ... | 1149 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1150 | ITR-RLOC-AFI n | ITR-RLOC Address n ... | 1151 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1152 / | Reserved | EID mask-len | EID-prefix-AFI | 1153 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1154 \ | EID-prefix ... | 1155 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1156 | Map-Reply Record ... | 1157 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1158 | Mapping Protocol Data | 1159 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1161 Packet field descriptions: 1163 Type: 1 (Map-Request) 1165 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 1166 Requests sent by an ITR. Set to 1 when an ITR wants the 1167 destination site to return the Map-Reply rather than the mapping 1168 database system. 1170 M: This is the map-data-present bit, when set, it indicates a Map- 1171 Reply Record segment is included in the Map-Request. 1173 P: This is the probe-bit which indicates that a Map-Request SHOULD be 1174 treated as a locator reachability probe. The receiver SHOULD 1175 respond with a Map-Reply with the probe-bit set, indicating the 1176 Map-Reply is a locator reachability probe reply, with the nonce 1177 copied from the Map-Request. See Section 6.3.2 for more details. 1179 S: This is the Solicit-Map-Request (SMR) bit. See Section 6.6.2 for 1180 details. 1182 p: This is the PITR bit. This bit is set to 1 when a PITR sends a 1183 Map-Request. 1185 s: This is the SMR-invoked bit. This bit is set to 1 when an xTR is 1186 sending a Map-Request in response to a received SMR-based Map- 1187 Request. 1189 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1190 receipt. 1192 IRC: This 5-bit field is the ITR-RLOC Count which encodes the 1193 additional number of (ITR-RLOC-AFI, ITR-RLOC Address) fields 1194 present in this message. At least one (ITR-RLOC-AFI, ITR-RLOC- 1195 Address) pair MUST be encoded. Multiple ITR-RLOC Address fields 1196 are used so a Map-Replier can select which destination address to 1197 use for a Map-Reply. The IRC value ranges from 0 to 31. For a 1198 value of 0, there is 1 ITR-RLOC address encoded, and for a value 1199 of 1, there are 2 ITR-RLOC addresses encoded and so on up to 31 1200 which encodes a total of 32 ITR-RLOC addresses. 1202 Record Count: The number of records in this Map-Request message. A 1203 record is comprised of the portion of the packet that is labeled 1204 'Rec' above and occurs the number of times equal to Record Count. 1205 For this version of the protocol, a receiver MUST accept and 1206 process Map-Requests that contain one or more records, but a 1207 sender MUST only send Map-Requests containing one record. Support 1208 for requesting multiple EIDs in a single Map-Request message will 1209 be specified in a future version of the protocol. 1211 Nonce: An 8-byte random value created by the sender of the Map- 1212 Request. This nonce will be returned in the Map-Reply. The 1213 security of the LISP mapping protocol depends critically on the 1214 strength of the nonce in the Map-Request message. The nonce 1215 SHOULD be generated by a properly seeded pseudo-random (or strong 1216 random) source. See [RFC4086] for advice on generating security- 1217 sensitive random data. 1219 Source-EID-AFI: Address family of the "Source EID Address" field. 1221 Source EID Address: This is the EID of the source host which 1222 originated the packet which is caused the Map-Request. When Map- 1223 Requests are used for refreshing a map-cache entry or for RLOC- 1224 probing, an AFI value 0 is used and this field is of zero length. 1226 ITR-RLOC-AFI: Address family of the "ITR-RLOC Address" field that 1227 follows this field. 1229 ITR-RLOC Address: Used to give the ETR the option of selecting the 1230 destination address from any address family for the Map-Reply 1231 message. This address MUST be a routable RLOC address of the 1232 sender of the Map-Request message. 1234 EID mask-len: Mask length for EID prefix. 1236 EID-prefix-AFI: Address family of EID-prefix according to [AFI] 1238 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1239 address-family. When a Map-Request is sent by an ITR because a 1240 data packet is received for a destination where there is no 1241 mapping entry, the EID-prefix is set to the destination IP address 1242 of the data packet. And the 'EID mask-len' is set to 32 or 128 1243 for IPv4 or IPv6, respectively. When an xTR wants to query a site 1244 about the status of a mapping it already has cached, the EID- 1245 prefix used in the Map-Request has the same mask-length as the 1246 EID-prefix returned from the site when it sent a Map-Reply 1247 message. 1249 Map-Reply Record: When the M bit is set, this field is the size of a 1250 single "Record" in the Map-Reply format. This Map-Reply record 1251 contains the EID-to-RLOC mapping entry associated with the Source 1252 EID. This allows the ETR which will receive this Map-Request to 1253 cache the data if it chooses to do so. 1255 Mapping Protocol Data: This field is optional and present when the 1256 UDP length indicates there is enough space in the packet to 1257 include it. 1259 6.1.3. EID-to-RLOC UDP Map-Request Message 1261 A Map-Request is sent from an ITR when it needs a mapping for an EID, 1262 wants to test an RLOC for reachability, or wants to refresh a mapping 1263 before TTL expiration. For the initial case, the destination IP 1264 address used for the Map-Request is the destination-EID from the 1265 packet which had a mapping cache lookup failure. For the latter two 1266 cases, the destination IP address used for the Map-Request is one of 1267 the RLOC addresses from the locator-set of the map cache entry. The 1268 source address is either an IPv4 or IPv6 RLOC address depending if 1269 the Map-Request is using an IPv4 versus IPv6 header, respectively. 1270 In all cases, the UDP source port number for the Map-Request message 1271 is an ITR/PITR selected 16-bit value and the UDP destination port 1272 number is set to the well-known destination port number 4342. A 1273 successful Map-Reply, which is one that has a nonce that matches an 1274 outstanding Map-Request nonce, will update the cached set of RLOCs 1275 associated with the EID prefix range. 1277 One or more Map-Request (ITR-RLOC-AFI, ITR-RLOC-Address) fields MUST 1278 be filled in by the ITR. The number of fields (minus 1) encoded MUST 1279 be placed in the IRC field. The ITR MAY include all locally 1280 configured locators in this list or just provide one locator address 1281 from each address family it supports. If the ITR erroneously 1282 provides no ITR-RLOC addresses, the Map-Replier MUST drop the Map- 1283 Request. 1285 Map-Requests can also be LISP encapsulated using UDP destination port 1286 4342 with a LISP type value set to "Encapsulated Control Message", 1287 when sent from an ITR to a Map-Resolver. Likewise, Map-Requests are 1288 LISP encapsulated the same way from a Map-Server to an ETR. Details 1289 on encapsulated Map-Requests and Map-Resolvers can be found in 1290 [LISP-MS]. 1292 Map-Requests MUST be rate-limited. It is RECOMMENDED that a Map- 1293 Request for the same EID-prefix be sent no more than once per second. 1295 An ITR that is configured with mapping database information (i.e. it 1296 is also an ETR) MAY optionally include those mappings in a Map- 1297 Request. When an ETR configured to accept and verify such 1298 "piggybacked" mapping data receives such a Map-Request and it does 1299 not have this mapping in the map-cache, it may originate a "verifying 1300 Map-Request", addressed to the map-requesting ITR. If the ETR has a 1301 map-cache entry that matches the "piggybacked" EID and the RLOC is in 1302 the locator-set for the entry, then it may send the "verifying Map- 1303 Request" directly to the originating Map-Request source. If the RLOC 1304 is not in the locator-set, then the ETR MUST send the "verifying Map- 1305 Request" to the "piggybacked" EID. Doing this forces the "verifying 1306 Map-Request" to go through the mapping database system to reach the 1307 authoritative source of information about that EID, guarding against 1308 RLOC-spoofing in in the "piggybacked" mapping data. 1310 6.1.4. Map-Reply Message Format 1312 0 1 2 3 1313 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 1314 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1315 |Type=2 |P|E|S| Reserved | Record Count | 1316 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1317 | Nonce . . . | 1318 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1319 | . . . Nonce | 1320 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1321 | | Record TTL | 1322 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1323 R | Locator Count | EID mask-len | ACT |A| Reserved | 1324 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1325 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1326 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1327 r | EID-prefix | 1328 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1329 | /| Priority | Weight | M Priority | M Weight | 1330 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1331 | o | Unused Flags |L|p|R| Loc-AFI | 1332 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1333 | \| Locator | 1334 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1335 | Mapping Protocol Data | 1336 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1338 Packet field descriptions: 1340 Type: 2 (Map-Reply) 1342 P: This is the probe-bit which indicates that the Map-Reply is in 1343 response to a locator reachability probe Map-Request. The nonce 1344 field MUST contain a copy of the nonce value from the original 1345 Map-Request. See Section 6.3.2 for more details. 1347 E: Indicates that the ETR which sends this Map-Reply message is 1348 advertising that the site is enabled for the Echo-Nonce locator 1349 reachability algorithm. See Section 6.3.1 for more details. 1351 S: This is the Security bit. When set to 1 the field following the 1352 Mapping Protocol Data field will have the following format. The 1353 detailed format of the Authentication Data Content is for further 1354 study. 1356 0 1 2 3 1357 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 1358 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1359 | AD Type | Authentication Data Content . . . | 1360 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1362 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1363 receipt. 1365 Record Count: The number of records in this reply message. A record 1366 is comprised of that portion of the packet labeled 'Record' above 1367 and occurs the number of times equal to Record count. 1369 Nonce: A 24-bit value set in a Data-Probe packet or a 64-bit value 1370 from the Map-Request is echoed in this Nonce field of the Map- 1371 Reply. When a 24-bit value is supplied, it resides in the low- 1372 order 64 bits of the nonce field. 1374 Record TTL: The time in minutes the recipient of the Map-Reply will 1375 store the mapping. If the TTL is 0, the entry SHOULD be removed 1376 from the cache immediately. If the value is 0xffffffff, the 1377 recipient can decide locally how long to store the mapping. 1379 Locator Count: The number of Locator entries. A locator entry 1380 comprises what is labeled above as 'Loc'. The locator count can 1381 be 0 indicating there are no locators for the EID-prefix. 1383 EID mask-len: Mask length for EID prefix. 1385 ACT: This 3-bit field describes negative Map-Reply actions. In any 1386 other message type, these bits are set to 0 and ignored on 1387 receipt. These bits are used only when the 'Locator Count' field 1388 is set to 0. The action bits are encoded only in Map-Reply 1389 messages. The actions defined are used by an ITR or PITR when a 1390 destination EID matches a negative mapping cache entry. 1391 Unassigned values should cause a map-cache entry to be created 1392 and, when packets match this negative cache entry, they will be 1393 dropped. The current assigned values are: 1395 (0) No-Action: The map-cache is kept alive and no packet 1396 encapsulation occurs. 1398 (1) Natively-Forward: The packet is not encapsulated or dropped 1399 but natively forwarded. 1401 (2) Send-Map-Request: The packet invokes sending a Map-Request. 1403 (3) Drop: A packet that matches this map-cache entry is dropped. 1404 An ICMP Unreachable message SHOULD be sent. 1406 A: The Authoritative bit, when sent is always set to 1 by an ETR. 1407 When a Map-Server is proxy Map-Replying [LISP-MS] for a LISP site, 1408 the Authoritative bit is set to 0. This indicates to requesting 1409 ITRs that the Map-Reply was not originated by a LISP node managed 1410 at the site that owns the EID-prefix. 1412 Map-Version Number: When this 12-bit value is non-zero the Map-Reply 1413 sender is informing the ITR what the version number is for the 1414 EID-record contained in the Map-Reply. The ETR can allocate this 1415 number internally but MUST coordinate this value with other ETRs 1416 for the site. When this value is 0, there is no versioning 1417 information conveyed. The Map-Version Number can be included in 1418 Map-Request and Map-Register messages. See Section 6.6.3 for more 1419 details. 1421 EID-prefix-AFI: Address family of EID-prefix according to [AFI]. 1423 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1424 address-family. 1426 Priority: each RLOC is assigned a unicast priority. Lower values 1427 are more preferable. When multiple RLOCs have the same priority, 1428 they MAY be used in a load-split fashion. A value of 255 means 1429 the RLOC MUST NOT be used for unicast forwarding. 1431 Weight: when priorities are the same for multiple RLOCs, the weight 1432 indicates how to balance unicast traffic between them. Weight is 1433 encoded as a relative weight of total unicast packets that match 1434 the mapping entry. For example if there are 4 locators in a 1435 locator set, where the weights assigned are 30, 20, 20, and 10, 1436 the first locator will get 37.5% of the traffic, the 2nd and 3rd 1437 locators will get 25% of traffic and the 4th locator will get 1438 12.5% of the traffic. If all weights for a locator-set are equal, 1439 receiver of the Map-Reply will decide how to load-split traffic. 1440 See Section 6.5 for a suggested hash algorithm to distribute load 1441 across locators with same priority and equal weight values. 1443 M Priority: each RLOC is assigned a multicast priority used by an 1444 ETR in a receiver multicast site to select an ITR in a source 1445 multicast site for building multicast distribution trees. A value 1446 of 255 means the RLOC MUST NOT be used for joining a multicast 1447 distribution tree. 1449 M Weight: when priorities are the same for multiple RLOCs, the 1450 weight indicates how to balance building multicast distribution 1451 trees across multiple ITRs. The weight is encoded as a relative 1452 weight (similar to the unicast Weights) of total number of trees 1453 built to the source site identified by the EID-prefix. If all 1454 weights for a locator-set are equal, the receiver of the Map-Reply 1455 will decide how to distribute multicast state across ITRs. 1457 Unused Flags: set to 0 when sending and ignored on receipt. 1459 L: when this bit is set, the locator is flagged as a local locator to 1460 the ETR that is sending the Map-Reply. When a Map-Server is doing 1461 proxy Map-Replying [LISP-MS] for a LISP site, the L bit is set to 1462 0 for all locators in this locator-set. 1464 p: when this bit is set, an ETR informs the RLOC-probing ITR that the 1465 locator address, for which this bit is set, is the one being RLOC- 1466 probed and MAY be different from the source address of the Map- 1467 Reply. An ITR that RLOC-probes a particular locator, MUST use 1468 this locator for retrieving the data structure used to store the 1469 fact that the locator is reachable. The "p" bit is set for a 1470 single locator in the same locator set. If an implementation sets 1471 more than one "p" bit erroneously, the receiver of the Map-Reply 1472 MUST select the first locator. The "p" bit MUST NOT be set for 1473 locator-set records sent in Map-Request and Map-Register messages. 