idnits 2.17.1 draft-ietf-lisp-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Sep 2009 rather than the newer Notice from 28 Dec 2009. (See https://trustee.ietf.org/license-info/) 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 5 instances of lines with non-RFC2606-compliant FQDNs in the document. == 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 (March 4, 2011) is 4800 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 1700 (Obsoleted by RFC 3232) ** Obsolete normative reference: RFC 3775 (Obsoleted by RFC 6275) ** Obsolete normative reference: RFC 4634 (Obsoleted by RFC 6234) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-01) exists of draft-eubanks-chimento-6man-00 == Outdated reference: A later version (-10) exists of draft-ietf-lisp-alt-06 == Outdated reference: A later version (-04) exists of draft-meyer-lisp-cons-03 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-interworking-02 == Outdated reference: A later version (-10) exists of draft-farinacci-lisp-lcaf-04 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-lig-01 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-mib-01 == Outdated reference: A later version (-16) exists of draft-meyer-lisp-mn-04 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-ms-07 == Outdated reference: A later version (-14) exists of draft-ietf-lisp-multicast-04 == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-08 == Outdated reference: A later version (-05) exists of draft-narten-radir-problem-statement-00 == Outdated reference: A later version (-10) exists of draft-ietf-mip4-rfc3344bis-05 -- No information found for draft-handley-p2ppush-unpublished-2007726 - is the name correct? == Outdated reference: A later version (-09) exists of draft-ietf-lisp-map-versioning-01 Summary: 5 errors (**), 0 flaws (~~), 17 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: September 5, 2011 D. Lewis 6 cisco Systems 7 March 4, 2011 9 Locator/ID Separation Protocol (LISP) 10 draft-ietf-lisp-10 12 Abstract 14 This draft describes a network-based protocol that enables separation 15 of IP addresses into two new numbering spaces: Endpoint Identifiers 16 (EIDs) and Routing Locators (RLOCs). No changes are required to 17 either host protocol stacks or to the "core" of the Internet 18 infrastructure. LISP can be incrementally deployed, without a "flag 19 day", and offers traffic engineering, multi-homing, and mobility 20 benefits even to early adopters, when there are relatively few LISP- 21 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 to IETF 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), its areas, and its working groups. Note that 34 other groups may also distribute working documents as Internet- 35 Drafts. 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 The list of current Internet-Drafts can be accessed at 43 http://www.ietf.org/ietf/1id-abstracts.txt. 45 The list of Internet-Draft Shadow Directories can be accessed at 46 http://www.ietf.org/shadow.html. 48 This Internet-Draft will expire on September 5, 2011. 50 Copyright Notice 52 Copyright (c) 2011 IETF Trust and the persons identified as the 53 document authors. All rights reserved. 55 This document is subject to BCP 78 and the IETF Trust's Legal 56 Provisions Relating to IETF Documents 57 (http://trustee.ietf.org/license-info) in effect on the date of 58 publication of this document. Please review these documents 59 carefully, as they describe your rights and restrictions with respect 60 to this document. Code Components extracted from this document must 61 include Simplified BSD License text as described in Section 4.e of 62 the Trust Legal Provisions and are provided without warranty as 63 described in the BSD License. 65 Table of Contents 67 1. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 68 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 69 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 7 70 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . . 12 71 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . . 14 72 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . . 16 73 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 17 74 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 17 75 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . . 19 76 5.4. Dealing with Large Encapsulated Packets . . . . . . . . . 22 77 5.4.1. A Stateless Solution to MTU Handling . . . . . . . . . 23 78 5.4.2. A Stateful Solution to MTU Handling . . . . . . . . . 24 79 5.5. Using Virtualization and Segmentation with LISP . . . . . 24 80 6. EID-to-RLOC Mapping . . . . . . . . . . . . . . . . . . . . . 26 81 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats . . . . . 26 82 6.1.1. LISP Packet Type Allocations . . . . . . . . . . . . . 28 83 6.1.2. Map-Request Message Format . . . . . . . . . . . . . . 28 84 6.1.3. EID-to-RLOC UDP Map-Request Message . . . . . . . . . 30 85 6.1.4. Map-Reply Message Format . . . . . . . . . . . . . . . 32 86 6.1.5. EID-to-RLOC UDP Map-Reply Message . . . . . . . . . . 36 87 6.1.6. Map-Register Message Format . . . . . . . . . . . . . 38 88 6.1.7. Map-Notify Message Format . . . . . . . . . . . . . . 40 89 6.1.8. Encapsulated Control Message Format . . . . . . . . . 41 90 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 43 91 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 45 92 6.3.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . 47 93 6.3.2. RLOC Probing Algorithm . . . . . . . . . . . . . . . . 48 94 6.4. EID Reachability within a LISP Site . . . . . . . . . . . 49 95 6.5. Routing Locator Hashing . . . . . . . . . . . . . . . . . 49 96 6.6. Changing the Contents of EID-to-RLOC Mappings . . . . . . 50 97 6.6.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 51 98 6.6.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 52 99 6.6.3. Database Map Versioning . . . . . . . . . . . . . . . 53 100 7. Router Performance Considerations . . . . . . . . . . . . . . 55 101 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 56 102 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 57 103 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 57 104 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 58 105 8.4. LISP Functionality with Conventional NATs . . . . . . . . 58 106 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 59 107 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 60 108 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 60 109 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 60 110 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 62 111 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 62 112 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 62 113 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 62 114 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 64 115 10.5. LISP Mobile Node Mobility . . . . . . . . . . . . . . . . 64 116 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 66 117 12. Security Considerations . . . . . . . . . . . . . . . . . . . 67 118 13. Network Management Considerations . . . . . . . . . . . . . . 68 119 14. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 69 120 14.1. LISP Address Type Codes . . . . . . . . . . . . . . . . . 69 121 14.2. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . . 69 122 15. References . . . . . . . . . . . . . . . . . . . . . . . . . . 70 123 15.1. Normative References . . . . . . . . . . . . . . . . . . . 70 124 15.2. Informative References . . . . . . . . . . . . . . . . . . 71 125 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 74 126 Appendix B. Document Change Log . . . . . . . . . . . . . . . . . 75 127 B.1. Changes to draft-ietf-lisp-10.txt . . . . . . . . . . . . 75 128 B.2. Changes to draft-ietf-lisp-09.txt . . . . . . . . . . . . 75 129 B.3. Changes to draft-ietf-lisp-08.txt . . . . . . . . . . . . 75 130 B.4. Changes to draft-ietf-lisp-07.txt . . . . . . . . . . . . 77 131 B.5. Changes to draft-ietf-lisp-06.txt . . . . . . . . . . . . 79 132 B.6. Changes to draft-ietf-lisp-05.txt . . . . . . . . . . . . 80 133 B.7. Changes to draft-ietf-lisp-04.txt . . . . . . . . . . . . 80 134 B.8. Changes to draft-ietf-lisp-03.txt . . . . . . . . . . . . 82 135 B.9. Changes to draft-ietf-lisp-02.txt . . . . . . . . . . . . 83 136 B.10. Changes to draft-ietf-lisp-01.txt . . . . . . . . . . . . 83 137 B.11. Changes to draft-ietf-lisp-00.txt . . . . . . . . . . . . 83 138 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 84 140 1. Requirements Notation 142 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 143 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 144 document are to be interpreted as described in [RFC2119]. 146 2. Introduction 148 This document describes the Locator/Identifier Separation Protocol 149 (LISP), which provides a set of functions for routers to exchange 150 information used to map from non-routeable Endpoint Identifiers 151 (EIDs) to routeable Routing Locators (RLOCs). It also defines a 152 mechanism for these LISP routers to encapsulate IP packets addressed 153 with EIDs for transmission across an Internet that uses RLOCs for 154 routing and forwarding. 156 Creation of LISP was initially motivated by discussions during the 157 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 158 October, 2006 (see [RFC4984]). A key conclusion of the workshop was 159 that the Internet routing and addressing system was not scaling well 160 in the face of the explosive growth of new sites; one reason for this 161 poor scaling is the increasing number of multi-homed and other sites 162 that cannot be addressed as part of topologically- or provider-based 163 aggregated prefixes. Additional work that more completely described 164 the problem statement may be found in [RADIR]. 166 A basic observation, made many years ago in early networking research 167 such as that documented in [CHIAPPA] and [RFC4984], is that using a 168 single address field for both identifying a device and for 169 determining where it is topologically located in the network requires 170 optimization along two conflicting axes: for routing to be efficient, 171 the address must be assigned topologically; for collections of 172 devices to be easily and effectively managed, without the need for 173 renumbering in response to topological change (such as that caused by 174 adding or removing attachment points to the network or by mobility 175 events), the address must explicitly not be tied to the topology. 177 The approach that LISP takes to solving the routing scalability 178 problem is to replace IP addresses with two new types of numbers: 179 Routing Locators (RLOCs), which are topologically assigned to network 180 attachment points (and are therefore amenable to aggregation) and 181 used for routing and forwarding of packets through the network; and 182 Endpoint Identifiers (EIDs), which are assigned independently from 183 the network topology, are used for numbering devices, and are 184 aggregated along administrative boundaries. LISP then defines 185 functions for mapping between the two numbering spaces and for 186 encapsulating traffic originated by devices using non-routeable EIDs 187 for transport across a network infrastructure that routes and 188 forwards using RLOCs. Both RLOCs and EIDs are syntactically- 189 identical to IP addresses; it is the semantics of how they are used 190 that differs. 192 This document describes the protocol that implements these functions. 193 The database which stores the mappings between EIDs and RLOCs is 194 explicitly a separate "module" to facilitate experimentation with a 195 variety of approaches. One database design that is being developed 196 and prototyped as part of the LISP working group work is [ALT]. 197 Others that have been described but not implemented include [CONS], 198 [EMACS], [RPMD], [NERD]. Finally, [LISP-MS], documents a general- 199 purpose service interface for accessing a mapping database; this 200 interface is intended to make the mapping database modular so that 201 different approaches can be tried without the need to modify 202 installed xTRs. 204 3. Definition of Terms 206 Provider Independent (PI) Addresses: PI addresses are an address 207 block assigned from a pool where blocks are not associated with 208 any particular location in the network (e.g. from a particular 209 service provider), and is therefore not topologically aggregatable 210 in the routing system. 212 Provider Assigned (PA) Addresses: PA addresses are an a address 213 block assigned to a site by each service provider to which a site 214 connects. Typically, each block is sub-block of a service 215 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 216 is aggregated into the larger block before being advertised into 217 the global Internet. Traditionally, IP multihoming has been 218 implemented by each multi-homed site acquiring its own, globally- 219 visible prefix. LISP uses only topologically-assigned and 220 aggregatable address blocks for RLOCs, eliminating this 221 demonstrably non-scalable practice. 223 Routing Locator (RLOC): A RLOC is an IPv4 or IPv6 address of an 224 egress tunnel router (ETR). A RLOC is the output of a EID-to-RLOC 225 mapping lookup. An EID maps to one or more RLOCs. Typically, 226 RLOCs are numbered from topologically-aggregatable blocks that are 227 assigned to a site at each point to which it attaches to the 228 global Internet; where the topology is defined by the connectivity 229 of provider networks, RLOCs can be thought of as PA addresses. 230 Multiple RLOCs can be assigned to the same ETR device or to 231 multiple ETR devices at a site. 233 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 234 IPv6) value used in the source and destination address fields of 235 the first (most inner) LISP header of a packet. The host obtains 236 a destination EID the same way it obtains an destination address 237 today, for example through a Domain Name System (DNS) [RFC1034] 238 lookup or Session Invitation Protocol (SIP) [RFC3261] exchange. 239 The source EID is obtained via existing mechanisms used to set a 240 host's "local" IP address. An EID is allocated to a host from an 241 EID-prefix block associated with the site where the host is 242 located. An EID can be used by a host to refer to other hosts. 243 EIDs MUST NOT be used as LISP RLOCs. Note that EID blocks may be 244 assigned in a hierarchical manner, independent of the network 245 topology, to facilitate scaling of the mapping database. In 246 addition, an EID block assigned to a site may have site-local 247 structure (subnetting) for routing within the site; this structure 248 is not visible to the global routing system. When used in 249 discussions with other Locator/ID separation proposals, a LISP EID 250 will be called a "LEID". Throughout this document, any references 251 to "EID" refers to an LEID. 253 EID-prefix: An EID-prefix is a power-of-two block of EIDs which are 254 allocated to a site by an address allocation authority. EID- 255 prefixes are associated with a set of RLOC addresses which make up 256 a "database mapping". EID-prefix allocations can be broken up 257 into smaller blocks when an RLOC set is to be associated with the 258 smaller EID-prefix. A globally routed address block (whether PI 259 or PA) is not an EID-prefix. However, a globally routed address 260 block may be removed from global routing and reused as an EID- 261 prefix. A site that receives an explicitly allocated EID-prefix 262 may not use that EID-prefix as a globally routed prefix assigned 263 to RLOCs. 265 End-system: An end-system is an IPv4 or IPv6 device that originates 266 packets with a single IPv4 or IPv6 header. The end-system 267 supplies an EID value for the destination address field of the IP 268 header when communicating globally (i.e. outside of its routing 269 domain). An end-system can be a host computer, a switch or router 270 device, or any network appliance. 272 Ingress Tunnel Router (ITR): An ITR is a router which accepts an IP 273 packet with a single IP header (more precisely, an IP packet that 274 does not contain a LISP header). The router treats this "inner" 275 IP destination address as an EID and performs an EID-to-RLOC 276 mapping lookup. The router then prepends an "outer" IP header 277 with one of its globally-routable RLOCs in the source address 278 field and the result of the mapping lookup in the destination 279 address field. Note that this destination RLOC may be an 280 intermediate, proxy device that has better knowledge of the EID- 281 to-RLOC mapping closer to the destination EID. In general, an ITR 282 receives IP packets from site end-systems on one side and sends 283 LISP-encapsulated IP packets toward the Internet on the other 284 side. 286 Specifically, when a service provider prepends a LISP header for 287 Traffic Engineering purposes, the router that does this is also 288 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 289 on the outer destination address (the originating ITR's supplied 290 RLOC) or the inner destination address (the originating hosts 291 supplied EID). 293 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 294 network that prepends an additional LISP header for Traffic 295 Engineering purposes. 297 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 298 packet where the destination address in the "outer" IP header is 299 one of its own RLOCs. The router strips the "outer" header and 300 forwards the packet based on the next IP header found. In 301 general, an ETR receives LISP-encapsulated IP packets from the 302 Internet on one side and sends decapsulated IP packets to site 303 end-systems on the other side. ETR functionality does not have to 304 be limited to a router device. A server host can be the endpoint 305 of a LISP tunnel as well. 307 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 308 network that strips an outer LISP header for Traffic Engineering 309 purposes. 311 xTR: A xTR is a reference to an ITR or ETR when direction of data 312 flow is not part of the context description. xTR refers to the 313 router that is the tunnel endpoint. Used synonymously with the 314 term "Tunnel Router". For example, "An xTR can be located at the 315 Customer Edge (CE) router", meaning both ITR and ETR functionality 316 is at the CE router. 318 EID-to-RLOC Cache: The EID-to-RLOC cache is a short-lived, on- 319 demand table in an ITR that stores, tracks, and is responsible for 320 timing-out and otherwise validating EID-to-RLOC mappings. This 321 cache is distinct from the full "database" of EID-to-RLOC 322 mappings, it is dynamic, local to the ITR(s), and relatively small 323 while the database is distributed, relatively static, and much 324 more global in scope. 326 EID-to-RLOC Database: The EID-to-RLOC database is a global 327 distributed database that contains all known EID-prefix to RLOC 328 mappings. Each potential ETR typically contains a small piece of 329 the database: the EID-to-RLOC mappings for the EID prefixes 330 "behind" the router. These map to one of the router's own, 331 globally-visible, IP addresses. The same database mapping entries 332 MUST be configured on all ETRs for a given site. That is, the 333 EID-prefixes for the site and locator-set for each EID-prefix MUST 334 be the same on all ETRs so they consistently send Map-Reply 335 messages with the same database mapping contents. 337 Recursive Tunneling: Recursive tunneling occurs when a packet has 338 more than one LISP IP header. Additional layers of tunneling may 339 be employed to implement traffic engineering or other re-routing 340 as needed. When this is done, an additional "outer" LISP header 341 is added and the original RLOCs are preserved in the "inner" 342 header. Any references to tunnels in this specification refers to 343 dynamic encapsulating tunnels and never are they statically 344 configured. 346 Reencapsulating Tunnels: Reencapsulating tunneling occurs when a 347 packet has no more than one LISP IP header (two IP headers total) 348 and when it needs to be diverted to new RLOC, an ETR can 349 decapsulate the packet (remove the LISP header) and prepends a new 350 tunnel header, with new RLOC, on to the packet. Doing this allows 351 a packet to be re-routed by the re-encapsulating router without 352 adding the overhead of additional tunnel headers. Any references 353 to tunnels in this specification refers to dynamic encapsulating 354 tunnels and never are they statically configured. 356 LISP Header: a term used in this document to refer to the outer 357 IPv4 or IPv6 header, a UDP header, and a LISP-specific 8-byte 358 header that follows the UDP header, an ITR prepends or an ETR 359 strips. 361 Address Family Identifier (AFI): a term used to describe an address 362 encoding in a packet. An address family currently pertains to an 363 IPv4 or IPv6 address. See [AFI] and [RFC1700] for details. An 364 AFI value of 0 used in this specification indicates an unspecified 365 encoded address where the length of the address is 0 bytes 366 following the 16-bit AFI value of 0. 368 Negative Mapping Entry: A negative mapping entry, also known as a 369 negative cache entry, is an EID-to-RLOC entry where an EID-prefix 370 is advertised or stored with no RLOCs. That is, the locator-set 371 for the EID-to-RLOC entry is empty or has an encoded locator count 372 of 0. This type of entry could be used to describe a prefix from 373 a non-LISP site, which is explicitly not in the mapping database. 374 There are a set of well defined actions that are encoded in a 375 Negative Map-Reply. 377 Data Probe: A data-probe is a LISP-encapsulated data packet where 378 the inner header destination address equals the outer header 379 destination address used to trigger a Map-Reply by a decapsulating 380 ETR. In addition, the original packet is decapsulated and 381 delivered to the destination host. A Data Probe is used in some 382 of the mapping database designs to "probe" or request a Map-Reply 383 from an ETR; in other cases, Map-Requests are used. See each 384 mapping database design for details. 386 Proxy ITR (PITR): A PITR is also known as a PTR is defined and 387 described in [INTERWORK], a PITR acts like an ITR but does so on 388 behalf of non-LISP sites which send packets to destinations at 389 LISP sites. 391 Proxy ETR (PETR): A PETR is defined and described in [INTERWORK], a 392 PETR acts like an ETR but does so on behalf of LISP sites which 393 send packets to destinations at non-LISP sites. 395 Route-returnability: is an assumption that the underlying routing 396 system will deliver packets to the destination. When combined 397 with a nonce that is provided by a sender and returned by a 398 receiver limits off-path data insertion. 400 LISP site: is a set of routers in an edge network that are under a 401 single technical administration. LISP routers which reside in the 402 edge network are the demarcation points to separate the edge 403 network from the core network. 405 Client-side: a term used in this document to indicate a connection 406 initiation attempt by an EID. The ITR(s) at the LISP site are the 407 first to get involved in obtaining database map cache entries by 408 sending Map-Request messages. 410 Server-side: a term used in this document to indicate a connection 411 initiation attempt is being accepted for a destination EID. The 412 ETR(s) at the destination LISP site are the first to send Map- 413 Replies to the source site initiating the connection. The ETR(s) 414 at this destination site can obtain mappings by gleaning 415 information from Map-Requests, Data-Probes, or encapsulated 416 packets. 418 4. Basic Overview 420 One key concept of LISP is that end-systems (hosts) operate the same 421 way they do today. The IP addresses that hosts use for tracking 422 sockets, connections, and for sending and receiving packets do not 423 change. In LISP terminology, these IP addresses are called Endpoint 424 Identifiers (EIDs). 426 Routers continue to forward packets based on IP destination 427 addresses. When a packet is LISP encapsulated, these addresses are 428 referred to as Routing Locators (RLOCs). Most routers along a path 429 between two hosts will not change; they continue to perform routing/ 430 forwarding lookups on the destination addresses. For routers between 431 the source host and the ITR as well as routers from the ETR to the 432 destination host, the destination address is an EID. For the routers 433 between the ITR and the ETR, the destination address is an RLOC. 435 Another key LISP concept is the "Tunnel Router". A tunnel router 436 prepends LISP headers on host-originated packets and strip them prior 437 to final delivery to their destination. The IP addresses in this 438 "outer header" are RLOCs. During end-to-end packet exchange between 439 two Internet hosts, an ITR prepends a new LISP header to each packet 440 and an egress tunnel router strips the new header. The ITR performs 441 EID-to-RLOC lookups to determine the routing path to the ETR, which 442 has the RLOC as one of its IP addresses. 