1475 R: set when the sender of a Map-Reply has a route to the locator in 1476 the locator data record. This receiver may find this useful to 1477 know if the locator is up but not necessarily reachable from the 1478 receiver's point of view. See also Section 6.4 for another way 1479 the R-bit may be used. 1481 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1482 assigned to an ETR. Note that the destination RLOC address MAY be 1483 an anycast address. A source RLOC can be an anycast address as 1484 well. The source or destination RLOC MUST NOT be the broadcast 1485 address (255.255.255.255 or any subnet broadcast address known to 1486 the router), and MUST NOT be a link-local multicast address. The 1487 source RLOC MUST NOT be a multicast address. The destination RLOC 1488 SHOULD be a multicast address if it is being mapped from a 1489 multicast destination EID. 1491 Mapping Protocol Data: This field is optional and present when the 1492 UDP length indicates there is enough space in the packet to 1493 include it. 1495 6.1.5. EID-to-RLOC UDP Map-Reply Message 1497 A Map-Reply returns an EID-prefix with a prefix length that is less 1498 than or equal to the EID being requested. The EID being requested is 1499 either from the destination field of an IP header of a Data-Probe or 1500 the EID record of a Map-Request. The RLOCs in the Map-Reply are 1501 globally-routable IP addresses of all ETRs for the LISP site. Each 1502 RLOC conveys status reachability but does not convey path 1503 reachability from a requesters perspective. Separate testing of path 1504 reachability is required, See Section 6.3 for details. 1506 Note that a Map-Reply may contain different EID-prefix granularity 1507 (prefix + length) than the Map-Request which triggers it. This might 1508 occur if a Map-Request were for a prefix that had been returned by an 1509 earlier Map-Reply. In such a case, the requester updates its cache 1510 with the new prefix information and granularity. For example, a 1511 requester with two cached EID-prefixes that are covered by a Map- 1512 Reply containing one, less-specific prefix, replaces the entry with 1513 the less-specific EID-prefix. Note that the reverse, replacement of 1514 one less-specific prefix with multiple more-specific prefixes, can 1515 also occur but not by removing the less-specific prefix rather by 1516 adding the more-specific prefixes which during a lookup will override 1517 the less-specific prefix. 1519 When an ETR is configured with overlapping EID-prefixes, a Map- 1520 Request with an EID that longest matches any EID-prefix MUST be 1521 returned in a single Map-Reply message. For instance, if an ETR had 1522 database mapping entries for EID-prefixes: 1524 10.0.0.0/8 1525 10.1.0.0/16 1526 10.1.1.0/24 1527 10.1.2.0/24 1529 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record 1530 count of 1 to be returned with a mapping record EID-prefix of 1531 10.1.1.0/24. 1533 A Map-Request for EID 10.1.5.5, would cause a Map-Reply with a record 1534 count of 3 to be returned with mapping records for EID-prefixes 1535 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24. 1537 Note that not all overlapping EID-prefixes need to be returned, only 1538 the more specifics (note in the second example above 10.0.0.0/8 was 1539 not returned for requesting EID 10.1.5.5) entries for the matching 1540 EID-prefix of the requesting EID. When more than one EID-prefix is 1541 returned, all SHOULD use the same Time-to-Live value so they can all 1542 time out at the same time. When a more specific EID-prefix is 1543 received later, its Time-to-Live value in the Map-Reply record can be 1544 stored even when other less specifics exist. When a less specific 1545 EID-prefix is received later, its map-cache expiration time SHOULD be 1546 set to the minimum expiration time of any more specific EID-prefix in 1547 the map-cache. This is done so the integrity of the EID-prefix set 1548 is wholly maintained so no more-specific entries are removed from the 1549 map-cache while keeping less-specific entries. 1551 Map-Replies SHOULD be sent for an EID-prefix no more often than once 1552 per second to the same requesting router. For scalability, it is 1553 expected that aggregation of EID addresses into EID-prefixes will 1554 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1555 prefix range thereby reducing the number of Map-Request messages. 1557 Map-Reply records can have an empty locator-set. A negative Map- 1558 Reply is a Map-Reply with an empty locator-set. Negative Map-Replies 1559 convey special actions by the sender to the ITR or PITR which have 1560 solicited the Map-Reply. There are two primary applications for 1561 Negative Map-Replies. The first is for a Map-Resolver to instruct an 1562 ITR or PITR when a destination is for a LISP site versus a non-LISP 1563 site. And the other is to source quench Map-Requests which are sent 1564 for non-allocated EIDs. 1566 For each Map-Reply record, the list of locators in a locator-set MUST 1567 appear in the same order for each ETR that originates a Map-Reply 1568 message. The locator-set MUST be sorted in order of ascending IP 1569 address where an IPv4 locator address is considered numerically 'less 1570 than' an IPv6 locator address. 1572 When sending a Map-Reply message, the destination address is copied 1573 from the one of the ITR-RLOC fields from the Map-Request. The ETR 1574 can choose a locator address from one of the address families it 1575 supports. For Data-Probes, the destination address of the Map-Reply 1576 is copied from the source address of the Data-Probe message which is 1577 invoking the reply. The source address of the Map-Reply is one of 1578 the local IP addresses chosen to allow uRPF checks to succeed in the 1579 upstream service provider. The destination port of a Map-Reply 1580 message is copied from the source port of the Map-Request or Data- 1581 Probe and the source port of the Map-Reply message is set to the 1582 well-known UDP port 4342. 1584 6.1.5.1. Traffic Redirection with Coarse EID-Prefixes 1586 When an ETR is misconfigured or compromised, it could return coarse 1587 EID-prefixes in Map-Reply messages it sends. The EID-prefix could 1588 cover EID-prefixes which are allocated to other sites redirecting 1589 their traffic to the locators of the compromised site. 1591 To solve this problem, there are two basic solutions that could be 1592 used. The first is to have Map-Servers proxy-map-reply on behalf of 1593 ETRs so their registered EID-prefixes are the ones returned in Map- 1594 Replies. Since the interaction between an ETR and Map-Server is 1595 secured with shared-keys, it is easier for an ETR to detect 1596 misbehavior. The second solution is to have ITRs and PITRs cache 1597 EID-prefixes with mask-lengths that are greater than or equal to a 1598 configured prefix length. This limits the damage to a specific width 1599 of any EID-prefix advertised, but needs to be coordinated with the 1600 allocation of site prefixes. These solutions can be used 1601 independently or at the same time. 1603 At the time of this writing, other approaches are being considered 1604 and researched. 1606 6.1.6. Map-Register Message Format 1608 The usage details of the Map-Register message can be found in 1609 specification [LISP-MS]. This section solely defines the message 1610 format. 1612 The message is sent in UDP with a destination UDP port of 4342 and a 1613 randomly selected UDP source port number. 1615 The Map-Register message format is: 1617 0 1 2 3 1618 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 1619 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1620 |Type=3 |P| Reserved |M| Record Count | 1621 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1622 | Nonce . . . | 1623 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1624 | . . . Nonce | 1625 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1626 | Key ID | Authentication Data Length | 1627 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1628 ~ Authentication Data ~ 1629 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1630 | | Record TTL | 1631 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1632 R | Locator Count | EID mask-len | ACT |A| Reserved | 1633 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1634 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1635 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1636 r | EID-prefix | 1637 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1638 | /| Priority | Weight | M Priority | M Weight | 1639 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1640 | o | Unused Flags |L|p|R| Loc-AFI | 1641 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1642 | \| Locator | 1643 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1645 Packet field descriptions: 1647 Type: 3 (Map-Register) 1649 P: This is the proxy-map-reply bit, when set to 1 an ETR sends a Map- 1650 Register message requesting for the Map-Server to proxy Map-Reply. 1651 The Map-Server will send non-authoritative Map-Replies on behalf 1652 of the ETR. Details on this usage can be found in [LISP-MS]. 1654 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1655 receipt. 1657 M: This is the want-map-notify bit, when set to 1 an ETR is 1658 requesting for a Map-Notify message to be returned in response to 1659 sending a Map-Register message. The Map-Notify message sent by a 1660 Map-Server is used to an acknowledge receipt of a Map-Register 1661 message. 1663 Record Count: The number of records in this Map-Register message. A 1664 record is comprised of that portion of the packet labeled 'Record' 1665 above and occurs the number of times equal to Record count. 1667 Nonce: This 8-byte Nonce field is set to 0 in Map-Register messages. 1668 Since the Map-Register message is authenticated, the nonce field 1669 is not needed for any security function. 1671 Key ID: A configured ID to find the configured Message 1672 Authentication Code (MAC) algorithm and key value used for the 1673 authentication function. See Section 14.4 for codepoint 1674 assignments. 1676 Authentication Data Length: The length in bytes of the 1677 Authentication Data field that follows this field. The length of 1678 the Authentication Data field is dependent on the Message 1679 Authentication Code (MAC) algorithm used. The length field allows 1680 a device that doesn't know the MAC algorithm to correctly parse 1681 the packet. 1683 Authentication Data: The message digest used from the output of the 1684 Message Authentication Code (MAC) algorithm. The entire Map- 1685 Register payload is authenticated with this field preset to 0. 1686 After the MAC is computed, it is placed in this field. 1687 Implementations of this specification MUST include support for 1688 HMAC-SHA-1-96 [RFC2404] and support for HMAC-SHA-256-128 [RFC6234] 1689 is RECOMMENDED. 1691 The definition of the rest of the Map-Register can be found in the 1692 Map-Reply section. 1694 6.1.7. Map-Notify Message Format 1696 The usage details of the Map-Notify message can be found in 1697 specification [LISP-MS]. This section solely defines the message 1698 format. 1700 The message is sent inside a UDP packet with source and destination 1701 UDP ports equal to 4342. 1703 The Map-Notify message format is: 1705 0 1 2 3 1706 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 1707 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1708 |Type=4 | Reserved | Record Count | 1709 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1710 | Nonce . . . | 1711 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1712 | . . . Nonce | 1713 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1714 | Key ID | Authentication Data Length | 1715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1716 ~ Authentication Data ~ 1717 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1718 | | Record TTL | 1719 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1720 R | Locator Count | EID mask-len | ACT |A| Reserved | 1721 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1722 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1723 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1724 r | EID-prefix | 1725 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1726 | /| Priority | Weight | M Priority | M Weight | 1727 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1728 | o | Unused Flags |L|p|R| Loc-AFI | 1729 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1730 | \| Locator | 1731 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1733 Packet field descriptions: 1735 Type: 4 (Map-Notify) 1737 The Map-Notify message has the same contents as a Map-Register 1738 message. See Map-Register section for field descriptions. 1740 6.1.8. Encapsulated Control Message Format 1742 An Encapsulated Control Message (ECM) is used to encapsulate control 1743 packets sent between xTRs and the mapping database system described 1744 in [LISP-MS]. 1746 0 1 2 3 1747 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 1748 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1749 / | IPv4 or IPv6 Header | 1750 OH | (uses RLOC addresses) | 1751 \ | | 1752 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1753 / | Source Port = xxxx | Dest Port = 4342 | 1754 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1755 \ | UDP Length | UDP Checksum | 1756 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1757 LH |Type=8 |S| Reserved | 1758 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1759 / | IPv4 or IPv6 Header | 1760 IH | (uses RLOC or EID addresses) | 1761 \ | | 1762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1763 / | Source Port = xxxx | Dest Port = yyyy | 1764 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1765 \ | UDP Length | UDP Checksum | 1766 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1767 LCM | LISP Control Message | 1768 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1770 Packet header descriptions: 1772 OH: The outer IPv4 or IPv6 header which uses RLOC addresses in the 1773 source and destination header address fields. 1775 UDP: The outer UDP header with destination port 4342. The source 1776 port is randomly allocated. The checksum field MUST be non-zero. 1778 LH: Type 8 is defined to be a "LISP Encapsulated Control Message" 1779 and what follows is either an IPv4 or IPv6 header as encoded by 1780 the first 4 bits after the reserved field. 1782 S: This is the Security bit. When set to 1 the field following the 1783 Reserved field will have the following format. The detailed 1784 format of the Authentication Data Content is for further study. 1786 0 1 2 3 1787 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 1788 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1789 | AD Type | Authentication Data Content . . . | 1790 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1792 IH: The inner IPv4 or IPv6 header which can use either RLOC or EID 1793 addresses in the header address fields. When a Map-Request is 1794 encapsulated in this packet format the destination address in this 1795 header is an EID. 1797 UDP: The inner UDP header where the port assignments depends on the 1798 control packet being encapsulated. When the control packet is a 1799 Map-Request or Map-Register, the source port is ITR/PITR selected 1800 and the destination port is 4342. When the control packet is a 1801 Map-Reply, the source port is 4342 and the destination port is 1802 assigned from the source port of the invoking Map-Request. Port 1803 number 4341 MUST NOT be assigned to either port. The checksum 1804 field MUST be non-zero. 1806 LCM: The format is one of the control message formats described in 1807 this section. At this time, only Map-Request messages are allowed 1808 to be encapsulated. And in the future, PIM Join-Prune messages 1809 [MLISP] might be allowed. Encapsulating other types of LISP 1810 control messages are for further study. When Map-Requests are 1811 sent for RLOC-probing purposes (i.e the probe-bit is set), they 1812 MUST NOT be sent inside Encapsulated Control Messages. 