444 Some basic rules governing LISP are: 446 o End-systems (hosts) only send to addresses which are EIDs. They 447 don't know addresses are EIDs versus RLOCs but assume packets get 448 to LISP routers, which in turn, deliver packets to the destination 449 the end-system has specified. 451 o EIDs are always IP addresses assigned to hosts. 453 o LISP routers mostly deal with Routing Locator addresses. See 454 details later in Section 4.1 to clarify what is meant by "mostly". 456 o RLOCs are always IP addresses assigned to routers; preferably, 457 topologically-oriented addresses from provider CIDR blocks. 459 o When a router originates packets it may use as a source address 460 either an EID or RLOC. When acting as a host (e.g. when 461 terminating a transport session such as SSH, TELNET, or SNMP), it 462 may use an EID that is explicitly assigned for that purpose. An 463 EID that identifies the router as a host MUST NOT be used as an 464 RLOC; an EID is only routable within the scope of a site. A 465 typical BGP configuration might demonstrate this "hybrid" EID/RLOC 466 usage where a router could use its "host-like" EID to terminate 467 iBGP sessions to other routers in a site while at the same time 468 using RLOCs to terminate eBGP sessions to routers outside the 469 site. 471 o EIDs are not expected to be usable for global end-to-end 472 communication in the absence of an EID-to-RLOC mapping operation. 473 They are expected to be used locally for intra-site communication. 475 o EID prefixes are likely to be hierarchically assigned in a manner 476 which is optimized for administrative convenience and to 477 facilitate scaling of the EID-to-RLOC mapping database. The 478 hierarchy is based on a address allocation hierarchy which is 479 independent of the network topology. 481 o EIDs may also be structured (subnetted) in a manner suitable for 482 local routing within an autonomous system. 484 An additional LISP header may be prepended to packets by a TE-ITR 485 when re-routing of the path for a packet is desired. An obvious 486 instance of this would be an ISP router that needs to perform traffic 487 engineering for packets flowing through its network. In such a 488 situation, termed Recursive Tunneling, an ISP transit acts as an 489 additional ingress tunnel router and the RLOC it uses for the new 490 prepended header would be either a TE-ETR within the ISP (along 491 intra-ISP traffic engineered path) or a TE-ETR within another ISP (an 492 inter-ISP traffic engineered path, where an agreement to build such a 493 path exists). 495 In order to avoid excessive packet overhead as well as possible 496 encapsulation loops, this document mandates that a maximum of two 497 LISP headers can be prepended to a packet. It is believed two 498 headers is sufficient, where the first prepended header is used at a 499 site for Location/Identity separation and second prepended header is 500 used inside a service provider for Traffic Engineering purposes. 502 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 503 For example, the ITR for a particular end-to-end packet exchange 504 might be the first-hop or default router within a site for the source 505 host. Similarly, the egress tunnel router might be the last-hop 506 router directly-connected to the destination host. Another example, 507 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 508 could be the site's border router at the service provider attachment 509 point. Mixing and matching of site-operated, ISP-operated, and other 510 tunnel routers is allowed for maximum flexibility. See Section 8 for 511 more details. 513 4.1. Packet Flow Sequence 515 This section provides an example of the unicast packet flow with the 516 following conditions: 518 o Source host "host1.abc.com" is sending a packet to 519 "host2.xyz.com", exactly what host1 would do if the site was not 520 using LISP. 522 o Each site is multi-homed, so each tunnel router has an address 523 (RLOC) assigned from the service provider address block for each 524 provider to which that particular tunnel router is attached. 526 o The ITR(s) and ETR(s) are directly connected to the source and 527 destination, respectively, but the source and destination can be 528 located anywhere in LISP site. 530 o Map-Requests can be sent on the underlying routing system topology 531 or over an alternative topology [ALT]. 533 o Map-Replies are sent on the underlying routing system topology. 535 Client host1.abc.com wants to communicate with server host2.xyz.com: 537 1. host1.abc.com wants to open a TCP connection to host2.xyz.com. 538 It does a DNS lookup on host2.xyz.com. An A/AAAA record is 539 returned. This address is the destination EID. The locally- 540 assigned address of host1.abc.com is used as the source EID. An 541 IPv4 or IPv6 packet is built and forwarded through the LISP site 542 as a normal IP packet until it reaches a LISP ITR. 544 2. The LISP ITR must be able to map the EID destination to an RLOC 545 of one of the ETRs at the destination site. The specific method 546 used to do this is not described in this example. See [ALT] or 547 [CONS] for possible solutions. 549 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be 550 rate-limited. 552 4. When an alternate mapping system is not in use, the Map-Request 553 packet is routed through the underlying routing system. 554 Otherwise, the Map-Request packet is routed on an alternate 555 logical topology. In either case, when the Map-Request arrives 556 at one of the ETRs at the destination site, it will process the 557 packet as a control message. 559 5. The ETR looks at the destination EID of the Map-Request and 560 matches it against the prefixes in the ETR's configured EID-to- 561 RLOC mapping database. This is the list of EID-prefixes the ETR 562 is supporting for the site it resides in. If there is no match, 563 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 564 returned to the ITR. 566 6. The ITR receives the Map-Reply message, parses the message (to 567 check for format validity) and stores the mapping information 568 from the packet. This information is stored in the ITR's EID-to- 569 RLOC mapping cache. Note that the map cache is an on-demand 570 cache. An ITR will manage its map cache in such a way that 571 optimizes for its resource constraints. 573 7. Subsequent packets from host1.abc.com to host2.xyz.com will have 574 a LISP header prepended by the ITR using the appropriate RLOC as 575 the LISP header destination address learned from the ETR. Note 576 the packet may be sent to a different ETR than the one which 577 returned the Map-Reply due to the source site's hashing policy or 578 the destination site's locator-set policy. 580 8. The ETR receives these packets directly (since the destination 581 address is one of its assigned IP addresses), strips the LISP 582 header and forwards the packets to the attached destination host. 584 In order to eliminate the need for a mapping lookup in the reverse 585 direction, an ETR MAY create a cache entry that maps the source EID 586 (inner header source IP address) to the source RLOC (outer header 587 source IP address) in a received LISP packet. Such a cache entry is 588 termed a "gleaned" mapping and only contains a single RLOC for the 589 EID in question. More complete information about additional RLOCs 590 SHOULD be verified by sending a LISP Map-Request for that EID. Both 591 ITR and the ETR may also influence the decision the other makes in 592 selecting an RLOC. See Section 6 for more details. 594 5. LISP Encapsulation Details 596 Since additional tunnel headers are prepended, the packet becomes 597 larger and can exceed the MTU of any link traversed from the ITR to 598 the ETR. It is recommended in IPv4 that packets do not get 599 fragmented as they are encapsulated by the ITR. Instead, the packet 600 is dropped and an ICMP Too Big message is returned to the source. 602 This specification recommends that implementations support for one of 603 the proposed fragmentation and reassembly schemes. These two 604 existing schemes are detailed in Section 5.4. 606 5.1. LISP IPv4-in-IPv4 Header Format 608 0 1 2 3 609 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 1 610 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 611 / |Version| IHL |Type of Service| Total Length | 612 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 613 | | Identification |Flags| Fragment Offset | 614 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 615 OH | Time to Live | Protocol = 17 | Header Checksum | 616 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 617 | | Source Routing Locator | 618 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 619 \ | Destination Routing Locator | 620 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 621 / | Source Port = xxxx | Dest Port = 4341 | 622 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 623 \ | UDP Length | UDP Checksum | 624 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 625 L |N|L|E|V|I|flags| Nonce/Map-Version | 626 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 627 S / | Instance ID/Locator Status Bits | 628 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 629 / |Version| IHL |Type of Service| Total Length | 630 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 631 | | Identification |Flags| Fragment Offset | 632 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 633 IH | Time to Live | Protocol | Header Checksum | 634 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 635 | | Source EID | 636 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 637 \ | Destination EID | 638 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 640 5.2. LISP IPv6-in-IPv6 Header Format 642 0 1 2 3 643 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 1 644 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 645 / |Version| Traffic Class | Flow Label | 646 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 648 | | Payload Length | Next Header=17| Hop Limit | 649 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 650 | | 651 O + + 652 u | | 653 t + Source Routing Locator + 654 e | | 655 r + + 656 | | 657 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 658 d | | 659 r + + 660 | | 661 ^ + Destination Routing Locator + 662 | | | 663 \ + + 664 \ | | 665 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 666 / | Source Port = xxxx | Dest Port = 4341 | 667 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 668 \ | UDP Length | UDP Checksum | 669 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 670 L |N|L|E|V|I|flags| Nonce/Map-Version | 671 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 672 S / | Instance ID/Locator Status Bits | 673 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 674 / |Version| Traffic Class | Flow Label | 675 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 676 / | Payload Length | Next Header | Hop Limit | 677 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 678 | | 679 I + + 680 n | | 681 n + Source EID + 682 e | | 683 r + + 684 | | 685 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 686 d | | 687 r + + 688 | | 689 ^ + Destination EID + 690 \ | | 691 \ + + 692 \ | | 693 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 695 5.3. Tunnel Header Field Descriptions 697 Inner Header: The inner header is the header on the datagram 698 received from the originating host. The source and destination IP 699 addresses are EIDs. 701 Outer Header: The outer header is a new header prepended by an ITR. 702 The address fields contain RLOCs obtained from the ingress 703 router's EID-to-RLOC cache. The IP protocol number is "UDP (17)" 704 from [RFC0768]. The DF bit of the Flags field is set to 0 when 705 the method in Section 5.4.1 is used and set to 1 when the method 706 in Section 5.4.2 is used. 708 UDP Header: The UDP header contains a ITR selected source port when 709 encapsulating a packet. See Section 6.5 for details on the hash 710 algorithm used to select a source port based on the 5-tuple of the 711 inner header. The destination port MUST be set to the well-known 712 IANA assigned port value 4341. 714 UDP Checksum: The UDP checksum field SHOULD be transmitted as zero 715 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 716 [UDP-TUNNELS]. When a packet with a zero UDP checksum is received 717 by an ETR, the ETR MUST accept the packet for decapsulation. When 718 an ITR transmits a non-zero value for the UDP checksum, it MUST 719 send a correctly computed value in this field. When an ETR 720 receives a packet with a non-zero UDP checksum, it MAY choose to 721 verify the checksum value. If it chooses to perform such 722 verification, and the verification fails, the packet MUST be 723 silently dropped. If the ETR chooses not to perform the 724 verification, or performs the verification successfully, the 725 packet MUST be accepted for decapsulation. The handling of UDP 726 checksums for all tunneling protocols, including LISP, is under 727 active discussion within the IETF. When that discussion 728 concludes, any necessary changes will be made to align LISP with 729 the outcome of the broader discussion. 731 UDP Length: The UDP length field is for an IPv4 encapsulated packet, 732 the inner header Total Length plus the UDP and LISP header lengths 733 are used. For an IPv6 encapsulated packet, the inner header 734 Payload Length plus the size of the IPv6 header (40 bytes) plus 735 the size of the UDP and LISP headers are used. The UDP header 736 length is 8 bytes. 738 N: The N bit is the nonce-present bit. When this bit is set to 1, 739 the low-order 24-bits of the first 32-bits of the LISP header 740 contains a Nonce. See Section 6.3.1 for details. Both N and V 741 bits MUST NOT be set in the same packet. If they are, a 742 decapsulating ETR MUST treat the "Nonce/Map-Version" field as 743 having a Nonce value present. 745 L: The L bit is the Locator-Status-Bits field enabled bit. When this 746 bit is set to 1, the Locator-Status-Bits in the second 32-bits of 747 the LISP header are in use. 749 x 1 x x 0 x x x 750 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 751 |N|L|E|V|I|flags| Nonce/Map-Version | 752 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 753 | Locator Status Bits | 754 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 756 E: The E bit is the echo-nonce-request bit. When this bit is set to 757 1, the N bit MUST be 1. This bit SHOULD be ignored and has no 758 meaning when the N bit is set to 0. See Section 6.3.1 for 759 details. 761 V: The V bit is the Map-Version present bit. When this bit is set to 762 1, the N bit MUST be 0. Refer to Section 6.6.3 for more details. 763 This bit indicates that the first 4 bytes of the LISP header is 764 encoded as: 766 0 x 0 1 x x x x 767 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 768 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version | 769 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 770 | Instance ID/Locator Status Bits | 771 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 773 I: The I bit is the Instance ID bit. See Section 5.5 for more 774 details. When this bit is set to 1, the Locator Status Bits field 775 is reduced to 8-bits and the high-order 24-bits are used as an 776 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 777 are transmitted as zero and ignored on receipt. The format of the 778 last 4 bytes of the LISP header would look like: 780 x x x x 1 x x x 781 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 782 |N|L|E|V|I|flags| Nonce/Map-Version | 783 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 784 | Instance ID | LSBs | 785 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 787 flags: The flags field is a 3-bit field is reserved for future flag 788 use. It is set to 0 on transmit and ignored on receipt. 790 LISP Nonce: The LISP nonce field is a 24-bit value that is randomly 791 generated by an ITR when the N-bit is set to 1. The nonce is also 792 used when the E-bit is set to request the nonce value to be echoed 793 by the other side when packets are returned. When the E-bit is 794 clear but the N-bit is set, a remote ITR is either echoing a 795 previously requested echo-nonce or providing a random nonce. See 796 Section 6.3.1 for more details. 798 LISP Locator Status Bits: The locator status bits field in the LISP 799 header is set by an ITR to indicate to an ETR the up/down status 800 of the Locators in the source site. Each RLOC in a Map-Reply is 801 assigned an ordinal value from 0 to n-1 (when there are n RLOCs in 802 a mapping entry). The Locator Status Bits are numbered from 0 to 803 n-1 from the least significant bit of field. The field is 32-bits 804 when the I-bit is set to 0 and is 8 bits when the I-bit is set to 805 1. When a Locator Status Bit is set to 1, the ITR is indicating 806 to the ETR the RLOC associated with the bit ordinal has up status. 807 See Section 6.3 for details on how an ITR can determine the status 808 of other ITRs at the same site. When a site has multiple EID- 809 prefixes which result in multiple mappings (where each could have 810 a different locator-set), the Locator Status Bits setting in an 811 encapsulated packet MUST reflect the mapping for the EID-prefix 812 that the inner-header source EID address matches. 814 When doing ITR/PITR encapsulation: 816 o The outer header Time to Live field (or Hop Limit field, in case 817 of IPv6) SHOULD be copied from the inner header Time to Live 818 field. 820 o The outer header Type of Service field (or the Traffic Class 821 field, in the case of IPv6) SHOULD be copied from the inner header 822 Type of Service field (with one caveat, see below). 824 When doing ETR/PETR decapsulation: 826 o The inner header Time to Live field (or Hop Limit field, in case 827 of IPv6) SHOULD be copied from the outer header Time to Live 828 field, when the Time to Live field of the outer header is less 829 than the Time to Live of the inner header. Failing to perform 830 this check can cause the Time to Live of the inner header to 831 increment across encapsulation/decapsulation cycle. This check is 832 also performed when doing initial encapsulation when a packet 833 comes to an ITR or PITR destined for a LISP site. 835 o The inner header Type of Service field (or the Traffic Class 836 field, in the case of IPv6) SHOULD be copied from the outer header 837 Type of Service field (with one caveat, see below). 839 Note if an ETR/PETR is also an ITR/PITR and choose to reencapsulate 840 after decapsulating, the net effect of this is that the new outer 841 header will carry the same Time to Live as the old outer header. 843 Copying the TTL serves two purposes: first, it preserves the distance 844 the host intended the packet to travel; second, and more importantly, 845 it provides for suppression of looping packets in the event there is 846 a loop of concatenated tunnels due to misconfiguration. See 847 Section 9.3 for TTL exception handling for traceroute packets. 849 The ECN field occupies bits 6 and 7 of both the IPv4 Type of Service 850 field and the IPv6 Traffic Class field [RFC3168]. The ECN field 851 requires special treatment in order to avoid discarding indications 852 of congestion [RFC3168]. ITR encapsulation MUST copy the 2-bit ECN 853 field from the inner header to the outer header. Re-encapsulation 854 MUST copy the 2-bit ECN field from the stripped outer header to the 855 new outer header. If the ECN field contains a congestion indication 856 codepoint (the value is '11', the Congestion Experienced (CE) 857 codepoint), then ETR decapsulation MUST copy the 2-bit ECN field from 858 the stripped outer header to the surviving inner header that is used 859 to forward the packet beyond the ETR. These requirements preserve 860 Congestion Experienced (CE) indications when a packet that uses ECN 861 traverses a LISP tunnel and becomes marked with a CE indication due 862 to congestion between the tunnel endpoints. 864 5.4. Dealing with Large Encapsulated Packets 866 This section proposes two mechanisms to deal with packets that exceed 867 the path MTU between the ITR and ETR. 869 It is left to the implementor to decide if the stateless or stateful 870 mechanism should be implemented. Both or neither can be used since 871 it is a local decision in the ITR regarding how to deal with MTU 872 issues, and sites can interoperate with differing mechanisms. 874 Both stateless and stateful mechanisms also apply to Reencapsulating 875 and Recursive Tunneling. So any actions below referring to an ITR 876 also apply to an TE-ITR. 878 5.4.1. A Stateless Solution to MTU Handling 880 An ITR stateless solution to handle MTU issues is described as 881 follows: 883 1. Define an architectural constant S for the maximum size of a 884 packet, in bytes, an ITR would like to receive from a source 885 inside of its site. 887 2. Define L to be the maximum size, in bytes, a packet of size S 888 would be after the ITR prepends the LISP header, UDP header, and 889 outer network layer header of size H. 891 3. Calculate: S + H = L. 893 When an ITR receives a packet from a site-facing interface and adds H 894 bytes worth of encapsulation to yield a packet size greater than L 895 bytes, it resolves the MTU issue by first splitting the original 896 packet into 2 equal-sized fragments. A LISP header is then prepended 897 to each fragment. The size of the encapsulated fragments is then 898 (S/2 + H), which is less than the ITR's estimate of the path MTU 899 between the ITR and its correspondent ETR. 901 When an ETR receives encapsulated fragments, it treats them as two 902 individually encapsulated packets. It strips the LISP headers then 903 forwards each fragment to the destination host of the destination 904 site. The two fragments are reassembled at the destination host into 905 the single IP datagram that was originated by the source host. 907 This behavior is performed by the ITR when the source host originates 908 a packet with the DF field of the IP header is set to 0. When the DF 909 field of the IP header is set to 1, or the packet is an IPv6 packet 910 originated by the source host, the ITR will drop the packet when the 911 size is greater than L, and sends an ICMP Too Big message to the 912 source with a value of S, where S is (L - H). 914 When the outer header encapsulation uses an IPv4 header, an 915 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 916 can be avoided. An implementation MAY set the DF bit in such headers 917 to 0 if it has good reason to believe there are unresolvable path MTU 918 issues between the sending ITR and the receiving ETR. 920 This specification recommends that L be defined as 1500. 922 5.4.2. A Stateful Solution to MTU Handling 924 An ITR stateful solution to handle MTU issues is described as follows 925 and was first introduced in [OPENLISP]: 927 1. The ITR will keep state of the effective MTU for each locator per 928 mapping cache entry. The effective MTU is what the core network 929 can deliver along the path between ITR and ETR. 931 2. When an IPv6 encapsulated packet or an IPv4 encapsulated packet 932 with DF bit set to 1, exceeds what the core network can deliver, 933 one of the intermediate routers on the path will send an ICMP Too 934 Big message to the ITR. The ITR will parse the ICMP message to 935 determine which locator is affected by the effective MTU change 936 and then record the new effective MTU value in the mapping cache 937 entry. 939 3. When a packet is received by the ITR from a source inside of the 940 site and the size of the packet is greater than the effective MTU 941 stored with the mapping cache entry associated with the 942 destination EID the packet is for, the ITR will send an ICMP Too 943 Big message back to the source. The packet size advertised by 944 the ITR in the ICMP Too Big message is the effective MTU minus 945 the LISP encapsulation length. 947 Even though this mechanism is stateful, it has advantages over the 948 stateless IP fragmentation mechanism, by not involving the 949 destination host with reassembly of ITR fragmented packets. 