1814 6.2. Routing Locator Selection 1816 Both client-side and server-side may need control over the selection 1817 of RLOCs for conversations between them. This control is achieved by 1818 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1819 messages. Alternatively, RLOC information MAY be gleaned from 1820 received tunneled packets or EID-to-RLOC Map-Request messages. 1822 The following enumerates different scenarios for choosing RLOCs and 1823 the controls that are available: 1825 o Server-side returns one RLOC. Client-side can only use one RLOC. 1826 Server-side has complete control of the selection. 1828 o Server-side returns a list of RLOC where a subset of the list has 1829 the same best priority. Client can only use the subset list 1830 according to the weighting assigned by the server-side. In this 1831 case, the server-side controls both the subset list and load- 1832 splitting across its members. The client-side can use RLOCs 1833 outside of the subset list if it determines that the subset list 1834 is unreachable (unless RLOCs are set to a Priority of 255). Some 1835 sharing of control exists: the server-side determines the 1836 destination RLOC list and load distribution while the client-side 1837 has the option of using alternatives to this list if RLOCs in the 1838 list are unreachable. 1840 o Server-side sets weight of 0 for the RLOC subset list. In this 1841 case, the client-side can choose how the traffic load is spread 1842 across the subset list. Control is shared by the server-side 1843 determining the list and the client determining load distribution. 1844 Again, the client can use alternative RLOCs if the server-provided 1845 list of RLOCs are unreachable. 1847 o Either side (more likely on the server-side ETR) decides not to 1848 send a Map-Request. For example, if the server-side ETR does not 1849 send Map-Requests, it gleans RLOCs from the client-side ITR, 1850 giving the client-side ITR responsibility for bidirectional RLOC 1851 reachability and preferability. Server-side ETR gleaning of the 1852 client-side ITR RLOC is done by caching the inner header source 1853 EID and the outer header source RLOC of received packets. The 1854 client-side ITR controls how traffic is returned and can alternate 1855 using an outer header source RLOC, which then can be added to the 1856 list the server-side ETR uses to return traffic. Since no 1857 Priority or Weights are provided using this method, the server- 1858 side ETR MUST assume each client-side ITR RLOC uses the same best 1859 Priority with a Weight of zero. In addition, since EID-prefix 1860 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1861 on tunnel routers can grow to be very large. 1863 o A "gleaned" map-cache entry, one learned from the source RLOC of a 1864 received encapsulated packet, is only stored and used for a few 1865 seconds, pending verification. Verification is performed by 1866 sending a Map-Request to the source EID (the inner header IP 1867 source address) of the received encapsulated packet. A reply to 1868 this "verifying Map-Request" is used to fully populate the map- 1869 cache entry for the "gleaned" EID and is stored and used for the 1870 time indicated from the TTL field of a received Map-Reply. When a 1871 verified map-cache entry is stored, data gleaning no longer occurs 1872 for subsequent packets which have a source EID that matches the 1873 EID-prefix of the verified entry. 1875 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1876 reachable when the R-bit for the locator record is set to 1. When 1877 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1878 RLOC. Neither the information contained in a Map-Reply or that 1879 stored in the mapping database system provides reachability 1880 information for RLOCs. Note that reachability is not part of the 1881 mapping system and is determined using one or more of the Routing 1882 Locator Reachability Algorithms described in the next section. 1884 6.3. Routing Locator Reachability 1886 Several mechanisms for determining RLOC reachability are currently 1887 defined: 1889 1. An ETR may examine the Locator Status Bits in the LISP header of 1890 an encapsulated data packet received from an ITR. If the ETR is 1891 also acting as an ITR and has traffic to return to the original 1892 ITR site, it can use this status information to help select an 1893 RLOC. 1895 2. An ITR may receive an ICMP Network or ICMP Host Unreachable 1896 message for an RLOC it is using. This indicates that the RLOC is 1897 likely down. Note, trusting ICMP messages may not be desirable 1898 but neither is ignoring them completely. Implementations are 1899 encouraged to follow current best practices in treating these 1900 conditions. 1902 3. An ITR which participates in the global routing system can 1903 determine that an RLOC is down if no BGP RIB route exists that 1904 matches the RLOC IP address. 1906 4. An ITR may receive an ICMP Port Unreachable message from a 1907 destination host. This occurs if an ITR attempts to use 1908 interworking [INTERWORK] and LISP-encapsulated data is sent to a 1909 non-LISP-capable site. 1911 5. An ITR may receive a Map-Reply from an ETR in response to a 1912 previously sent Map-Request. The RLOC source of the Map-Reply is 1913 likely up since the ETR was able to send the Map-Reply to the 1914 ITR. 1916 6. When an ETR receives an encapsulated packet from an ITR, the 1917 source RLOC from the outer header of the packet is likely up. 1919 7. An ITR/ETR pair can use the Locator Reachability Algorithms 1920 described in this section, namely Echo-Noncing or RLOC-Probing. 1922 When determining Locator up/down reachability by examining the 1923 Locator Status Bits from the LISP encapsulated data packet, an ETR 1924 will receive up to date status from an encapsulating ITR about 1925 reachability for all ETRs at the site. CE-based ITRs at the source 1926 site can determine reachability relative to each other using the site 1927 IGP as follows: 1929 o Under normal circumstances, each ITR will advertise a default 1930 route into the site IGP. 1932 o If an ITR fails or if the upstream link to its PE fails, its 1933 default route will either time-out or be withdrawn. 1935 Each ITR can thus observe the presence or lack of a default route 1936 originated by the others to determine the Locator Status Bits it sets 1937 for them. 1939 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1940 Locator Status Bits in a LISP encapsulated packet are numbered from 0 1941 to n-1 starting with the least significant bit. For example, if an 1942 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1943 value 2), then all ITRs at the site will clear the 3rd least 1944 significant bit (xxxx x0xx) of the Locator Status Bits field for the 1945 packets they encapsulate. 1947 When an ETR decapsulates a packet, it will check for any change in 1948 the Locator Status Bits field. When a bit goes from 1 to 0, the ETR 1949 if acting also as an ITR, will refrain from encapsulating packets to 1950 an RLOC that is indicated as down. It will only resume using that 1951 RLOC if the corresponding Locator Status Bit returns to a value of 1. 1952 Locator Status Bits are associated with a locator-set per EID-prefix. 1953 Therefore, when a locator becomes unreachable, the Locator Status Bit 1954 that corresponds to that locator's position in the list returned by 1955 the last Map-Reply will be set to zero for that particular EID- 1956 prefix. 1958 When ITRs at the site are not deployed in CE routers, the IGP can 1959 still be used to determine the reachability of Locators provided they 1960 are injected into the IGP. This is typically done when a /32 address 1961 is configured on a loopback interface. 1963 When ITRs receive ICMP Network or Host Unreachable messages as a 1964 method to determine unreachability, they will refrain from using 1965 Locators which are described in Locator lists of Map-Replies. 1966 However, using this approach is unreliable because many network 1967 operators turn off generation of ICMP Unreachable messages. 1969 If an ITR does receive an ICMP Network or Host Unreachable message, 1970 it MAY originate its own ICMP Unreachable message destined for the 1971 host that originated the data packet the ITR encapsulated. 1973 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1974 locator address from a locator-set in a mapping entry matches a 1975 prefix. If it does not find one and BGP is running in the Default 1976 Free Zone (DFZ), it can decide to not use the locator even though the 1977 Locator Status Bits indicate the locator is up. In this case, the 1978 path from the ITR to the ETR that is assigned the locator is not 1979 available. More details are in [LOC-ID-ARCH]. 1981 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 1982 Reply is returned, reachability of the Locator has been determined. 1983 Obviously, sending such probes increases the number of control 1984 messages originated by tunnel routers for active flows, so Locators 1985 are assumed to be reachable when they are advertised. 1987 This assumption does create a dependency: Locator unreachability is 1988 detected by the receipt of ICMP Host Unreachable messages. When an 1989 Locator has been determined to be unreachable, it is not used for 1990 active traffic; this is the same as if it were listed in a Map-Reply 1991 with priority 255. 1993 The ITR can test the reachability of the unreachable Locator by 1994 sending periodic Requests. Both Requests and Replies MUST be rate- 1995 limited. Locator reachability testing is never done with data 1996 packets since that increases the risk of packet loss for end-to-end 1997 sessions. 1999 When an ETR decapsulates a packet, it knows that it is reachable from 2000 the encapsulating ITR because that is how the packet arrived. In 2001 most cases, the ETR can also reach the ITR but cannot assume this to 2002 be true due to the possibility of path asymmetry. In the presence of 2003 unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD NOT 2004 use the lack of return traffic as an indication that the ETR is 2005 unreachable. Instead, it MUST use an alternate mechanisms to 2006 determine reachability. 2008 6.3.1. Echo Nonce Algorithm 2010 When data flows bidirectionally between locators from different 2011 sites, a data-plane mechanism called "nonce echoing" can be used to 2012 determine reachability between an ITR and ETR. When an ITR wants to 2013 solicit a nonce echo, it sets the N and E bits and places a 24-bit 2014 nonce [RFC4086] in the LISP header of the next encapsulated data 2015 packet. 2017 When this packet is received by the ETR, the encapsulated packet is 2018 forwarded as normal. When the ETR next sends a data packet to the 2019 ITR, it includes the nonce received earlier with the N bit set and E 2020 bit cleared. The ITR sees this "echoed nonce" and knows the path to 2021 and from the ETR is up. 2023 The ITR will set the E-bit and N-bit for every packet it sends while 2024 in echo-nonce-request state. The time the ITR waits to process the 2025 echoed nonce before it determines the path is unreachable is variable 2026 and a choice left for the implementation. 2028 If the ITR is receiving packets from the ETR but does not see the 2029 nonce echoed while being in echo-nonce-request state, then the path 2030 to the ETR is unreachable. This decision may be overridden by other 2031 locator reachability algorithms. Once the ITR determines the path to 2032 the ETR is down it can switch to another locator for that EID-prefix. 2034 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 2035 be implementing both ITR and ETR functionality for the echo nonce 2036 mechanism to operate. 2038 The ITR and ETR may both go into echo-nonce-request state at the same 2039 time. The number of packets sent or the time during which echo nonce 2040 requests are sent is an implementation specific setting. However, 2041 when an ITR is in echo-nonce-request state, it can echo the ETR's 2042 nonce in the next set of packets that it encapsulates and then 2043 subsequently, continue sending echo-nonce-request packets. 2045 This mechanism does not completely solve the forward path 2046 reachability problem as traffic may be unidirectional. That is, the 2047 ETR receiving traffic at a site may not be the same device as an ITR 2048 which transmits traffic from that site or the site to site traffic is 2049 unidirectional so there is no ITR returning traffic. 2051 The echo-nonce algorithm is bilateral. That is, if one side sets the 2052 E-bit and the other side is not enabled for echo-noncing, then the 2053 echoing of the nonce does not occur and the requesting side may 2054 regard the locator unreachable erroneously. An ITR SHOULD only set 2055 the E-bit in a encapsulated data packet when it knows the ETR is 2056 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 2057 Reply message. 2059 Note that other locator reachability mechanisms are being researched 2060 and can be used to compliment or even override the Echo Nonce 2061 Algorithm. See next section for an example of control-plane probing. 2063 6.3.2. RLOC Probing Algorithm 2065 RLOC Probing is a method that an ITR or PITR can use to determine the 2066 reachability status of one or more locators that it has cached in a 2067 map-cache entry. The probe-bit of the Map-Request and Map-Reply 2068 messages are used for RLOC Probing. 2070 RLOC probing is done in the control-plane on a timer basis where an 2071 ITR or PITR will originate a Map-Request destined to a locator 2072 address from one of its own locator addresses. A Map-Request used as 2073 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or on 2074 the ALT like one would when soliciting mapping data. The EID record 2075 encoded in the Map-Request is the EID-prefix of the map-cache entry 2076 cached by the ITR or PITR. The ITR may include a mapping data record 2077 for its own database mapping information which contains the local 2078 EID-prefixes and RLOCs for its site. 2080 When an ETR receives a Map-Request message with the probe-bit set, it 2081 returns a Map-Reply with the probe-bit set. The source address of 2082 the Map-Reply is set according to the procedure described in 2083 Section 6.1.5. The Map-Reply SHOULD contain mapping data for the 2084 EID-prefix contained in the Map-Request. This provides the 2085 opportunity for the ITR or PITR, which sent the RLOC-probe to get 2086 mapping updates if there were changes to the ETR's database mapping 2087 entries. 2089 There are advantages and disadvantages of RLOC Probing. The greatest 2090 benefit of RLOC Probing is that it can handle many failure scenarios 2091 allowing the ITR to determine when the path to a specific locator is 2092 reachable or has become unreachable, thus providing a robust 2093 mechanism for switching to using another locator from the cached 2094 locator. RLOC Probing can also provide rough RTT estimates between a 2095 pair of locators which can be useful for network management purposes 2096 as well as for selecting low delay paths. The major disadvantage of 2097 RLOC Probing is in the number of control messages required and the 2098 amount of bandwidth used to obtain those benefits, especially if the 2099 requirement for failure detection times are very small. 2101 Continued research and testing will attempt to characterize the 2102 tradeoffs of failure detection times versus message overhead. 2104 6.4. EID Reachability within a LISP Site 2106 A site may be multihomed using two or more ETRs. The hosts and 2107 infrastructure within a site will be addressed using one or more EID 2108 prefixes that are mapped to the RLOCs of the relevant ETRs in the 2109 mapping system. One possible failure mode is for an ETR to lose 2110 reachability to one or more of the EID prefixes within its own site. 2111 When this occurs when the ETR sends Map-Replies, it can clear the 2112 R-bit associated with its own locator. And when the ETR is also an 2113 ITR, it can clear its locator-status-bit in the encapsulation data 2114 header. 