951 5.5. Using Virtualization and Segmentation with LISP 953 When multiple organizations inside of a LISP site are using private 954 addresses [RFC1918] as EID-prefixes, their address spaces MUST remain 955 segregated due to possible address duplication. An Instance ID in 956 the address encoding can aid in making the entire AFI based address 957 unique. See IANA Considerations Section 14.1 for details for 958 possible address encodings. 960 An Instance ID can be carried in a LISP encapsulated packet. An ITR 961 that prepends a LISP header, will copy a 24-bit value, used by the 962 LISP router to uniquely identify the address space. The value is 963 copied to the Instance ID field of the LISP header and the I-bit is 964 set to 1. 966 When an ETR decapsulates a packet, the Instance ID from the LISP 967 header is used as a table identifier to locate the forwarding table 968 to use for the inner destination EID lookup. 970 For example, a 802.1Q VLAN tag or VPN identifier could be used as a 971 24-bit Instance ID. 973 6. EID-to-RLOC Mapping 975 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats 977 The following new UDP packet types are used to retrieve EID-to-RLOC 978 mappings: 980 0 1 2 3 981 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 1 982 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 983 |Version| IHL |Type of Service| Total Length | 984 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 985 | Identification |Flags| Fragment Offset | 986 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 987 | Time to Live | Protocol = 17 | Header Checksum | 988 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 989 | Source Routing Locator | 990 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 991 | Destination Routing Locator | 992 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 993 / | Source Port | Dest Port | 994 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 995 \ | UDP Length | UDP Checksum | 996 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 997 | | 998 | LISP Message | 999 | | 1000 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1002 0 1 2 3 1003 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 1004 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1005 |Version| Traffic Class | Flow Label | 1006 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1007 | Payload Length | Next Header=17| Hop Limit | 1008 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1009 | | 1010 + + 1011 | | 1012 + Source Routing Locator + 1013 | | 1014 + + 1015 | | 1016 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1017 | | 1018 + + 1019 | | 1020 + Destination Routing Locator + 1021 | | 1022 + + 1023 | | 1024 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1025 / | Source Port | Dest Port | 1026 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1027 \ | UDP Length | UDP Checksum | 1028 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1029 | | 1030 | LISP Message | 1031 | | 1032 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1034 The LISP UDP-based messages are the Map-Request and Map-Reply 1035 messages. When a UDP Map-Request is sent, the UDP source port is 1036 chosen by the sender and the destination UDP port number is set to 1037 4342. When a UDP Map-Reply is sent, the source UDP port number is 1038 set to 4342 and the destination UDP port number is copied from the 1039 source port of either the Map-Request or the invoking data packet. 1040 Implementations MUST be prepared to accept packets when either the 1041 source port or destination UDP port is set to 4342 due to NATs 1042 changing port number values. 1044 The UDP Length field will reflect the length of the UDP header and 1045 the LISP Message payload. 1047 The UDP Checksum is computed and set to non-zero for Map-Request, 1048 Map-Reply, Map-Register and ECM control messages. It MUST be checked 1049 on receipt and if the checksum fails, the packet MUST be dropped. 1051 LISP-CONS [CONS] uses TCP to send LISP control messages. The format 1052 of control messages includes the UDP header so the checksum and 1053 length fields can be used to protect and delimit message boundaries. 1055 This main LISP specification is the authoritative source for message 1056 format definitions for the Map-Request and Map-Reply messages. 1058 6.1.1. LISP Packet Type Allocations 1060 This section will be the authoritative source for allocating LISP 1061 Type values. Current allocations are: 1063 Reserved: 0 b'0000' 1064 LISP Map-Request: 1 b'0001' 1065 LISP Map-Reply: 2 b'0010' 1066 LISP Map-Register: 3 b'0011' 1067 LISP Map-Notify: 4 b'0100' 1068 LISP Encapsulated Control Message: 8 b'1000' 1070 6.1.2. Map-Request Message Format 1072 0 1 2 3 1073 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 1074 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1075 |Type=1 |A|M|P|S|p| Reserved | IRC | Record Count | 1076 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1077 | Nonce . . . | 1078 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1079 | . . . Nonce | 1080 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1081 | Source-EID-AFI | Source EID Address ... | 1082 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1083 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... | 1084 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1085 | ... | 1086 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1087 | ITR-RLOC-AFI n | ITR-RLOC Address n ... | 1088 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1089 / | Reserved | EID mask-len | EID-prefix-AFI | 1090 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1091 \ | EID-prefix ... | 1092 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1093 | Map-Reply Record ... | 1094 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1095 | Mapping Protocol Data | 1096 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1098 Packet field descriptions: 1100 Type: 1 (Map-Request) 1102 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 1103 Requests sent by an ITR. 1105 M: When set, it indicates a Map-Reply Record segment is included in 1106 the Map-Request. 1108 P: This is the probe-bit which indicates that a Map-Request SHOULD be 1109 treated as a locator reachability probe. The receiver SHOULD 1110 respond with a Map-Reply with the probe-bit set, indicating the 1111 Map-Reply is a locator reachability probe reply, with the nonce 1112 copied from the Map-Request. See Section 6.3.2 for more details. 1114 S: This is the SMR bit. See Section 6.6.2 for details. 1116 p: This is the PITR bit. This bit is set to 1 when a PITR sends a 1117 Map-Request. 1119 Reserved: Set to 0 on transmission and ignored on receipt. 1121 IRC: This 5-bit field is the ITR-RLOC Count which encodes the 1122 additional number of (ITR-RLOC-AFI, ITR-RLOC Address) fields 1123 present in this message. At least one (ITR-RLOC-AFI, ITR-RLOC- 1124 Address) pair must always be encoded. Multiple ITR-RLOC Address 1125 fields are used so a Map-Replier can select which destination 1126 address to use for a Map-Reply. The IRC value ranges from 0 to 1127 31, and for a value of 1, there are 2 ITR-RLOC addresses encoded 1128 and so on up to 31 which encodes a total of 32 ITR-RLOC addresses. 1130 Record Count: The number of records in this Map-Request message. A 1131 record is comprised of the portion of the packet that is labeled 1132 'Rec' above and occurs the number of times equal to Record Count. 1133 For this version of the protocol, a receiver MUST accept and 1134 process Map-Requests that contain one or more records, but a 1135 sender MUST only send Map-Requests containing one record. Support 1136 for requesting multiple EIDs in a single Map-Request message will 1137 be specified in a future version of the protocol. 1139 Nonce: An 8-byte random value created by the sender of the Map- 1140 Request. This nonce will be returned in the Map-Reply. The 1141 security of the LISP mapping protocol depends critically on the 1142 strength of the nonce in the Map-Request message. The nonce 1143 SHOULD be generated by a properly seeded pseudo-random (or strong 1144 random) source. See [RFC4086] for advice on generating security- 1145 sensitive random data. 1147 Source-EID-AFI: Address family of the "Source EID Address" field. 1149 Source EID Address: This is the EID of the source host which 1150 originated the packet which is invoking this Map-Request. When 1151 Map-Requests are used for refreshing a map-cache entry or for 1152 RLOC-probing, an AFI value 0 is used and this field is of zero 1153 length. 1155 ITR-RLOC-AFI: Address family of the "ITR-RLOC Address" field that 1156 follows this field. 1158 ITR-RLOC Address: Used to give the ETR the option of selecting the 1159 destination address from any address family for the Map-Reply 1160 message. This address MUST be a routable RLOC address of the 1161 sender of the Map-Request message. 1163 EID mask-len: Mask length for EID prefix. 1165 EID-prefix-AFI: Address family of EID-prefix according to [RFC5226] 1167 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1168 address-family. When a Map-Request is sent by an ITR because a 1169 data packet is received for a destination where there is no 1170 mapping entry, the EID-prefix is set to the destination IP address 1171 of the data packet. And the 'EID mask-len' is set to 32 or 128 1172 for IPv4 or IPv6, respectively. When an xTR wants to query a site 1173 about the status of a mapping it already has cached, the EID- 1174 prefix used in the Map-Request has the same mask-length as the 1175 EID-prefix returned from the site when it sent a Map-Reply 1176 message. 1178 Map-Reply Record: When the M bit is set, this field is the size of a 1179 single "Record" in the Map-Reply format. This Map-Reply record 1180 contains the EID-to-RLOC mapping entry associated with the Source 1181 EID. This allows the ETR which will receive this Map-Request to 1182 cache the data if it chooses to do so. 1184 Mapping Protocol Data: See [CONS] for details. This field is 1185 optional and present when the UDP length indicates there is enough 1186 space in the packet to include it. 1188 6.1.3. EID-to-RLOC UDP Map-Request Message 1190 A Map-Request is sent from an ITR when it needs a mapping for an EID, 1191 wants to test an RLOC for reachability, or wants to refresh a mapping 1192 before TTL expiration. For the initial case, the destination IP 1193 address used for the Map-Request is the destination-EID from the 1194 packet which had a mapping cache lookup failure. For the latter 2 1195 cases, the destination IP address used for the Map-Request is one of 1196 the RLOC addresses from the locator-set of the map cache entry. The 1197 source address is either an IPv4 or IPv6 RLOC address depending if 1198 the Map-Request is using an IPv4 versus IPv6 header, respectively. 1199 In all cases, the UDP source port number for the Map-Request message 1200 is an ITR/PITR selected 16-bit value and the UDP destination port 1201 number is set to the well-known destination port number 4342. A 1202 successful Map-Reply updates the cached set of RLOCs associated with 1203 the EID prefix range. 1205 One or more Map-Request (ITR-RLOC-AFI, ITR-RLOC-Address) fields MUST 1206 be filled in by the ITR. The number of fields (minus 1) encoded MUST 1207 be placed in the IRC field. The ITR MAY include all locally 1208 configured locators in this list or just provide one locator address 1209 from each address family it supports. If the ITR erroneously 1210 provides no ITR-RLOC addresses, the Map-Replier MUST drop the Map- 1211 Request. 1213 Map-Requests can also be LISP encapsulated using UDP destination port 1214 4342 with a LISP type value set to "Encapsulated Control Message", 1215 when sent from an ITR to a Map-Resolver. Likewise, Map-Requests are 1216 LISP encapsulated the same way from a Map-Server to an ETR. Details 1217 on encapsulated Map-Requests and Map-Resolvers can be found in 1218 [LISP-MS]. 1220 Map-Requests MUST be rate-limited. It is recommended that a Map- 1221 Request for the same EID-prefix be sent no more than once per second. 1223 An ITR that is configured with mapping database information (i.e. it 1224 is also an ETR) may optionally include those mappings in a Map- 1225 Request. When an ETR configured to accept and verify such 1226 "piggybacked" mapping data receives such a Map-Request and it does 1227 not have this mapping in the map-cache, it may originate a "verifying 1228 Map-Request", addressed to the map-requesting ITR. If the ETR has a 1229 map-cache entry that matches the "piggybacked" EID and the RLOC is in 1230 the locator-set for the entry, then it may send the "verifying Map- 1231 Request" directly to the originating Map-Request source. If the RLOC 1232 is not in the locator-set, then the ETR MUST send the "verifying Map- 1233 Request" to the "piggybacked" EID. Doing this forces the "verifying 1234 Map-Request" to go through the mapping database system to reach the 1235 authoritative source of information about that EID, guarding against 1236 RLOC-spoofing in in the "piggybacked" mapping data. 1238 6.1.4. Map-Reply Message Format 1240 0 1 2 3 1241 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 1242 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1243 |Type=2 |P|E|S| Reserved | Record Count | 1244 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1245 | Nonce . . . | 1246 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1247 | . . . Nonce | 1248 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1249 | | Record TTL | 1250 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1251 R | Locator Count | EID mask-len | ACT |A| Reserved | 1252 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1253 c | Rsvd | Map-Version Number | EID-AFI | 1254 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1255 r | EID-prefix | 1256 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1257 | /| Priority | Weight | M Priority | M Weight | 1258 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1259 | o | Unused Flags |L|p|R| Loc-AFI | 1260 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1261 | \| Locator | 1262 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1263 | Mapping Protocol Data | 1264 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1266 Packet field descriptions: 1268 Type: 2 (Map-Reply) 1270 P: This is the probe-bit which indicates that the Map-Reply is in 1271 response to a locator reachability probe Map-Request. The nonce 1272 field MUST contain a copy of the nonce value from the original 1273 Map-Request. See Section 6.3.2 for more details. 1275 E: Indicates that the ETR which sends this Map-Reply message is 1276 advertising that the site is enabled for the Echo-Nonce locator 1277 reachability algorithm. See Section 6.3.1 for more details. 1279 S: This is the Security bit. When set to 1 the field following the 1280 Mapping Protocol Data field will have the following format. The 1281 detailed format of the Authentication Data Content field can be 1282 found in [LISP-SEC] when AD Type is equal to 1. 1284 0 1 2 3 1285 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 1286 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1287 | AD Type | Authentication Data Content . . . | 1288 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1290 Reserved: Set to 0 on transmission and ignored on receipt. 1292 Record Count: The number of records in this reply message. A record 1293 is comprised of that portion of the packet labeled 'Record' above 1294 and occurs the number of times equal to Record count. 1296 Nonce: A 24-bit value set in a Data-Probe packet or a 64-bit value 1297 from the Map-Request is echoed in this Nonce field of the Map- 1298 Reply. 1300 Record TTL: The time in minutes the recipient of the Map-Reply will 1301 store the mapping. If the TTL is 0, the entry SHOULD be removed 1302 from the cache immediately. If the value is 0xffffffff, the 1303 recipient can decide locally how long to store the mapping. 1305 Locator Count: The number of Locator entries. A locator entry 1306 comprises what is labeled above as 'Loc'. The locator count can 1307 be 0 indicating there are no locators for the EID-prefix. 1309 EID mask-len: Mask length for EID prefix. 1311 ACT: This 3-bit field describes negative Map-Reply actions. These 1312 bits are used only when the 'Locator Count' field is set to 0. 1313 The action bits are encoded only in Map-Reply messages. The 1314 actions defined are used by an ITR or PTR when a destination EID 1315 matches a negative mapping cache entry. Unassigned values should 1316 cause a map-cache entry to be created and, when packets match this 1317 negative cache entry, they will be dropped. The current assigned 1318 values are: 1320 (0) No-Action: The map-cache is kept alive and packet 1321 encapsulation occurs. 1323 (1) Natively-Forward: The packet is not encapsulated or dropped 1324 but natively forwarded. 1326 (2) Send-Map-Request: The packet invokes sending a Map-Request. 1328 (3) Drop: A packet that matches this map-cache entry is dropped. 1330 A: The Authoritative bit, when sent by a UDP-based message is always 1331 set to 1 by an ETR. See [CONS] for TCP-based Map-Replies. When a 1332 Map-Server is proxy Map-Replying [LISP-MS] for a LISP site, the 1333 Authoritative bit is set to 0. This indicates to requesting ITRs 1334 that the Map-Reply was not originated by a LISP node managed at 1335 the site that owns the EID-prefix. 1337 Map-Version Number: When this 12-bit value is non-zero the Map-Reply 1338 sender is informing the ITR what the version number is for the 1339 EID-record contained in the Map-Reply. The ETR can allocate this 1340 number internally but MUST coordinate this value with other ETRs 1341 for the site. When this value is 0, there is no versioning 1342 information conveyed. The Map-Version Number can be included in 1343 Map-Request and Map-Register messages. See Section 6.6.3 for more 1344 details. 1346 EID-AFI: Address family of EID-prefix according to [RFC5226]. 1348 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1349 address-family. 1351 Priority: each RLOC is assigned a unicast priority. Lower values 1352 are more preferable. When multiple RLOCs have the same priority, 1353 they may be used in a load-split fashion. A value of 255 means 1354 the RLOC MUST NOT be used for unicast forwarding. 1356 Weight: when priorities are the same for multiple RLOCs, the weight 1357 indicates how to balance unicast traffic between them. Weight is 1358 encoded as a relative weight of total unicast packets that match 1359 the mapping entry. If a non-zero weight value is used for any 1360 RLOC, then all RLOCs MUST use a non-zero weight value and then the 1361 sum of all weight values MUST equal 100. If a zero value is used 1362 for any RLOC weight, then all weights MUST be zero and the 1363 receiver of the Map-Reply will decide how to load-split traffic. 1364 See Section 6.5 for a suggested hash algorithm to distribute load 1365 across locators with same priority and equal weight values. 1367 M Priority: each RLOC is assigned a multicast priority used by an 1368 ETR in a receiver multicast site to select an ITR in a source 1369 multicast site for building multicast distribution trees. A value 1370 of 255 means the RLOC MUST NOT be used for joining a multicast 1371 distribution tree. 1373 M Weight: when priorities are the same for multiple RLOCs, the 1374 weight indicates how to balance building multicast distribution 1375 trees across multiple ITRs. The weight is encoded as a relative 1376 weight of total number of trees built to the source site 1377 identified by the EID-prefix. If a non-zero weight value is used 1378 for any RLOC, then all RLOCs MUST use a non-zero weight value and 1379 then the sum of all weight values MUST equal 100. If a zero value 1380 is used for any RLOC weight, then all weights MUST be zero and the 1381 receiver of the Map-Reply will decide how to distribute multicast 1382 state across ITRs. 1384 Unused Flags: set to 0 when sending and ignored on receipt. 1386 L: when this bit is set, the locator is flagged as a local locator to 1387 the ETR that is sending the Map-Reply. When a Map-Server is doing 1388 proxy Map-Replying [LISP-MS] for a LISP site, the L bit is set to 1389 0 for all locators in this locator-set. 1391 p: when this bit is set, an ETR informs the RLOC-probing ITR that the 1392 locator address, for which this bit is set, is the one being RLOC- 1393 probed and may be different from the source address of the Map- 1394 Reply. An ITR that RLOC-probes a particular locator, MUST use 1395 this locator for retrieving the data structure used to store the 1396 fact that the locator is reachable. The "p" bit is set for a 1397 single locator in the same locator set. If an implementation sets 1398 more than one "p" bit erroneously, the receiver of the Map-Reply 1399 MUST select the first locator. The "p" bit MUST NOT be set for 1400 locator-set records sent in Map-Request and Map-Register messages. 1402 R: set when the sender of a Map-Reply has a route to the locator in 1403 the locator data record. This receiver may find this useful to 1404 know when determining if the locator is reachable from the 1405 receiver. See also Section 6.4 for another way the R-bit may be 1406 used. 1408 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1409 assigned to an ETR. Note that the destination RLOC address MAY be 1410 an anycast address. A source RLOC can be an anycast address as 1411 well. The source or destination RLOC MUST NOT be the broadcast 1412 address (255.255.255.255 or any subnet broadcast address known to 1413 the router), and MUST NOT be a link-local multicast address. The 1414 source RLOC MUST NOT be a multicast address. The destination RLOC 1415 SHOULD be a multicast address if it is being mapped from a 1416 multicast destination EID. 1418 Mapping Protocol Data: See [CONS] or [ALT] for details. This field 1419 is optional and present when the UDP length indicates there is 1420 enough space in the packet to include it. 1422 6.1.5. EID-to-RLOC UDP Map-Reply Message 1424 A Map-Reply returns an EID-prefix with a prefix length that is less 1425 than or equal to the EID being requested. The EID being requested is 1426 either from the destination field of an IP header of a Data-Probe or 1427 the EID record of a Map-Request. The RLOCs in the Map-Reply are 1428 globally-routable IP addresses of all ETRs for the LISP site. Each 1429 RLOC conveys status reachability but does not convey path 1430 reachability from a requesters perspective. Separate testing of path 1431 reachability is required, See Section 6.3 for details. 1433 Note that a Map-Reply may contain different EID-prefix granularity 1434 (prefix + length) than the Map-Request which triggers it. This might 1435 occur if a Map-Request were for a prefix that had been returned by an 1436 earlier Map-Reply. In such a case, the requester updates its cache 1437 with the new prefix information and granularity. For example, a 1438 requester with two cached EID-prefixes that are covered by a Map- 1439 Reply containing one, less-specific prefix, replaces the entry with 1440 the less-specific EID-prefix. Note that the reverse, replacement of 1441 one less-specific prefix with multiple more-specific prefixes, can 1442 also occur but not by removing the less-specific prefix rather by 1443 adding the more-specific prefixes which during a lookup will override 1444 the less-specific prefix. 1446 When an ETR is configured with overlapping EID-prefixes, a Map- 1447 Request with an EID that longest matches any EID-prefix MUST be 1448 returned in a single Map-Reply message. For instance, if an ETR had 1449 database mapping entries for EID-prefixes: 1451 10.0.0.0/8 1452 10.1.0.0/16 1453 10.1.1.0/24 1454 10.1.2.0/24 1456 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record 1457 count of 1 to be returned with a mapping record EID-prefix of 1458 10.1.1.0/24. 1460 A Map-Request for EID 10.1.5.5, would cause a Map-Reply with a record 1461 count of 3 to be returned with mapping records for EID-prefixes 1462 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24. 1464 Note that not all overlapping EID-prefixes need to be returned, only 1465 the more specifics (note in the second example above 10.0.0.0/8 was 1466 not returned for requesting EID 10.1.5.5) entries for the matching 1467 EID-prefix of the requesting EID. When more than one EID-prefix is 1468 returned, all SHOULD use the same Time-to-Live value so they can all 1469 time out at the same time. When a more specific EID-prefix is 1470 received later, its Time-to-Live value in the Map-Reply record can be 1471 stored even when other less specifics exist. When a less specific 1472 EID-prefix is received later, its map-cache expiration time SHOULD be 1473 set to the minimum expiration time of any more specific EID-prefix in 1474 the map-cache. 