2116 It is recognized there are no simple solutions to the site 2117 partitioning problem because it is hard to know which part of the 2118 EID-prefix range is partitioned. And which locators can reach any 2119 sub-ranges of the EID-prefixes. This problem is under investigation 2120 with the expectation that experiments will tell us more. Note, this 2121 is not a new problem introduced by the LISP architecture. The 2122 problem exists today when a multi-homed site uses BGP to advertise 2123 its reachability upstream. 2125 6.5. Routing Locator Hashing 2127 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 2128 a requesting ITR, the locator-set for the EID-prefix may contain 2129 different priority values for each locator address. When more than 2130 one best priority locator exists, the ITR can decide how to load 2131 share traffic against the corresponding locators. 2133 The following hash algorithm may be used by an ITR to select a 2134 locator for a packet destined to an EID for the EID-to-RLOC mapping: 2136 1. Either a source and destination address hash can be used or the 2137 traditional 5-tuple hash which includes the source and 2138 destination addresses, source and destination TCP, UDP, or SCTP 2139 port numbers and the IP protocol number field or IPv6 next- 2140 protocol fields of a packet a host originates from within a LISP 2141 site. When a packet is not a TCP, UDP, or SCTP packet, the 2142 source and destination addresses only from the header are used to 2143 compute the hash. 2145 2. Take the hash value and divide it by the number of locators 2146 stored in the locator-set for the EID-to-RLOC mapping. 2148 3. The remainder will yield a value of 0 to "number of locators 2149 minus 1". Use the remainder to select the locator in the 2150 locator-set. 2152 Note that when a packet is LISP encapsulated, the source port number 2153 in the outer UDP header needs to be set. Selecting a hashed value 2154 allows core routers which are attached to Link Aggregation Groups 2155 (LAGs) to load-split the encapsulated packets across member links of 2156 such LAGs. Otherwise, core routers would see a single flow, since 2157 packets have a source address of the ITR, for packets which are 2158 originated by different EIDs at the source site. A suggested setting 2159 for the source port number computed by an ITR is a 5-tuple hash 2160 function on the inner header, as described above. 2162 Many core router implementations use a 5-tuple hash to decide how to 2163 balance packet load across members of a LAG. The 5-tuple hash 2164 includes the source and destination addresses of the packet and the 2165 source and destination ports when the protocol number in the packet 2166 is TCP or UDP. For this reason, UDP encoding is used for LISP 2167 encapsulation. 2169 6.6. Changing the Contents of EID-to-RLOC Mappings 2171 Since the LISP architecture uses a caching scheme to retrieve and 2172 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 2173 date mapping is to re-request the mapping. However, the ITRs do not 2174 know when the mappings change and the ETRs do not keep track of which 2175 ITRs requested its mappings. For scalability reasons, we want to 2176 maintain this approach but need to provide a way for ETRs change 2177 their mappings and inform the sites that are currently communicating 2178 with the ETR site using such mappings. 2180 When adding a new locator record in lexicographic order to the end of 2181 a locator-set, it is easy to update mappings. We assume new mappings 2182 will maintain the same locator ordering as the old mapping but just 2183 have new locators appended to the end of the list. So some ITRs can 2184 have a new mapping while other ITRs have only an old mapping that is 2185 used until they time out. When an ITR has only an old mapping but 2186 detects bits set in the loc-status-bits that correspond to locators 2187 beyond the list it has cached, it simply ignores them. However, this 2188 can only happen for locator addresses that are lexicographically 2189 greater than the locator addresses in the existing locator-set. 2191 When a locator record is inserted in the middle of a locator-set, to 2192 maintain lexicographic order, the SMR procedure in Section 6.6.2 is 2193 used to inform ITRs and PITRs of the new locator-status-bit mappings. 2195 When a locator record is removed from a locator-set, ITRs that have 2196 the mapping cached will not use the removed locator because the xTRs 2197 will set the loc-status-bit to 0. So even if the locator is in the 2198 list, it will not be used. For new mapping requests, the xTRs can 2199 set the locator AFI to 0 (indicating an unspecified address), as well 2200 as setting the corresponding loc-status-bit to 0. This forces ITRs 2201 with old or new mappings to avoid using the removed locator. 2203 If many changes occur to a mapping over a long period of time, one 2204 will find empty record slots in the middle of the locator-set and new 2205 records appended to the locator-set. At some point, it would be 2206 useful to compact the locator-set so the loc-status-bit settings can 2207 be efficiently packed. 2209 We propose here three approaches for locator-set compaction, one 2210 operational and two protocol mechanisms. The operational approach 2211 uses a clock sweep method. The protocol approaches use the concept 2212 of Solicit-Map-Requests and Map-Versioning. 2214 6.6.1. Clock Sweep 2216 The clock sweep approach uses planning in advance and the use of 2217 count-down TTLs to time out mappings that have already been cached. 2218 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 2219 there is a 24 hour window to time out old mappings. The following 2220 clock sweep procedure is used: 2222 1. 24 hours before a mapping change is to take effect, a network 2223 administrator configures the ETRs at a site to start the clock 2224 sweep window. 2226 2. During the clock sweep window, ETRs continue to send Map-Reply 2227 messages with the current (unchanged) mapping records. The TTL 2228 for these mappings is set to 1 hour. 2230 3. 24 hours later, all previous cache entries will have timed out, 2231 and any active cache entries will time out within 1 hour. During 2232 this 1 hour window the ETRs continue to send Map-Reply messages 2233 with the current (unchanged) mapping records with the TTL set to 2234 1 minute. 2236 4. At the end of the 1 hour window, the ETRs will send Map-Reply 2237 messages with the new (changed) mapping records. So any active 2238 caches can get the new mapping contents right away if not cached, 2239 or in 1 minute if they had the mapping cached. The new mappings 2240 are cached with a time to live equal to the TTL in the Map-Reply. 2242 6.6.2. Solicit-Map-Request (SMR) 2244 Soliciting a Map-Request is a selective way for ETRs, at the site 2245 where mappings change, to control the rate they receive requests for 2246 Map-Reply messages. SMRs are also used to tell remote ITRs to update 2247 the mappings they have cached. 2249 Since the ETRs don't keep track of remote ITRs that have cached their 2250 mappings, they do not know which ITRs need to have their mappings 2251 updated. As a result, an ETR will solicit Map-Requests (called an 2252 SMR message) from those sites to which it has been sending 2253 encapsulated data to for the last minute. In particular, an ETR will 2254 send an SMR an ITR to which it has recently sent encapsulated data. 2256 An SMR message is simply a bit set in a Map-Request message. An ITR 2257 or PITR will send a Map-Request when they receive an SMR message. 2258 Both the SMR sender and the Map-Request responder MUST rate-limited 2259 these messages. Rate-limiting can be implemented as a global rate- 2260 limiter or one rate-limiter per SMR destination. 2262 The following procedure shows how a SMR exchange occurs when a site 2263 is doing locator-set compaction for an EID-to-RLOC mapping: 2265 1. When the database mappings in an ETR change, the ETRs at the site 2266 begin to send Map-Requests with the SMR bit set for each locator 2267 in each map-cache entry the ETR caches. 2269 2. A remote ITR which receives the SMR message will schedule sending 2270 a Map-Request message to the source locator address of the SMR 2271 message or to the mapping database system. A newly allocated 2272 random nonce is selected and the EID-prefix used is the one 2273 copied from the SMR message. If the source locator is the only 2274 locator in the cached locator-set, the remote ITR SHOULD send a 2275 Map-Request to the database mapping system just in case the 2276 single locator has changed and may no longer be reachable to 2277 accept the Map-Request. 2279 3. The remote ITR MUST rate-limit the Map-Request until it gets a 2280 Map-Reply while continuing to use the cached mapping. When Map 2281 Versioning is used, described in Section 6.6.3, an SMR sender can 2282 detect if an ITR is using the most up to date database mapping. 2284 4. The ETRs at the site with the changed mapping will reply to the 2285 Map-Request with a Map-Reply message that has a nonce from the 2286 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate 2287 limited. This is important to avoid Map-Reply implosion. 2289 5. The ETRs, at the site with the changed mapping, record the fact 2290 that the site that sent the Map-Request has received the new 2291 mapping data in the mapping cache entry for the remote site so 2292 the loc-status-bits are reflective of the new mapping for packets 2293 going to the remote site. The ETR then stops sending SMR 2294 messages. 2296 Experimentation is in progress to determine the appropriate rate- 2297 limit parameters. 2299 For security reasons an ITR MUST NOT process unsolicited Map-Replies. 2300 To avoid map-cache entry corruption by a third-party, a sender of an 2301 SMR-based Map-Request MUST be verified. If an ITR receives an SMR- 2302 based Map-Request and the source is not in the locator-set for the 2303 stored map-cache entry, then the responding Map-Request MUST be sent 2304 with an EID destination to the mapping database system. Since the 2305 mapping database system is more secure to reach an authoritative ETR, 2306 it will deliver the Map-Request to the authoritative source of the 2307 mapping data. 2309 When an ITR receives an SMR-based Map-Request for which it does not 2310 have a cached mapping for the EID in the SMR message, it MAY not send 2311 a SMR-invoked Map-Request. This scenario can occur when an ETR sends 2312 SMR messages to all locators in the locator-set it has stored in its 2313 map-cache but the remote ITRs that receive the SMR may not be sending 2314 packets to the site. There is no point in updating the ITRs until 2315 they need to send, in which case, they will send Map-Requests to 2316 obtain a map-cache entry. 2318 6.6.3. Database Map Versioning 2320 When there is unidirectional packet flow between an ITR and ETR, and 2321 the EID-to-RLOC mappings change on the ETR, it needs to inform the 2322 ITR so encapsulation can stop to a removed locator and start to a new 2323 locator in the locator-set. 2325 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 2326 number. This is known as the Destination Map-Version Number. ITRs 2327 include the Destination Map-Version Number in packets they 2328 encapsulate to the site. When an ETR decapsulates a packet and 2329 detects the Destination Map-Version Number is less than the current 2330 version for its mapping, the SMR procedure described in Section 6.6.2 2331 occurs. 2333 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 2334 Version number. This is known as the Source Map-Version Number. 2335 When an ETR decapsulates a packet and detects the Source Map-Version 2336 Number is greater than the last Map-Version Number sent in a Map- 2337 Reply from the ITR's site, the ETR will send a Map-Request to one of 2338 the ETRs for the source site. 2340 A Map-Version Number is used as a sequence number per EID-prefix. So 2341 values that are greater, are considered to be more recent. A value 2342 of 0 for the Source Map-Version Number or the Destination Map-Version 2343 Number conveys no versioning information and an ITR does no 2344 comparison with previously received Map-Version Numbers. 2346 A Map-Version Number can be included in Map-Register messages as 2347 well. This is a good way for the Map-Server can assure that all ETRs 2348 for a site registering to it will be Map-Version number synchronized. 2350 See [VERSIONING] for a more detailed analysis and description of 2351 Database Map Versioning. 2353 7. Router Performance Considerations 2355 LISP is designed to be very hardware-based forwarding friendly. A 2356 few implementation techniques can be used to incrementally implement 2357 LISP: 2359 o When a tunnel encapsulated packet is received by an ETR, the outer 2360 destination address may not be the address of the router. This 2361 makes it challenging for the control plane to get packets from the 2362 hardware. This may be mitigated by creating special FIB entries 2363 for the EID-prefixes of EIDs served by the ETR (those for which 2364 the router provides an RLOC translation). These FIB entries are 2365 marked with a flag indicating that control plane processing should 2366 be performed. The forwarding logic of testing for particular IP 2367 protocol number value is not necessary. There are a few proven 2368 cases where no changes to existing deployed hardware were needed 2369 to support the LISP data-plane. 2371 o On an ITR, prepending a new IP header consists of adding more 2372 bytes to a MAC rewrite string and prepending the string as part of 2373 the outgoing encapsulation procedure. Routers that support GRE 2374 tunneling [RFC2784] or 6to4 tunneling [RFC3056] may already 2375 support this action. 2377 o A packet's source address or interface the packet was received on 2378 can be used to select a VRF (Virtual Routing/Forwarding). The 2379 VRF's routing table can be used to find EID-to-RLOC mappings. 2381 For performance issues related to map-cache management, see section 2382 Section 12. 2384 8. Deployment Scenarios 2386 This section will explore how and where ITRs and ETRs can be deployed 2387 and will discuss the pros and cons of each deployment scenario. For 2388 a more detailed deployment recommendation, refer to [LISP-DEPLOY]. 2390 There are two basic deployment trade-offs to consider: centralized 2391 versus distributed caches and flat, recursive, or re-encapsulating 2392 tunneling. When deciding on centralized versus distributed caching, 2393 the following issues should be considered: 2395 o Are the tunnel routers spread out so that the caches are spread 2396 across all the memories of each router? A centralized cache is 2397 when an ITR keeps a cache for all the EIDs it is encapsulating to. 2398 The packet takes a direct path to the destination locator. A 2399 distributed cache is when an ITR needs help from other re- 2400 encapsulating routers because it does not store all the cache 2401 entries for the EIDs is it encapsulating to. So the packet takes 2402 a path through re-encapsulating routers that have a different set 2403 of cache entries. 2405 o Should management "touch points" be minimized by choosing few 2406 tunnel routers, just enough for redundancy? 2408 o In general, using more ITRs doesn't increase management load, 2409 since caches are built and stored dynamically. On the other hand, 2410 more ETRs does require more management since EID-prefix-to-RLOC 2411 mappings need to be explicitly configured. 