1476 Map-Replies SHOULD be sent for an EID-prefix no more often than once 1477 per second to the same requesting router. For scalability, it is 1478 expected that aggregation of EID addresses into EID-prefixes will 1479 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1480 prefix range thereby reducing the number of Map-Request messages. 1482 Map-Reply records can have an empty locator-set. A negative Map- 1483 Reply is a Map-Reply with an empty locator-set. Negative Map-Replies 1484 convey special actions by the sender to the ITR or PTR which have 1485 solicited the Map-Reply. There are two primary applications for 1486 Negative Map-Replies. The first is for a Map-Resolver to instruct an 1487 ITR or PTR when a destination is for a LISP site versus a non-LISP 1488 site. And the other is to source quench Map-Requests which are sent 1489 for non-allocated EIDs. 1491 For each Map-Reply record, the list of locators in a locator-set MUST 1492 appear in the same order for each ETR that originates a Map-Reply 1493 message. The locator-set MUST be sorted in order of ascending IP 1494 address where an IPv4 locator address is considered numerically 'less 1495 than' an IPv6 locator address. 1497 When sending a Map-Reply message, the destination address is copied 1498 from the one of the ITR-RLOC fields from the Map-Request. The ETR 1499 can choose a locator address from one of the address families it 1500 supports. For Data-Probes, the destination address of the Map-Reply 1501 is copied from the source address of the Data-Probe message which is 1502 invoking the reply. The source address of the Map-Reply is one of 1503 the local locator addresses listed in the locator-set of any mapping 1504 record in the message and SHOULD be chosen to allow uRPF checks to 1505 succeed in the upstream service provider. The destination port of a 1506 Map-Reply message is copied from the source port of the Map-Request 1507 or Data-Probe and the source port of the Map-Reply message is set to 1508 the well-known UDP port 4342. 1510 6.1.5.1. Traffic Redirection with Coarse EID-Prefixes 1512 When an ETR is misconfigured or compromised, it could return coarse 1513 EID-prefixes in Map-Reply messages it sends. The EID-prefix could 1514 cover EID-prefixes which are allocated to other sites redirecting 1515 their traffic to the locators of the compromised site. 1517 To solve this problem, there are two basic solutions that could be 1518 used. The first is to have Map-Servers proxy-map-reply on behalf of 1519 ETRs so their registered EID-prefixes are the ones returned in Map- 1520 Replies. Since the interaction between an ETR and Map-Server is 1521 secured with shared-keys, it is more difficult for an ETR to 1522 misbehave. The second solution is to have ITRs and PTRs cache EID- 1523 prefixes with mask-lengths that are greater than or equal to a 1524 configured prefix length. This limits the damage to a specific width 1525 of any EID-prefix advertised, but needs to be coordinated with the 1526 allocation of site prefixes. These solutions can be used 1527 independently or at the same time. 1529 At the time of this writing, other approaches are being considered 1530 and researched. 1532 6.1.6. Map-Register Message Format 1534 The usage details of the Map-Register message can be found in 1535 specification [LISP-MS]. This section solely defines the message 1536 format. 1538 The message is sent in UDP with a destination UDP port of 4342 and a 1539 randomly selected UDP source port number. 1541 The Map-Register message format is: 1543 0 1 2 3 1544 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 1545 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1546 |Type=3 |P| Reserved |M| Record Count | 1547 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1548 | Nonce . . . | 1549 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1550 | . . . Nonce | 1551 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1552 | Key ID | Authentication Data Length | 1553 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1554 ~ Authentication Data ~ 1555 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1556 | | Record TTL | 1557 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1558 R | Locator Count | EID mask-len | ACT |A| Reserved | 1559 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1560 c | Rsvd | Map-Version Number | EID-AFI | 1561 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1562 r | EID-prefix | 1563 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1564 | /| Priority | Weight | M Priority | M Weight | 1565 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1566 | o | Unused Flags |L|p|R| Loc-AFI | 1567 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1568 | \| Locator | 1569 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1571 Packet field descriptions: 1573 Type: 3 (Map-Register) 1575 P: This is the proxy-map-reply bit, when set to 1 an ETR sends a Map- 1576 Register message requesting for the Map-Server to proxy Map-Reply. 1577 The Map-Server will send non-authoritative Map-Replies on behalf 1578 of the ETR. Details on this usage will be provided in a future 1579 version of this draft. 1581 Reserved: Set to 0 on transmission and ignored on receipt. 1583 M: This is the want-map-notify bit, when set to 1 an ETR is 1584 requesting for a Map-Notify message to be returned in response to 1585 sending a Map-Register message. The Map-Notify message sent by a 1586 Map-Server is used to an acknowledge receipt of a Map-Register 1587 message. 1589 Record Count: The number of records in this Map-Register message. A 1590 record is comprised of that portion of the packet labeled 'Record' 1591 above and occurs the number of times equal to Record count. 1593 Nonce: This 8-byte Nonce field is set to 0 in Map-Register messages. 1595 Key ID: A configured ID to find the configured Message 1596 Authentication Code (MAC) algorithm and key value used for the 1597 authentication function. 1599 Authentication Data Length: The length in bytes of the 1600 Authentication Data field that follows this field. The length of 1601 the Authentication Data field is dependent on the Message 1602 Authentication Code (MAC) algorithm used. The length field allows 1603 a device that doesn't know the MAC algorithm to correctly parse 1604 the packet. 1606 Authentication Data: The message digest used from the output of the 1607 Message Authentication Code (MAC) algorithm. The entire Map- 1608 Register payload is authenticated with this field preset to 0. 1609 After the MAC is computed, it is placed in this field. 1610 Implementations of this specification MUST include support for 1611 HMAC-SHA-1-96 [RFC2404] and support for HMAC-SHA-128-256 [RFC4634] 1612 is recommended. 1614 The definition of the rest of the Map-Register can be found in the 1615 Map-Reply section. 1617 6.1.7. Map-Notify Message Format 1619 The usage details of the Map-Notify message can be found in 1620 specification [LISP-MS]. This section solely defines the message 1621 format. 1623 The message is sent inside a UDP packet with a source UDP port equal 1624 to 4342 and a destination port equal to the source port from the Map- 1625 Register message this Map-Notify message is responding to. 1627 The Map-Notify message format is: 1629 0 1 2 3 1630 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 1631 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1632 |Type=4 | Reserved | Record Count | 1633 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1634 | Nonce . . . | 1635 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1636 | . . . Nonce | 1637 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1638 | Key ID | Authentication Data Length | 1639 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1640 ~ Authentication Data ~ 1641 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1642 | | Record TTL | 1643 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1644 R | Locator Count | EID mask-len | ACT |A| Reserved | 1645 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1646 c | Rsvd | Map-Version Number | EID-AFI | 1647 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1648 r | EID-prefix | 1649 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1650 | /| Priority | Weight | M Priority | M Weight | 1651 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1652 | o | Unused Flags |L|p|R| Loc-AFI | 1653 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1654 | \| Locator | 1655 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1657 Packet field descriptions: 1659 Type: 4 (Map-Notify) 1661 The Map-Notify message has the same contents as a Map-Register 1662 message. See Map-Register section for field descriptions. 1664 6.1.8. Encapsulated Control Message Format 1666 An Encapsulated Control Message is used to encapsulate control 1667 packets sent between xTRs and the mapping database system described 1668 in [LISP-MS]. 1670 0 1 2 3 1671 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 1672 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1673 / | IPv4 or IPv6 Header | 1674 OH | (uses RLOC addresses) | 1675 \ | | 1676 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1677 / | Source Port = xxxx | Dest Port = 4342 | 1678 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1679 \ | UDP Length | UDP Checksum | 1680 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1681 LH |Type=8 |S| Reserved | 1682 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1683 / | IPv4 or IPv6 Header | 1684 IH | (uses RLOC or EID addresses) | 1685 \ | | 1686 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1687 / | Source Port = xxxx | Dest Port = yyyy | 1688 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1689 \ | UDP Length | UDP Checksum | 1690 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1691 LCM | LISP Control Message | 1692 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1694 Packet header descriptions: 1696 OH: The outer IPv4 or IPv6 header which uses RLOC addresses in the 1697 source and destination header address fields. 1699 UDP: The outer UDP header with destination port 4342. The source 1700 port is randomly allocated. The checksum field MUST be non-zero. 1702 LH: Type 8 is defined to be a "LISP Encapsulated Control Message" 1703 and what follows is either an IPv4 or IPv6 header as encoded by 1704 the first 4 bits after the reserved field. 1706 S: This is the Security bit. When set to 1 the field following the 1707 Reserved field will have the following format. The detailed 1708 format of the Authentication Data Content field can be found in 1709 [LISP-SEC] when AD Type is equal to 1. 1711 0 1 2 3 1712 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 1713 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1714 | AD Type | Authentication Data Content . . . | 1715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1717 IH: The inner IPv4 or IPv6 header which can use either RLOC or EID 1718 addresses in the header address fields. When a Map-Request is 1719 encapsulated in this packet format the destination address in this 1720 header is an EID. 1722 UDP: The inner UDP header where the port assignments depends on the 1723 control packet being encapsulated. When the control packet is a 1724 Map-Request or Map-Register, the source port is ITR/PITR selected 1725 and the destination port is 4342. When the control packet is a 1726 Map-Reply, the source port is 4342 and the destination port is 1727 assigned from the source port of the invoking Map-Request. Port 1728 number 4341 MUST NOT be assigned to either port. The checksum 1729 field MUST be non-zero. 1731 LCM: The format is one of the control message formats described in 1732 this section. At this time, only Map-Request messages and PIM 1733 Join-Prune messages [MLISP] are allowed to be encapsulated. 1734 Encapsulating other types of LISP control messages are for further 1735 study. When Map-Requests are sent for RLOC-probing purposes (i.e 1736 the probe-bit is set), they MUST NOT be sent inside Encapsulated 1737 Control Messages. 