2413 When deciding on flat, recursive, or re-encapsulation tunneling, the 2414 following issues should be considered: 2416 o Flat tunneling implements a single tunnel between source site and 2417 destination site. This generally offers better paths between 2418 sources and destinations with a single tunnel path. 2420 o Recursive tunneling is when tunneled traffic is again further 2421 encapsulated in another tunnel, either to implement VPNs or to 2422 perform Traffic Engineering. When doing VPN-based tunneling, the 2423 site has some control since the site is prepending a new tunnel 2424 header. In the case of TE-based tunneling, the site may have 2425 control if it is prepending a new tunnel header, but if the site's 2426 ISP is doing the TE, then the site has no control. Recursive 2427 tunneling generally will result in suboptimal paths but at the 2428 benefit of steering traffic to resource available parts of the 2429 network. 2431 o The technique of re-encapsulation ensures that packets only 2432 require one tunnel header. So if a packet needs to be rerouted, 2433 it is first decapsulated by the ETR and then re-encapsulated with 2434 a new tunnel header using a new RLOC. 2436 The next sub-sections will survey where tunnel routers can reside in 2437 the network. 2439 8.1. First-hop/Last-hop Tunnel Routers 2441 By locating tunnel routers close to hosts, the EID-prefix set is at 2442 the granularity of an IP subnet. So at the expense of more EID- 2443 prefix-to-RLOC sets for the site, the caches in each tunnel router 2444 can remain relatively small. But caches always depend on the number 2445 of non-aggregated EID destination flows active through these tunnel 2446 routers. 2448 With more tunnel routers doing encapsulation, the increase in control 2449 traffic grows as well: since the EID-granularity is greater, more 2450 Map-Requests and Map-Replies are traveling between more routers. 2452 The advantage of placing the caches and databases at these stub 2453 routers is that the products deployed in this part of the network 2454 have better price-memory ratios then their core router counterparts. 2455 Memory is typically less expensive in these devices and fewer routes 2456 are stored (only IGP routes). These devices tend to have excess 2457 capacity, both for forwarding and routing state. 2459 LISP functionality can also be deployed in edge switches. These 2460 devices generally have layer-2 ports facing hosts and layer-3 ports 2461 facing the Internet. Spare capacity is also often available in these 2462 devices as well. 2464 8.2. Border/Edge Tunnel Routers 2466 Using customer-edge (CE) routers for tunnel endpoints allows the EID 2467 space associated with a site to be reachable via a small set of RLOCs 2468 assigned to the CE routers for that site. This is the default 2469 behavior envisioned in the rest of this specification. 2471 This offers the opposite benefit of the first-hop/last-hop tunnel 2472 router scenario: the number of mapping entries and network management 2473 touch points are reduced, allowing better scaling. 2475 One disadvantage is that less of the network's resources are used to 2476 reach host endpoints thereby centralizing the point-of-failure domain 2477 and creating network choke points at the CE router. 2479 Note that more than one CE router at a site can be configured with 2480 the same IP address. In this case an RLOC is an anycast address. 2481 This allows resilience between the CE routers. That is, if a CE 2482 router fails, traffic is automatically routed to the other routers 2483 using the same anycast address. However, this comes with the 2484 disadvantage where the site cannot control the entrance point when 2485 the anycast route is advertised out from all border routers. Another 2486 disadvantage of using anycast locators is the limited advertisement 2487 scope of /32 (or /128 for IPv6) routes. 2489 8.3. ISP Provider-Edge (PE) Tunnel Routers 2491 Use of ISP PE routers as tunnel endpoint routers is not the typical 2492 deployment scenario envisioned in the specification. This section 2493 attempts to capture some of reasoning behind this preference of 2494 implementing LISP on CE routers. 2496 Use of ISP PE routers as tunnel endpoint routers gives an ISP, rather 2497 than a site, control over the location of the egress tunnel 2498 endpoints. That is, the ISP can decide if the tunnel endpoints are 2499 in the destination site (in either CE routers or last-hop routers 2500 within a site) or at other PE edges. The advantage of this case is 2501 that two tunnel headers can be avoided. By having the PE be the 2502 first router on the path to encapsulate, it can choose a TE path 2503 first, and the ETR can decapsulate and re-encapsulate for a tunnel to 2504 the destination end site. 2506 An obvious disadvantage is that the end site has no control over 2507 where its packets flow or the RLOCs used. Other disadvantages 2508 include the difficulty in synchronizing path liveness updates between 2509 CE and PE routers. 2511 As mentioned in earlier sections a combination of these scenarios is 2512 possible at the expense of extra packet header overhead, if both site 2513 and provider want control, then recursive or re-encapsulating tunnels 2514 are used. 2516 8.4. LISP Functionality with Conventional NATs 2518 LISP routers can be deployed behind Network Address Translator (NAT) 2519 devices to provide the same set of packet services hosts have today 2520 when they are addressed out of private address space. 2522 It is important to note that a locator address in any LISP control 2523 message MUST be a globally routable address and therefore SHOULD NOT 2524 contain [RFC1918] addresses. If a LISP router is configured with 2525 private addresses, they MUST be used only in the outer IP header so 2526 the NAT device can translate properly. Otherwise, EID addresses MUST 2527 be translated before encapsulation is performed. Both NAT 2528 translation and LISP encapsulation functions could be co-located in 2529 the same device. 2531 More details on LISP address translation can be found in [INTERWORK]. 2533 8.5. Packets Egressing a LISP Site 2535 When a LISP site is using two ITRs for redundancy, the failure of one 2536 ITR will likely shift outbound traffic to the second. This second 2537 ITR's cache may not not be populated with the same EID-to-RLOC 2538 mapping entries as the first. If this second ITR does not have these 2539 mappings, traffic will be dropped while the mappings are retrieved 2540 from the mapping system. The retrieval of these messages may 2541 increase the load of requests being sent into the mapping system. 2542 Deployment and experimentation will determine whether this issue 2543 requires more attention. 2545 9. Traceroute Considerations 2547 When a source host in a LISP site initiates a traceroute to a 2548 destination host in another LISP site, it is highly desirable for it 2549 to see the entire path. Since packets are encapsulated from ITR to 2550 ETR, the hop across the tunnel could be viewed as a single hop. 2551 However, LISP traceroute will provide the entire path so the user can 2552 see 3 distinct segments of the path from a source LISP host to a 2553 destination LISP host: 2555 Segment 1 (in source LISP site based on EIDs): 2557 source-host ---> first-hop ... next-hop ---> ITR 2559 Segment 2 (in the core network based on RLOCs): 2561 ITR ---> next-hop ... next-hop ---> ETR 2563 Segment 3 (in the destination LISP site based on EIDs): 2565 ETR ---> next-hop ... last-hop ---> destination-host 2567 For segment 1 of the path, ICMP Time Exceeded messages are returned 2568 in the normal manner as they are today. The ITR performs a TTL 2569 decrement and test for 0 before encapsulating. So the ITR hop is 2570 seen by the traceroute source has an EID address (the address of 2571 site-facing interface). 2573 For segment 2 of the path, ICMP Time Exceeded messages are returned 2574 to the ITR because the TTL decrement to 0 is done on the outer 2575 header, so the destination of the ICMP messages are to the ITR RLOC 2576 address, the source RLOC address of the encapsulated traceroute 2577 packet. The ITR looks inside of the ICMP payload to inspect the 2578 traceroute source so it can return the ICMP message to the address of 2579 the traceroute client as well as retaining the core router IP address 2580 in the ICMP message. This is so the traceroute client can display 2581 the core router address (the RLOC address) in the traceroute output. 2582 The ETR returns its RLOC address and responds to the TTL decrement to 2583 0 like the previous core routers did. 2585 For segment 3, the next-hop router downstream from the ETR will be 2586 decrementing the TTL for the packet that was encapsulated, sent into 2587 the core, decapsulated by the ETR, and forwarded because it isn't the 2588 final destination. If the TTL is decremented to 0, any router on the 2589 path to the destination of the traceroute, including the next-hop 2590 router or destination, will send an ICMP Time Exceeded message to the 2591 source EID of the traceroute client. The ICMP message will be 2592 encapsulated by the local ITR and sent back to the ETR in the 2593 originated traceroute source site, where the packet will be delivered 2594 to the host. 2596 9.1. IPv6 Traceroute 2598 IPv6 traceroute follows the procedure described above since the 2599 entire traceroute data packet is included in ICMP Time Exceeded 2600 message payload. Therefore, only the ITR needs to pay special 2601 attention for forwarding ICMP messages back to the traceroute source. 2603 9.2. IPv4 Traceroute 2605 For IPv4 traceroute, we cannot follow the above procedure since IPv4 2606 ICMP Time Exceeded messages only include the invoking IP header and 8 2607 bytes that follow the IP header. Therefore, when a core router sends 2608 an IPv4 Time Exceeded message to an ITR, all the ITR has in the ICMP 2609 payload is the encapsulated header it prepended followed by a UDP 2610 header. The original invoking IP header, and therefore the identity 2611 of the traceroute source is lost. 2613 The solution we propose to solve this problem is to cache traceroute 2614 IPv4 headers in the ITR and to match them up with corresponding IPv4 2615 Time Exceeded messages received from core routers and the ETR. The 2616 ITR will use a circular buffer for caching the IPv4 and UDP headers 2617 of traceroute packets. It will select a 16-bit number as a key to 2618 find them later when the IPv4 Time Exceeded messages are received. 2619 When an ITR encapsulates an IPv4 traceroute packet, it will use the 2620 16-bit number as the UDP source port in the encapsulating header. 2621 When the ICMP Time Exceeded message is returned to the ITR, the UDP 2622 header of the encapsulating header is present in the ICMP payload 2623 thereby allowing the ITR to find the cached headers for the 2624 traceroute source. The ITR puts the cached headers in the payload 2625 and sends the ICMP Time Exceeded message to the traceroute source 2626 retaining the source address of the original ICMP Time Exceeded 2627 message (a core router or the ETR of the site of the traceroute 2628 destination). 2630 The signature of a traceroute packet comes in two forms. The first 2631 form is encoded as a UDP message where the destination port is 2632 inspected for a range of values. The second form is encoded as an 2633 ICMP message where the IP identification field is inspected for a 2634 well-known value. 2636 9.3. Traceroute using Mixed Locators 2638 When either an IPv4 traceroute or IPv6 traceroute is originated and 2639 the ITR encapsulates it in the other address family header, you 2640 cannot get all 3 segments of the traceroute. Segment 2 of the 2641 traceroute can not be conveyed to the traceroute source since it is 2642 expecting addresses from intermediate hops in the same address format 2643 for the type of traceroute it originated. Therefore, in this case, 2644 segment 2 will make the tunnel look like one hop. All the ITR has to 2645 do to make this work is to not copy the inner TTL to the outer, 2646 encapsulating header's TTL when a traceroute packet is encapsulated 2647 using an RLOC from a different address family. This will cause no 2648 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2650 10. Mobility Considerations 2652 There are several kinds of mobility of which only some might be of 2653 concern to LISP. Essentially they are as follows. 2655 10.1. Site Mobility 2657 A site wishes to change its attachment points to the Internet, and 2658 its LISP Tunnel Routers will have new RLOCs when it changes upstream 2659 providers. Changes in EID-RLOC mappings for sites are expected to be 2660 handled by configuration, outside of the LISP protocol. 2662 10.2. Slow Endpoint Mobility 2664 An individual endpoint wishes to move, but is not concerned about 2665 maintaining session continuity. Renumbering is involved. LISP can 2666 help with the issues surrounding renumbering [RFC4192] [LISA96] by 2667 decoupling the address space used by a site from the address spaces 2668 used by its ISPs. [RFC4984] 2670 10.3. Fast Endpoint Mobility 2672 Fast endpoint mobility occurs when an endpoint moves relatively 2673 rapidly, changing its IP layer network attachment point. Maintenance 2674 of session continuity is a goal. This is where the Mobile IPv4 2675 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used, 2676 and primarily where interactions with LISP need to be explored. 2678 The problem is that as an endpoint moves, it may require changes to 2679 the mapping between its EID and a set of RLOCs for its new network 2680 location. When this is added to the overhead of mobile IP binding 2681 updates, some packets might be delayed or dropped. 2683 In IPv4 mobility, when an endpoint is away from home, packets to it 2684 are encapsulated and forwarded via a home agent which resides in the 2685 home area the endpoint's address belongs to. The home agent will 2686 encapsulate and forward packets either directly to the endpoint or to 2687 a foreign agent which resides where the endpoint has moved to. 2688 Packets from the endpoint may be sent directly to the correspondent 2689 node, may be sent via the foreign agent, or may be reverse-tunneled 2690 back to the home agent for delivery to the mobile node. As the 2691 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 2692 mappings are required for communication between the mobile node and 2693 the home agent, whether via foreign agent or not. As a mobile 2694 endpoint changes networks, up to three LISP mapping changes may be 2695 required: 2697 o The mobile node moves from an old location to a new visited 2698 network location and notifies its home agent that it has done so. 2699 The Mobile IPv4 control packets the mobile node sends pass through 2700 one of the new visited network's ITRs, which needs an EID-RLOC 2701 mapping for the home agent. 2703 o The home agent might not have the EID-RLOC mappings for the mobile 2704 node's "care-of" address or its foreign agent in the new visited 2705 network, in which case it will need to acquire them. 2707 o When packets are sent directly to the correspondent node, it may 2708 be that no traffic has been sent from the new visited network to 2709 the correspondent node's network, and the new visited network's 2710 ITR will need to obtain an EID-RLOC mapping for the correspondent 2711 node's site. 2713 In addition, if the IPv4 endpoint is sending packets from the new 2714 visited network using its original EID, then LISP will need to 2715 perform a route-returnability check on the new EID-RLOC mapping for 2716 that EID. 2718 In IPv6 mobility, packets can flow directly between the mobile node 2719 and the correspondent node in either direction. The mobile node uses 2720 its "care-of" address (EID). In this case, the route-returnability 2721 check would not be needed but one more LISP mapping lookup may be 2722 required instead: 2724 o As above, three mapping changes may be needed for the mobile node 2725 to communicate with its home agent and to send packets to the 2726 correspondent node. 2728 o In addition, another mapping will be needed in the correspondent 2729 node's ITR, in order for the correspondent node to send packets to 2730 the mobile node's "care-of" address (EID) at the new network 2731 location. 2733 When both endpoints are mobile the number of potential mapping 2734 lookups increases accordingly. 2736 As a mobile node moves there are not only mobility state changes in 2737 the mobile node, correspondent node, and home agent, but also state 2738 changes in the ITRs and ETRs for at least some EID-prefixes. 2740 The goal is to support rapid adaptation, with little delay or packet 2741 loss for the entire system. Also IP mobility can be modified to 2742 require fewer mapping changes. In order to increase overall system 2743 performance, there may be a need to reduce the optimization of one 2744 area in order to place fewer demands on another. 2746 In LISP, one possibility is to "glean" information. When a packet 2747 arrives, the ETR could examine the EID-RLOC mapping and use that 2748 mapping for all outgoing traffic to that EID. It can do this after 2749 performing a route-returnability check, to ensure that the new 2750 network location does have a internal route to that endpoint. 2751 However, this does not cover the case where an ITR (the node assigned 2752 the RLOC) at the mobile-node location has been compromised. 2754 Mobile IP packet exchange is designed for an environment in which all 2755 routing information is disseminated before packets can be forwarded. 2756 In order to allow the Internet to grow to support expected future 2757 use, we are moving to an environment where some information may have 2758 to be obtained after packets are in flight. Modifications to IP 2759 mobility should be considered in order to optimize the behavior of 2760 the overall system. Anything which decreases the number of new EID- 2761 RLOC mappings needed when a node moves, or maintains the validity of 2762 an EID-RLOC mapping for a longer time, is useful. 2764 10.4. Fast Network Mobility 2766 In addition to endpoints, a network can be mobile, possibly changing 2767 xTRs. A "network" can be as small as a single router and as large as 2768 a whole site. This is different from site mobility in that it is 2769 fast and possibly short-lived, but different from endpoint mobility 2770 in that a whole prefix is changing RLOCs. However, the mechanisms 2771 are the same and there is no new overhead in LISP. A map request for 2772 any endpoint will return a binding for the entire mobile prefix. 2774 If mobile networks become a more common occurrence, it may be useful 2775 to revisit the design of the mapping service and allow for dynamic 2776 updates of the database. 2778 The issue of interactions between mobility and LISP needs to be 2779 explored further. Specific improvements to the entire system will 2780 depend on the details of mapping mechanisms. Mapping mechanisms 2781 should be evaluated on how well they support session continuity for 2782 mobile nodes. 2784 10.5. LISP Mobile Node Mobility 2786 A mobile device can use the LISP infrastructure to achieve mobility 2787 by implementing the LISP encapsulation and decapsulation functions 2788 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 2789 node" can use topologically-independent EID IP addresses that are not 2790 advertised into and do not impose a cost on the global routing 2791 system. These EIDs are maintained at the edges of the mapping system 2792 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 2793 only the correspondents of the LISP mobile node. 2795 Refer to the LISP Mobility Architecture specification [LISP-MN] for 2796 more details. 2798 11. Multicast Considerations 2800 A multicast group address, as defined in the original Internet 2801 architecture is an identifier of a grouping of topologically 2802 independent receiver host locations. The address encoding itself 2803 does not determine the location of the receiver(s). The multicast 2804 routing protocol, and the network-based state the protocol creates, 2805 determines where the receivers are located. 2807 In the context of LISP, a multicast group address is both an EID and 2808 a Routing Locator. Therefore, no specific semantic or action needs 2809 to be taken for a destination address, as it would appear in an IP 2810 header. Therefore, a group address that appears in an inner IP 2811 header built by a source host will be used as the destination EID. 2812 The outer IP header (the destination Routing Locator address), 2813 prepended by a LISP router, will use the same group address as the 2814 destination Routing Locator. 2816 Having said that, only the source EID and source Routing Locator 2817 needs to be dealt with. Therefore, an ITR merely needs to put its 2818 own IP address in the source Routing Locator field when prepending 2819 the outer IP header. This source Routing Locator address, like any 2820 other Routing Locator address MUST be globally routable. 2822 Therefore, an EID-to-RLOC mapping does not need to be performed by an 2823 ITR when a received data packet is a multicast data packet or when 2824 processing a source-specific Join (either by IGMPv3 or PIM). But the 2825 source Routing Locator is decided by the multicast routing protocol 2826 in a receiver site. That is, an EID to Routing Locator translation 2827 is done at control-time. 2829 Another approach is to have the ITR not encapsulate a multicast 2830 packet and allow the host built packet to flow into the core even if 2831 the source address is allocated out of the EID namespace. If the 2832 RPF-Vector TLV [RFC5496] is used by PIM in the core, then core 2833 routers can RPF to the ITR (the Locator address which is injected 2834 into core routing) rather than the host source address (the EID 2835 address which is not injected into core routing). 2837 To avoid any EID-based multicast state in the network core, the first 2838 approach is chosen for LISP-Multicast. Details for LISP-Multicast 2839 and Interworking with non-LISP sites is described in specification 2840 [MLISP]. 2842 12. Security Considerations 2844 It is believed that most of the security mechanisms will be part of 2845 the mapping database service when using control plane procedures for 2846 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 2847 as described in this specification, protection is provided against 2848 ETR spoofing by using Return-Routability (see Section 3) mechanisms 2849 evidenced by the use of a 24-bit Nonce field in the LISP 2850 encapsulation header and a 64-bit Nonce field in the LISP control 2851 message. 2853 The nonce, coupled with the ITR accepting only solicited Map-Replies 2854 provides a basic level of security, in many ways similar to the 2855 security experienced in the current Internet routing system. It is 2856 hard for off-path attackers to launch attacks against these LISP 2857 mechanisms, as they do not have the nonce values. Sending a large 2858 number of packets to accidentally find the right nonce value is 2859 possible, but would already by itself be a denial-of-service attack. 2860 On-path attackers can perform far more serious attacks, but on-path 2861 attackers can launch serious attacks in the current Internet as well, 2862 including eavesdropping, blocking or redirecting traffic. See more 2863 discussion on this topic in Section 6.1.5.1. 2865 LISP does not rely on a PKI or a more heavy weight authentication 2866 system. These systems challenge the scalability of LISP which was a 2867 primary design goal. 2869 DoS attack prevention will depend on implementations rate-limiting 2870 Map-Requests and Map-Replies to the control plane as well as rate- 2871 limiting the number of data-triggered Map-Replies. 2873 An incorrectly implemented or malicious ITR might choose to ignore 2874 the priority and weights provided by the ETR in its Map-Reply. This 2875 traffic steering would be limited to the traffic that is sent by this 2876 ITR's site, and no more severe than if the site initiated a bandwidth 2877 DoS attack on (one of) the ETR's ingress links. The ITR's site would 2878 typically gain no benefit from not respecting the weights, and would 2879 likely to receive better service by abiding by them. 2881 To deal with map-cache exhaustion attempts in an ITR/PITR, the 2882 implementation should consider putting a maximum cap on the number of 2883 entries stored with a reserve list for special or frequently accessed 2884 sites. This should be a configuration policy control set by the 2885 network administrator who manages ITRs and PITRs. When overlapping 2886 EID-prefixes occur across multiple map-cache entries, the integrity 2887 of the set must be wholly maintained. So if a more-specific entry 2888 cannot be added due to reaching the maximum cap, then none of the 2889 less specifics should be stored in the map-cache. 2891 Given that the ITR/PITR maintains a cache of EID-to-RLOC mappings, 2892 cache sizing and maintenance is an issue to be kept in mind during 2893 implementation. It is a good idea to have instrumentation in place 2894 to detect thrashing of the cache. Implementation experimentation 2895 will be used to determine which cache management strategies work 2896 best. In general, it is difficult to defend against cache trashing 2897 attacks. It should be noted that an undersized cache in an ITR/PITR 2898 not only causes adverse affect on the site or region they support, 2899 but may also cause increased Map-Request load on the mapping system. 2901 There is a security risk implicit in the fact that ETRs generate the 2902 EID prefix to which they are responding. An ETR can claim a shorter 2903 prefix than it is actually responsible for. Various mechanisms to 2904 ameliorate or resolve this issue will be examined in the future, 2905 [LISP-SEC]. 2907 Spoofing of inner header addresses of LISP encapsulated packets is 2908 possible like with any tunneling mechanism. ITRs MUST verify the 2909 source address of a packet to be an EID that belongs to the site's 2910 EID-prefix range prior to encapsulation. An ETR must only 2911 decapsulate and forward datagrams with an inner header destination 2912 that matches one of its EID-prefix ranges. If, upon receipt and 2913 decapsulation, the destination EID of a datagram does not match one 2914 of the ETR's configured EID-prefixes, the ETR MUST drop the datagram. 2915 If a LISP encapsulated packet arrives at an ETR, it SHOULD compare 2916 the inner header source EID address and the outer header source RLOC 2917 address with the mapping that exists in the mapping database. Then 2918 when spoofing attacks occur, the outer header source RLOC address can 2919 be used to trace back the attack to the source site, using existing 2920 operational tools. 2922 This experimental specification does not address automated key 2923 management (AKM). BCP 107 provides guidance in this area. In 2924 addition, at the time of this writing, substantial work is being 2925 undertaken to improve security of the routing system [KARP], [RPKI], 2926 [BGP-SEC], [LISP-SEC]. Future work on LISP should address BCP-107 as 2927 well as other open security considerations, which may require changes 2928 to this specification. 2930 13. Network Management Considerations 2932 Considerations for Network Management tools exist so the LISP 2933 protocol suite can be operationally managed. The mechanisms can be 2934 found in [LISP-MIB] and [LISP-LIG]. 2936 14. IANA Considerations 2938 This section provides guidance to the Internet Assigned Numbers 2939 Authority (IANA) regarding registration of values related to the LISP 2940 specification, in accordance with BCP 26 and RFC 5226 [RFC5226]. 2942 There are two name spaces in LISP that require registration: 2944 o LISP IANA registry allocations should not be made for purposes 2945 unrelated to LISP routing or transport protocols. 2947 o The following policies are used here with the meanings defined in 2948 BCP 26: "Specification Required", "IETF Review", "Experimental 2949 Use", "First Come First Served". 2951 14.1. LISP ACT and Flag Fields 2953 New ACT values (Section 6.1.4) can be allocated through IETF review 2954 or IESG approval. Four values have already been allocated by this 2955 specification (Section 6.1.4). 2957 In addition, the LISP protocol has a number of flag and reserved 2958 fields, such as the LISP header flags field (Section 5.3). New bits 2959 for flags can be taken into use from these fields through IETF review 2960 or IESG approval, but these need not be managed by IANA. 2962 14.2. LISP Address Type Codes 2964 LISP Address [LCAF] type codes have a range from 0 to 255. New type 2965 codes MUST be allocated consecutively starting at 0. Type Codes 0 - 2966 127 are to be assigned by IETF review or IESG approval. 2968 Type Codes 128 - 255 are available on a First Come First Served 2969 policy. 2971 This registry, initially empty, is constructed for future-use 2972 experimental work of LCAF values. See [LCAF] for details for other 2973 possible unapproved address encodings. The unapproved LCAF encodings 2974 are an area for further study and experimentation. 2976 14.3. LISP UDP Port Numbers 2978 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2979 LISP data-plane and control-plane operation, respectively. 2981 14.4. LISP Key ID Numbers 2983 The following Key ID values are defined by this specification as used 2984 in any packet type that references a Key ID field: 2986 Name Number Defined in 2987 ----------------------------------------------- 2988 None 0 n/a 2989 HMAC-SHA-1-96 1 [RFC2404] 2990 HMAC-SHA-256-128 2 [RFC6234] 2992 15. Known Open Issues and Areas of Future Work 2994 As an experimental specification, this work is, by definition, 2995 incomplete. Specific areas where additional experience and work are 2996 needed include: 2998 o At present, only [ALT] is defined for implementing a database of 2999 EID-to-RLOC mapping information. Additional research on other 3000 mapping database systems is strongly encouraged. 3002 o Failure and recovery of LISP site partitioning (see Section 6.4), 3003 in the presence of redundant configuration (see Section 8.5) needs 3004 further research and experimentation. 3006 o The characteristics of map-cache management under exceptional 3007 conditions, such as denial-of-service attacks are not fully 3008 understood. Further experience is needed to determine whether 3009 current caching methods are practical or in need of further 3010 development. In particular, the performance, scaling and security 3011 characteristics of the map-cache will be discovered as part of 3012 this experiment. Performance metrics to be observed are packet 3013 reordering associated with the LISP data probe and loss of the 3014 first packet in a flow associated with map-caching. The impact of 3015 these upon TCP will be observed. See Section 12 for additional 3016 thoughts and considerations. 3018 o Preliminary work has been done to ensure that sites employing LISP 3019 can interconnect with the rest of the Internet. This work is 3020 documented in [INTERWORK] but further experimentation and 3021 experience is needed. 3023 o At present, no mechanism for automated key management for message 3024 authentication is defined. Addressing automated key management is 3025 necessary before this specification could be developed into a 3026 standards track RFC. See Section 12 for further details regarding 3027 security considerations. 3029 o In order to maintain security and stability, Internet Protocols 3030 typically isolate the control and data planes. Therefore, user 3031 activity cannot cause control plane state to be created or 3032 destroyed. LISP does not maintain this separation. The degree to 3033 which the loss of separation impacts security and stability is a 3034 topic for experimental observation. 3036 o LISP allows for different mapping database systems to be used. 3037 While only one [ALT] is currently well-defined, each mapping 3038 database will likely have some impact on the security of the EID- 3039 to-RLOC mappings. How each mapping database system's security 3040 properties impact on LISP overall is for further study. 3042 16. References 3044 16.1. Normative References 3046 [ALT] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "LISP 3047 Alternative Topology (LISP-ALT)", 3048 draft-ietf-lisp-alt-09.txt (work in progress). 3050 [LISP-MS] Farinacci, D. and V. Fuller, "LISP Map Server", 3051 draft-ietf-lisp-ms-12.txt (work in progress). 3053 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 3054 August 1980. 3056 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3057 September 1981. 3059 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 3060 E. Lear, "Address Allocation for Private Internets", 3061 BCP 5, RFC 1918, February 1996. 3063 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3064 Requirement Levels", BCP 14, RFC 2119, March 1997. 3066 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 3067 ESP and AH", RFC 2404, November 1998. 3069 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3070 (IPv6) Specification", RFC 2460, December 1998. 3072 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3073 of Explicit Congestion Notification (ECN) to IP", 3074 RFC 3168, September 2001. 3076 [RFC3232] Reynolds, J., "Assigned Numbers: RFC 1700 is Replaced by 3077 an On-line Database", RFC 3232, January 2002. 3079 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 3080 Requirements for Security", BCP 106, RFC 4086, June 2005. 3082 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 3083 (CIDR): The Internet Address Assignment and Aggregation 3084 Plan", BCP 122, RFC 4632, August 2006. 3086 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 3087 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 3088 May 2008. 3090 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 3091 Forwarding (RPF) Vector TLV", RFC 5496, March 2009. 3093 [RFC5944] Perkins, C., "IP Mobility Support for IPv4, Revised", 3094 RFC 5944, November 2010. 3096 [RFC6234] Eastlake, D. and T. Hansen, "US Secure Hash Algorithms 3097 (SHA and SHA-based HMAC and HKDF)", RFC 6234, May 2011. 3099 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 3100 in IPv6", RFC 6275, July 2011. 3102 [UDP-TUNNELS] 3103 Eubanks, M. and P. Chimento, "UDP Checksums for Tunneled 3104 Packets", draft-eubanks-chimento-6man-01.txt (work in 3105 progress), October 2010. 3107 [UDP-ZERO] 3108 Fairhurst, G. and M. Westerland, "IPv6 UDP Checksum 3109 Considerations", draft-ietf-6man-udpzero-04.txt (work in 3110 progress), October 2011. 3112 [VERSIONING] 3113 Iannone, L., Saucez, D., and O. Bonaventure, "LISP Mapping 3114 Versioning", draft-ietf-lisp-map-versioning-05.txt (work 3115 in progress). 3117 16.2. Informative References 3119 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3120 NUMBERS 3121 http://www.iana.org/assignments/address-family-numbers. 3123 [AFI-REGISTRY] 3124 IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3125 NUMBER registry http://www.iana.org/assignments/ 3126 address-family-numbers/ 3127 address-family-numbers.xml#address-family-numbers-1. 3129 [BGP-SEC] Lepinski, M., "An Overview of BGPSEC", 3130 draft-lepinski-bgpsec-overview-00.txt (work in progress), 3131 March 2011. 3133 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 3134 Enhancement to the Internet Architecture", Internet- 3135 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt. 3137 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 3138 Content distribution Overlay Network Service for LISP", 3139 draft-meyer-lisp-cons-04.txt (work in progress). 3141 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 3142 Mappings Multicast Across Cooperating Systems for LISP", 3143 draft-curran-lisp-emacs-00.txt (work in progress). 3145 [INTERWORK] 3146 Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 3147 "Interworking LISP with IPv4 and IPv6", 3148 draft-ietf-lisp-interworking-02.txt (work in progress). 3150 [KARP] Lebovitz, G. and M. Bhatia, "Keying and Authentication for 3151 Routing Protocols (KARP)Design Guidelines", 3152 draft-ietf-karp-design-guide-06.txt (work in progress), 3153 October 2011. 3155 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 3156 Address Format", draft-farinacci-lisp-lcaf-06.txt (work in 3157 progress). 3159 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 3160 "Renumbering: Threat or Menace?", Usenix . 3162 [LISP-DEPLOY] 3163 Jakab, L., Coras, F., Domingo-Pascual, J., and D. Lewis, 3164 "LISP Network Element Deployment Considerations", 3165 draft-ietf-lisp-deployment-02.txt (work in progress). 3167 [LISP-LIG] 3168 Farinacci, D. and D. Meyer, "LISP Internet Groper (LIG)", 3169 draft-ietf-lisp-lig-06.txt (work in progress). 3171 [LISP-MAIN] 3172 Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 3173 "Locator/ID Separation Protocol (LISP)", 3174 draft-farinacci-lisp-12.txt (work in progress). 3176 [LISP-MIB] 3177 Schudel, G., Jain, A., and V. Moreno, "LISP MIB", 3178 draft-ietf-lisp-mib-02.txt (work in progress). 3180 [LISP-MN] Farinacci, D., Fuller, V., Lewis, D., and D. Meyer, "LISP 3181 Mobility Architecture", draft-meyer-lisp-mn-06.txt (work 3182 in progress). 3184 [LISP-SEC] 3185 Maino, F., Ermagon, V., Cabellos, A., Sausez, D., and O. 3187 Bonaventure, "LISP-Security (LISP-SEC)", 3188 draft-ietf-lisp-sec-00.txt (work in progress). 3190 [LOC-ID-ARCH] 3191 Meyer, D. and D. Lewis, "Architectural Implications of 3192 Locator/ID Separation", 3193 draft-meyer-loc-id-implications-02.txt (work in progress). 3195 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 3196 "LISP for Multicast Environments", 3197 draft-ietf-lisp-multicast-10.txt (work in progress). 3199 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 3200 draft-lear-lisp-nerd-08.txt (work in progress). 3202 [OPENLISP] 3203 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 3204 Report", draft-iannone-openlisp-implementation-01.txt 3205 (work in progress). 3207 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 3208 draft-narten-radir-problem-statement-05.txt (work in 3209 progress). 3211 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 3212 STD 13, RFC 1034, November 1987. 3214 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3215 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3216 March 2000. 3218 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 3219 via IPv4 Clouds", RFC 3056, February 2001. 3221 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 3222 A., Peterson, J., Sparks, R., Handley, M., and E. 3223 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 3224 June 2002. 3226 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 3227 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 3228 September 2005. 3230 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 3231 Optimization for Mobile IPv6", RFC 4866, May 2007. 3233 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 3234 Workshop on Routing and Addressing", RFC 4984, 3235 September 2007. 3237 [RPKI] Lepinski, M., "An Infrastructure to Support Secure 3238 Internet Routing", draft-ietf-sidr-arch-13.txt (work in 3239 progress), February 2011. 3241 Appendix A. Acknowledgments 3243 An initial thank you goes to Dave Oran for planting the seeds for the 3244 initial ideas for LISP. His consultation continues to provide value 3245 to the LISP authors. 3247 A special and appreciative thank you goes to Noel Chiappa for 3248 providing architectural impetus over the past decades on separation 3249 of location and identity, as well as detailed review of the LISP 3250 architecture and documents, coupled with enthusiasm for making LISP a 3251 practical and incremental transition for the Internet. 3253 The authors would like to gratefully acknowledge many people who have 3254 contributed discussion and ideas to the making of this proposal. 3255 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 3256 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 3257 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 3258 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 3259 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 3260 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 3261 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 3262 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 3263 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 3264 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 3265 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 3266 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 3267 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 3268 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 3269 Ermagan, Albert Cabellos, Fabio Maino, Victor Moreno, Chris White, 3270 Clarence Filsfils, and Alia Atlas. 3272 This work originated in the Routing Research Group (RRG) of the IRTF. 3273 The individual submission [LISP-MAIN] was converted into this IETF 3274 LISP working group draft. 3276 The LISP working group would like to give a special thanks to Jari 3277 Arkko, the Internet Area AD at the time the set of LISP documents 3278 were being prepared for IESG last call, for his meticulous review and 3279 detail commentary on the 7 working group last call drafts progressing 3280 toward experimental RFCs. 3282 Appendix B. Document Change Log 3284 B.1. Changes to draft-ietf-lisp-18.txt 3286 o Posted December 2011 after reflecting comments from IANA. 3288 o Create reference to sections 5.4.1 and 5.4.2 about DF bit setting 3289 from section 5.3. 3291 o Inserted two references for Route-Returnability and on-path 3292 attacks in Security Considerations section. 3294 B.2. Changes to draft-ietf-lisp-17.txt 3296 o Posted December 2011 after IETF last call comments. 3298 o Make Map-Notify port assignment be 4342 in both source and 3299 destination ports. This change was agreed on and put in [LISP-MS] 3300 but was not updated in this spec. 3302 B.3. Changes to draft-ietf-lisp-16.txt 3304 o Posted October 2011 after AD review by Jari. 3306 B.4. Changes to draft-ietf-lisp-15.txt 3308 o Posted July 2011. Fixing IDnits errors. 3310 o Change description on how to select a source address for RLOC- 3311 probe Map-Replies to refer to the "EID-to-RLOC Map-Reply Message" 3312 section. 3314 B.5. Changes to draft-ietf-lisp-14.txt 3316 o Post working group last call and pre-IESG last call review. 3318 o Indicate that an ICMP Unreachable message should be sent when a 3319 packet matches a drop-based negative map-cache entry. 3321 o Indicate how a map-cache set of overlapping EID-prefixes must 3322 maintain integrity when the map-cache maximum cap is reached. 3324 o Add Joel's description for the definition of an EID, that the bit 3325 string value can be an RLOC for another device in abstract but the 3326 architecture allows it to be an EID of one device and the same 3327 value as an RLOC for another device. 3329 o In the "Tunnel Encapsulation Details" section, indicate that 4 3330 combinations of encapsulation are supported. 3332 o Add what ETR should do for a Data-Probe when received for a 3333 destination EID outside of its EID-prefix range. This was added 3334 in the Data Probe definition section. 3336 o Added text indicating that more-specific EID-prefixes must not be 3337 removed when less-specific entries stay in the map-cache. This is 3338 to preserve the integrity of the EID-prefix set. 3340 o Add clarifying text in the Security Considerations section about 3341 how an ETR must not decapsulate and forward a packet that is not 3342 for its configured EID-prefix range. 3344 B.6. Changes to draft-ietf-lisp-13.txt 3346 o Posted June 2011 to complete working group last call. 3348 o Tracker item 87. Put Yakov suggested wording in the EID-prefix 3349 definition section to reference [INTERWORK] and [LISP-DEPLOY] 3350 about discussion on transition and access mechanisms. 3352 o Change "ITRs" to "ETRs" in the Locator Status Bit definition 3353 section and data packet description section per Damien's comment. 3355 o Remove the normative reference to [LISP-SEC] when describing the 3356 S-bit in the ECM and Map-Reply headers. 3358 o Tracker item 54. Added text from John Scudder in the "Packets 3359 Egressing a LISP Site" section. 3361 o Add sentence to the "Reencapsulating Tunnel" definition about how 3362 reencapsulation loops can occur when not coordinating among 3363 multiple mapping database systems. 3365 o Remove "In theory" from a sentence in the Security Considerations 3366 section. 3368 o Remove Security Area Statement title and reword section with 3369 Eliot's provided text. The text was agreed upon by LISP-WG chairs 3370 and Security ADs. 3372 o Remove word "potential" from the over-claiming paragraph of the 3373 Security Considerations section per Stephen's request. 3375 o Wordsmithing and other editorial comments from Alia. 3377 B.7. Changes to draft-ietf-lisp-12.txt 3379 o Posted April 2011. 3381 o Tracker item 87. Provided rewording how an EID-prefix can be 3382 reused in the definition section of "EID-prefix". 3384 o Tracker item 95. Change "eliminate" to "defer" in section 4.1. 3386 o Tracker item 110. Added that the Mapping Protocol Data field in 3387 the Map-Reply message is only used when needed by the particular 3388 Mapping Database System. 3390 o Tracker item 111. Indicate that if an LSB that is associated with 3391 an anycast address, that there is at least one RLOC that is up. 3393 o Tracker item 108. Make clear the R-bit does not define RLOC path 3394 reachability. 3396 o Tracker item 107. Indicate that weights are relative to each 3397 other versus requiring an addition of up to 100%. 3399 o Tracker item 46. Add a sentence how LISP products should be sized 3400 for the appropriate demand so cache thrashing is avoided. 3402 o Change some references of RFC 5226 to [AFI] per Luigi. 3404 o Per Luigi, make reference to "EID-AFI" consistent to "EID-prefix- 3405 AFI". 3407 o Tracker item 66. Indicate that appending locators to a locator- 3408 set is done when the added locators are lexicographically greater 3409 than the previous ones in the set. 3411 o Tracker item 87. Once again reword the definition of the EID- 3412 prefix to reflect recent comments. 3414 o Tracker item 70. Added text to security section on what the 3415 implications could be if an ITR does not obey priority and weights 3416 from a Map-Reply message. 3418 o Tracker item 54. Added text to the new section titled "Packets 3419 Egressing a LISP Site" to describe the implications when two or 3420 more ITRs exist at a site where only one ITR is used for egress 3421 traffic and when there is a shift of traffic to the others, how 3422 the map-cache will need to be populated in those new egress ITRs. 3424 o Tracker item 33. Make more clear in the Routing Locator Selection 3425 section what an ITR should do when it sees an R-bit of 0 in a 3426 locator-record of a Map-Reply. 3428 o Tracker item 33. Add paragraph to the EID Reachability section 3429 indicating that site partitioning is under investigation. 3431 o Tracker item 58. Added last paragraph of Security Considerations 3432 section about how to protect inner header EID address spoofing 3433 attacks. 3435 o Add suggested Sam text to indicate that all security concerns need 3436 not be addressed for moving document to Experimental RFC status. 3437 Put this in a subsection of the Security Considerations section. 3439 B.8. Changes to draft-ietf-lisp-11.txt 3441 o Posted March 30, 2011. 3443 o Change IANA URL. The URL we had pointed to a general protocol 3444 numbers page. 3446 o Added the "s" bit to the Map-Request to allow SMR-invoked Map- 3447 Requests to be sent to a MN ETR via the map-server. 3449 o Generalize text for the definition of Reencapsuatling tunnels. 3451 o Add paragraph suggested by Joel to explain how implementation 3452 experimentation will be used to determine the proper cache 3453 management techniques. 3455 o Add Yakov provided text for the definition of "EID-to-RLOC 3456 "Database". 3458 o Add reference in Section 8, Deployment Scenarios, to the 3459 draft-jakab-lisp-deploy-02.txt draft. 3461 o Clarify sentence about no hardware changes needed to support LISP 3462 encapsulation. 3464 o Add paragraph about what is the procedure when a locator is 3465 inserted in the middle of a locator-set. 3467 o Add a definition for Locator Status Bits so we can emphasize they 3468 are used as a hint for router up/down status and not path 3469 reachability. 3471 o Change "BGP RIB" to "RIB" per Clarence's comment. 3473 o Fixed complaints by IDnits. 3475 o Add subsection to Security Considerations section indicating how 3476 EID-prefix overclaiming in Map-Replies is for further study and 3477 add a reference to LISP-SEC. 3479 B.9. Changes to draft-ietf-lisp-10.txt 3481 o Posted March 2011. 3483 o Add p-bit to Map-Request so there is documentary reasons to know 3484 when a PITR has sent a Map-Request to an ETR. 3486 o Add Map-Notify message which is used to acknowledge a Map-Register 3487 message sent to a Map-Server. 3489 o Add M-bit to the Map-Register message so an ETR that wants an 3490 acknowledgment for the Map-Register can request one. 3492 o Add S-bit to the ECM and Map-Reply messages to describe security 3493 data that can be present in each message. Then refer to 3494 [LISP-SEC] for expansive details. 3496 o Add Network Management Considerations section and point to the MIB 3497 and LIG drafts. 3499 o Remove the word "simple" per Yakov's comments. 3501 B.10. Changes to draft-ietf-lisp-09.txt 3503 o Posted October 2010. 3505 o Add to IANA Consideration section about the use of LCAF Type 3506 values that accepted and maintained by the IANA registry and not 3507 the LCAF specification. 3509 o Indicate that implementations should be able to receive LISP 3510 control messages when either UDP port is 4342, so they can be 3511 robust in the face of intervening NAT boxes. 3513 o Add paragraph to SMR section to indicate that an ITR does not need 3514 to respond to an SMR-based Map-Request when it has no map-cache 3515 entry for the SMR source's EID-prefix. 3517 B.11. Changes to draft-ietf-lisp-08.txt 3519 o Posted August 2010. 3521 o In section 6.1.6, remove statement about setting TTL to 0 in Map- 3522 Register messages. 3524 o Clarify language in section 6.1.5 about Map-Replying to Data- 3525 Probes or Map-Requests. 3527 o Indicate that outer TTL should only be copied to inner TTL when it 3528 is less than inner TTL. 3530 o Indicate a source-EID for RLOC-probes are encoded with an AFI 3531 value of 0. 3533 o Indicate that SMRs can have a global or per SMR destination rate- 3534 limiter. 3536 o Add clarifications to the SMR procedures. 3538 o Add definitions for "client-side" and 'server-side" terms used in 3539 this specification. 3541 o Clear up language in section 6.4, last paragraph. 3543 o Change ACT of value 0 to "no-action". This is so we can RLOC- 3544 probe a PETR and have it return a Map-Reply with a locator-set of 3545 size 0. The way it is spec'ed the map-cache entry has action 3546 "dropped". Drop-action is set to 3. 3548 o Add statement about normalizing locator weights. 3550 o Clarify R-bit definition in the Map-Reply locator record. 3552 o Add section on EID Reachability within a LISP site. 3554 o Clarify another disadvantage of using anycast locators. 3556 o Reworded Abstract. 3558 o Change section 2.0 Introduction to remove obsolete information 3559 such as the LISP variant definitions. 3561 o Change section 5 title from "Tunneling Details" to "LISP 3562 Encapsulation Details". 3564 o Changes to section 5 to include results of network deployment 3565 experience with MTU. Recommend that implementations use either 3566 the stateful or stateless handling. 3568 o Make clarification wordsmithing to Section 7 and 8. 3570 o Identify that if there is one locator in the locator-set of a map- 3571 cache entry, that an SMR from that locator should be responded to 3572 by sending the the SMR-invoked Map-Request to the database mapping 3573 system rather than to the RLOC itself (which may be unreachable). 3575 o When describing Unicast and Multicast Weights indicate the the 3576 values are relative weights rather than percentages. So it 3577 doesn't imply the sum of all locator weights in the locator-set 3578 need to be 100. 3580 o Do some wordsmithing on copying TTL and TOS fields. 3582 o Numerous wordsmithing changes from Dave Meyer. He fine toothed 3583 combed the spec. 3585 o Removed Section 14 "Prototype Plans and Status". We felt this 3586 type of section is no longer appropriate for a protocol 3587 specification. 3589 o Add clarification text for the IRC description per Damien's 3590 commentary. 3592 o Remove text on copying nonce from SMR to SMR-invoked Map- Request 3593 per Vina's comment about a possible DoS vector. 3595 o Clarify (S/2 + H) in the stateless MTU section. 3597 o Add text to reflect Damien's comment about the description of the 3598 "ITR-RLOC Address" field in the Map-Request. that the list of RLOC 3599 addresses are local addresses of the Map-Requester. 3601 B.12. Changes to draft-ietf-lisp-07.txt 3603 o Posted April 2010. 3605 o Added I-bit to data header so LSB field can also be used as an 3606 Instance ID field. When this occurs, the LSB field is reduced to 3607 8-bits (from 32-bits). 3609 o Added V-bit to the data header so the 24-bit nonce field can also 3610 be used for source and destination version numbers. 3612 o Added Map-Version 12-bit value to the EID-record to be used in all 3613 of Map-Request, Map-Reply, and Map-Register messages. 3615 o Added multiple ITR-RLOC fields to the Map-Request packet so an ETR 3616 can decide what address to select for the destination of a Map- 3617 Reply. 3619 o Added L-bit (Local RLOC bit) and p-bit (Probe-Reply RLOC bit) to 3620 the Locator-Set record of an EID-record for a Map-Reply message. 3621 The L-bit indicates which RLOCs in the locator-set are local to 3622 the sender of the message. The P-bit indicates which RLOC is the 3623 source of a RLOC-probe Reply (Map-Reply) message. 3625 o Add reference to the LISP Canonical Address Format [LCAF] draft. 3627 o Made editorial and clarification changes based on comments from 3628 Dhirendra Trivedi. 3630 o Added wordsmithing comments from Joel Halpern on DF=1 setting. 3632 o Add John Zwiebel clarification to Echo Nonce Algorithm section 3633 6.3.1. 3635 o Add John Zwiebel comment about expanding on proxy-map-reply bit 3636 for Map-Register messages. 3638 o Add NAT section per Ron Bonica comments. 3640 o Fix IDnits issues per Ron Bonica. 3642 o Added section on Virtualization and Segmentation to explain the 3643 use if the Instance ID field in the data header. 3645 o There are too many P-bits, keep their scope to the packet format 3646 description and refer to them by name every where else in the 3647 spec. 3649 o Scanned all occurrences of "should", "should not", "must" and 3650 "must not" and uppercased them. 3652 o John Zwiebel offered text for section 4.1 to modernize the 3653 example. Thanks Z! 3655 o Make it more clear in the definition of "EID-to-RLOC Database" 3656 that all ETRs need to have the same database mapping. This 3657 reflects a comment from John Scudder. 3659 o Add a definition "Route-returnability" to the Definition of Terms 3660 section. 3662 o In section 9.2, add text to describe what the signature of 3663 traceroute packets can look like. 3665 o Removed references to Data Probe for introductory example. Data- 3666 probes are still part of the LISP design but not encouraged. 3668 o Added the definition for "LISP site" to the Definition of Terms" 3669 section. 3671 B.13. Changes to draft-ietf-lisp-06.txt 3673 Editorial based changes: 3675 o Posted December 2009. 3677 o Fix typo for flags in LISP data header. Changed from "4" to "5". 3679 o Add text to indicate that Map-Register messages must contain a 3680 computed UDP checksum. 3682 o Add definitions for PITR and PETR. 3684 o Indicate an AFI value of 0 is an unspecified address. 3686 o Indicate that the TTL field of a Map-Register is not used and set 3687 to 0 by the sender. This change makes this spec consistent with 3688 [LISP-MS]. 3690 o Change "... yield a packet size of L bytes" to "... yield a packet 3691 size greater than L bytes". 3693 o Clarify section 6.1.5 on what addresses and ports are used in Map- 3694 Reply messages. 3696 o Clarify that LSBs that go beyond the number of locators do not to 3697 be SMRed when the locator addresses are greater lexicographically 3698 than the locator in the existing locator-set. 3700 o Add Gregg, Srini, and Amit to acknowledgment section. 3702 o Clarify in the definition of a LISP header what is following the 3703 UDP header. 3705 o Clarify "verifying Map-Request" text in section 6.1.3. 3707 o Add Xu Xiaohu to the acknowledgment section for introducing the 3708 problem of overlapping EID-prefixes among multiple sites in an RRG 3709 email message. 3711 Design based changes: 3713 o Use stronger language to have the outer IPv4 header set DF=1 so we 3714 can avoid fragment reassembly in an ETR or PETR. This will also 3715 make IPv4 and IPv6 encapsulation have consistent behavior. 3717 o Map-Requests should not be sent in ECM with the Probe bit is set. 3718 These type of Map-Requests are used as RLOC-probes and are sent 3719 directly to locator addresses in the underlying network. 3721 o Add text in section 6.1.5 about returning all EID-prefixes in a 3722 Map-Reply sent by an ETR when there are overlapping EID-prefixes 3723 configure. 3725 o Add text in a new subsection of section 6.1.5 about dealing with 3726 Map-Replies with coarse EID-prefixes. 3728 B.14. Changes to draft-ietf-lisp-05.txt 3730 o Posted September 2009. 3732 o Added this Document Change Log appendix. 3734 o Added section indicating that encapsulated Map-Requests must use 3735 destination UDP port 4342. 3737 o Don't use AH in Map-Registers. Put key-id, auth-length, and auth- 3738 data in Map-Register payload. 3740 o Added Jari to acknowledgment section. 3742 o State the source-EID is set to 0 when using Map-Requests to 3743 refresh or RLOC-probe. 3745 o Make more clear what source-RLOC should be for a Map-Request. 3747 o The LISP-CONS authors thought that the Type definitions for CONS 3748 should be removed from this specification. 3750 o Removed nonce from Map-Register message, it wasn't used so no need 3751 for it. 3753 o Clarify what to do for unspecified Action bits for negative Map- 3754 Replies. Since No Action is a drop, make value 0 Drop. 3756 B.15. Changes to draft-ietf-lisp-04.txt 3758 o Posted September 2009. 3760 o How do deal with record count greater than 1 for a Map-Request. 3761 Damien and Joel comment. Joel suggests: 1) Specify that senders 3762 compliant with the current document will always set the count to 3763 1, and note that the count is included for future extensibility. 3764 2) Specify what a receiver compliant with the draft should do if 3765 it receives a request with a count greater than 1. Presumably, it 3766 should send some error back? 3768 o Add Fred Templin in acknowledgment section. 3770 o Add Margaret and Sam to the acknowledgment section for their great 3771 comments. 3773 o Say more about LAGs in the UDP section per Sam Hartman's comment. 3775 o Sam wants to use MAY instead of SHOULD for ignoring checksums on 3776 ETR. From the mailing list: "You'd need to word it as an ITR MAY 3777 send a zero checksum, an ETR MUST accept a 0 checksum and MAY 3778 ignore the checksum completely. And of course we'd need to 3779 confirm that can actually be implemented. In particular, hardware 3780 that verifies UDP checksums on receive needs to be checked to make 3781 sure it permits 0 checksums." 3783 o Margaret wants a reference to 3784 http://www.ietf.org/id/draft-eubanks-chimento-6man-00.txt. 3786 o Fix description in Map-Request section. Where we describe Map- 3787 Reply Record, change "R-bit" to "M-bit". 3789 o Add the mobility bit to Map-Replies. So PITRs don't probe so 3790 often for MNs but often enough to get mapping updates. 3792 o Indicate SHA1 can be used as well for Map-Registers. 3794 o More Fred comments on MTU handling. 3796 o Isidor comment about spec'ing better periodic Map-Registers. Will 3797 be fixed in draft-ietf-lisp-ms-02.txt. 3799 o Margaret's comment on gleaning: "The current specification does 3800 not make it clear how long gleaned map entries should be retained 3801 in the cache, nor does it make it clear how/ when they will be 3802 validated. The LISP spec should, at the very least, include a 3803 (short) default lifetime for gleaned entries, require that they be 3804 validated within a short period of time, and state that a new 3805 gleaned entry should never overwrite an entry that was obtained 3806 from the mapping system. The security implications of storing 3807 "gleaned" entries should also be explored in detail." 3809 o Add section on RLOC-probing per working group feedback. 3811 o Change "loc-reach-bits" to "loc-status-bits" per comment from 3812 Noel. 3814 o Remove SMR-bit from data-plane. Dino prefers to have it in the 3815 control plane only. 3817 o Change LISP header to allow a "Research Bit" so the Nonce and LSB 3818 fields can be turned off and used for another future purpose. For 3819 Luigi et al versioning convergence. 3821 o Add a N-bit to the data header suggested by Noel. Then the nonce 3822 field could be used when N is not 1. 3824 o Clarify that when E-bit is 0, the nonce field can be an echoed 3825 nonce or a random nonce. Comment from Jesper. 3827 o Indicate when doing data-gleaning that a verifying Map-Request is 3828 sent to the source-EID of the gleaned data packet so we can avoid 3829 map-cache corruption by a 3rd party. Comment from Pedro. 3831 o Indicate that a verifying Map-Request, for accepting mapping data, 3832 should be sent over the ALT (or to the EID). 3834 o Reference IPsec RFC 4302. Comment from Sam and Brian Weis. 3836 o Put E-bit in Map-Reply to tell ITRs that the ETR supports echo- 3837 noncing. Comment by Pedro and Dino. 3839 o Jesper made a comment to loosen the language about requiring the 3840 copy of inner TTL to outer TTL since the text to get mixed-AF 3841 traceroute to work would violate the "MUST" clause. Changed from 3842 MUST to SHOULD in section 5.3. 3844 B.16. Changes to draft-ietf-lisp-03.txt 3846 o Posted July 2009. 3848 o Removed loc-reach-bits longword from control packets per Damien 3849 comment. 3851 o Clarifications in MTU text from Roque. 3853 o Added text to indicate that the locator-set be sorted by locator 3854 address from Isidor. 3856 o Clarification text from John Zwiebel in Echo-Nonce section. 3858 B.17. Changes to draft-ietf-lisp-02.txt 3860 o Posted July 2009. 3862 o Encapsulation packet format change to add E-bit and make loc- 3863 reach-bits 32-bits in length. 3865 o Added Echo-Nonce Algorithm section. 3867 o Clarification how ECN bits are copied. 3869 o Moved S-bit in Map-Request. 3871 o Added P-bit in Map-Request and Map-Reply messages to anticipate 3872 RLOC-Probe Algorithm. 3874 o Added to Mobility section to reference [LISP-MN]. 3876 B.18. Changes to draft-ietf-lisp-01.txt 3878 o Posted 2 days after draft-ietf-lisp-00.txt in May 2009. 3880 o Defined LEID to be a "LISP EID". 3882 o Indicate encapsulation use IPv4 DF=0. 3884 o Added negative Map-Reply messages with drop, native-forward, and 3885 send-map-request actions. 3887 o Added Proxy-Map-Reply bit to Map-Register. 3889 B.19. Changes to draft-ietf-lisp-00.txt 3891 o Posted May 2009. 3893 o Rename of draft-farinacci-lisp-12.txt. 3895 o Acknowledgment to RRG. 3897 Authors' Addresses 3899 Dino Farinacci 3900 cisco Systems 3901 Tasman Drive 3902 San Jose, CA 95134 3903 USA 3905 Email: dino@cisco.com 3907 Vince Fuller 3908 cisco Systems 3909 Tasman Drive 3910 San Jose, CA 95134 3911 USA 3913 Email: vaf@cisco.com 3915 Dave Meyer 3916 cisco Systems 3917 170 Tasman Drive 3918 San Jose, CA 3919 USA 3921 Email: dmm@cisco.com 3923 Darrel Lewis 3924 cisco Systems 3925 170 Tasman Drive 3926 San Jose, CA 3927 USA 3929 Email: darlewis@cisco.com