1739 6.2. Routing Locator Selection 1741 Both client-side and server-side may need control over the selection 1742 of RLOCs for conversations between them. This control is achieved by 1743 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1744 messages. Alternatively, RLOC information may be gleaned from 1745 received tunneled packets or EID-to-RLOC Map-Request messages. 1747 The following enumerates different scenarios for choosing RLOCs and 1748 the controls that are available: 1750 o Server-side returns one RLOC. Client-side can only use one RLOC. 1751 Server-side has complete control of the selection. 1753 o Server-side returns a list of RLOC where a subset of the list has 1754 the same best priority. Client can only use the subset list 1755 according to the weighting assigned by the server-side. In this 1756 case, the server-side controls both the subset list and load- 1757 splitting across its members. The client-side can use RLOCs 1758 outside of the subset list if it determines that the subset list 1759 is unreachable (unless RLOCs are set to a Priority of 255). Some 1760 sharing of control exists: the server-side determines the 1761 destination RLOC list and load distribution while the client-side 1762 has the option of using alternatives to this list if RLOCs in the 1763 list are unreachable. 1765 o Server-side sets weight of 0 for the RLOC subset list. In this 1766 case, the client-side can choose how the traffic load is spread 1767 across the subset list. Control is shared by the server-side 1768 determining the list and the client determining load distribution. 1769 Again, the client can use alternative RLOCs if the server-provided 1770 list of RLOCs are unreachable. 1772 o Either side (more likely on the server-side ETR) decides not to 1773 send a Map-Request. For example, if the server-side ETR does not 1774 send Map-Requests, it gleans RLOCs from the client-side ITR, 1775 giving the client-side ITR responsibility for bidirectional RLOC 1776 reachability and preferability. Server-side ETR gleaning of the 1777 client-side ITR RLOC is done by caching the inner header source 1778 EID and the outer header source RLOC of received packets. The 1779 client-side ITR controls how traffic is returned and can alternate 1780 using an outer header source RLOC, which then can be added to the 1781 list the server-side ETR uses to return traffic. Since no 1782 Priority or Weights are provided using this method, the server- 1783 side ETR MUST assume each client-side ITR RLOC uses the same best 1784 Priority with a Weight of zero. In addition, since EID-prefix 1785 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1786 on tunnel routers can grow to be very large. 1788 o A "gleaned" map-cache entry, one learned from the source RLOC of a 1789 received encapsulated packet, is only stored and used for a few 1790 seconds, pending verification. Verification is performed by 1791 sending a Map-Request to the source EID (the inner header IP 1792 source address) of the received encapsulated packet. A reply to 1793 this "verifying Map-Request" is used to fully populate the map- 1794 cache entry for the "gleaned" EID and is stored and used for the 1795 time indicated from the TTL field of a received Map-Reply. When a 1796 verified map-cache entry is stored, data gleaning no longer occurs 1797 for subsequent packets which have a source EID that matches the 1798 EID-prefix of the verified entry. 1800 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1801 reachable when the R-bit for the locator record is set to 1. Neither 1802 the information contained in a Map-Reply or that stored in the 1803 mapping database system provides reachability information for RLOCs. 1804 Note that reachability is not part of the mapping system and is 1805 determined using one or more of the Routing Locator Reachability 1806 Algorithms described in the next section. 1808 6.3. Routing Locator Reachability 1810 Several mechanisms for determining RLOC reachability are currently 1811 defined: 1813 1. An ETR may examine the Loc-Status-Bits in the LISP header of an 1814 encapsulated data packet received from an ITR. If the ETR is 1815 also acting as an ITR and has traffic to return to the original 1816 ITR site, it can use this status information to help select an 1817 RLOC. 1819 2. An ITR may receive an ICMP Network or ICMP Host Unreachable 1820 message for an RLOC it is using. This indicates that the RLOC is 1821 likely down. 1823 3. An ITR which participates in the global routing system can 1824 determine that an RLOC is down if no BGP RIB route exists that 1825 matches the RLOC IP address. 1827 4. An ITR may receive an ICMP Port Unreachable message from a 1828 destination host. This occurs if an ITR attempts to use 1829 interworking [INTERWORK] and LISP-encapsulated data is sent to a 1830 non-LISP-capable site. 1832 5. An ITR may receive a Map-Reply from a ETR in response to a 1833 previously sent Map-Request. The RLOC source of the Map-Reply is 1834 likely up since the ETR was able to send the Map-Reply to the 1835 ITR. 1837 6. When an ETR receives an encapsulated packet from an ITR, the 1838 source RLOC from the outer header of the packet is likely up. 1840 7. An ITR/ETR pair can use the Locator Reachability Algorithms 1841 described in this section, namely Echo-Noncing or RLOC-Probing. 1843 When determining Locator up/down reachability by examining the Loc- 1844 Status-Bits from the LISP encapsulated data packet, an ETR will 1845 receive up to date status from an encapsulating ITR about 1846 reachability for all ETRs at the site. CE-based ITRs at the source 1847 site can determine reachability relative to each other using the site 1848 IGP as follows: 1850 o Under normal circumstances, each ITR will advertise a default 1851 route into the site IGP. 1853 o If an ITR fails or if the upstream link to its PE fails, its 1854 default route will either time-out or be withdrawn. 1856 Each ITR can thus observe the presence or lack of a default route 1857 originated by the others to determine the Locator Status Bits it sets 1858 for them. 1860 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1861 Loc-Status-Bits in a LISP encapsulated packet are numbered from 0 to 1862 n-1 starting with the least significant bit. For example, if an RLOC 1863 listed in the 3rd position of the Map-Reply goes down (ordinal value 1864 2), then all ITRs at the site will clear the 3rd least significant 1865 bit (xxxx x0xx) of the Loc-Status-Bits field for the packets they 1866 encapsulate. 1868 When an ETR decapsulates a packet, it will check for any change in 1869 the Loc-Status-Bits field. When a bit goes from 1 to 0, the ETR will 1870 refrain from encapsulating packets to an RLOC that is indicated as 1871 down. It will only resume using that RLOC if the corresponding Loc- 1872 Status-Bit returns to a value of 1. Loc-Status-Bits are associated 1873 with a locator-set per EID-prefix. Therefore, when a locator becomes 1874 unreachable, the Loc-Status-Bit that corresponds to that locator's 1875 position in the list returned by the last Map-Reply will be set to 1876 zero for that particular EID-prefix. 1878 When ITRs at the site are not deployed in CE routers, the IGP can 1879 still be used to determine the reachability of Locators provided they 1880 are injected into the IGP. This is typically done when a /32 address 1881 is configured on a loopback interface. 1883 When ITRs receive ICMP Network or Host Unreachable messages as a 1884 method to determine unreachability, they will refrain from using 1885 Locators which are described in Locator lists of Map-Replies. 1886 However, using this approach is unreliable because many network 1887 operators turn off generation of ICMP Unreachable messages. 1889 If an ITR does receive an ICMP Network or Host Unreachable message, 1890 it MAY originate its own ICMP Unreachable message destined for the 1891 host that originated the data packet the ITR encapsulated. 1893 Also, BGP-enabled ITRs can unilaterally examine the BGP RIB to see if 1894 a locator address from a locator-set in a mapping entry matches a 1895 prefix. If it does not find one and BGP is running in the Default 1896 Free Zone (DFZ), it can decide to not use the locator even though the 1897 Loc-Status-Bits indicate the locator is up. In this case, the path 1898 from the ITR to the ETR that is assigned the locator is not 1899 available. More details are in [LOC-ID-ARCH]. 1901 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 1902 Reply is returned, reachability of the Locator has been determined. 1903 Obviously, sending such probes increases the number of control 1904 messages originated by tunnel routers for active flows, so Locators 1905 are assumed to be reachable when they are advertised. 1907 This assumption does create a dependency: Locator unreachability is 1908 detected by the receipt of ICMP Host Unreachable messages. When an 1909 Locator has been determined to be unreachable, it is not used for 1910 active traffic; this is the same as if it were listed in a Map-Reply 1911 with priority 255. 1913 The ITR can test the reachability of the unreachable Locator by 1914 sending periodic Requests. Both Requests and Replies MUST be rate- 1915 limited. Locator reachability testing is never done with data 1916 packets since that increases the risk of packet loss for end-to-end 1917 sessions. 1919 When an ETR decapsulates a packet, it knows that it is reachable from 1920 the encapsulating ITR because that is how the packet arrived. In 1921 most cases, the ETR can also reach the ITR but cannot assume this to 1922 be true due to the possibility of path asymmetry. In the presence of 1923 unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD NOT 1924 use the lack of return traffic as an indication that the ETR is 1925 unreachable. Instead, it MUST use an alternate mechanisms to 1926 determine reachability. 1928 6.3.1. Echo Nonce Algorithm 1930 When data flows bidirectionally between locators from different 1931 sites, a data-plane mechanism called "nonce echoing" can be used to 1932 determine reachability between an ITR and ETR. When an ITR wants to 1933 solicit a nonce echo, it sets the N and E bits and places a 24-bit 1934 nonce in the LISP header of the next encapsulated data packet. 1936 When this packet is received by the ETR, the encapsulated packet is 1937 forwarded as normal. When the ETR next sends a data packet to the 1938 ITR, it includes the nonce received earlier with the N bit set and E 1939 bit cleared. The ITR sees this "echoed nonce" and knows the path to 1940 and from the ETR is up. 1942 The ITR will set the E-bit and N-bit for every packet it sends while 1943 in echo-nonce-request state. The time the ITR waits to process the 1944 echoed nonce before it determines the path is unreachable is variable 1945 and a choice left for the implementation. 1947 If the ITR is receiving packets from the ETR but does not see the 1948 nonce echoed while being in echo-nonce-request state, then the path 1949 to the ETR is unreachable. This decision may be overridden by other 1950 locator reachability algorithms. Once the ITR determines the path to 1951 the ETR is down it can switch to another locator for that EID-prefix. 1953 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 1954 be implementing both ITR and ETR functionality for the echo nonce 1955 mechanism to operate. 1957 The ITR and ETR may both go into echo-nonce-request state at the same 1958 time. The number of packets sent or the time during which echo nonce 1959 requests are sent is an implementation specific setting. However, 1960 when an ITR is in echo-nonce-request state, it can echo the ETR's 1961 nonce in the next set of packets that it encapsulates and then 1962 subsequently, continue sending echo-nonce-request packets. 1964 This mechanism does not completely solve the forward path 1965 reachability problem as traffic may be unidirectional. That is, the 1966 ETR receiving traffic at a site may not be the same device as an ITR 1967 which transmits traffic from that site or the site to site traffic is 1968 unidirectional so there is no ITR returning traffic. 1970 The echo-nonce algorithm is bilateral. That is, if one side sets the 1971 E-bit and the other side is not enabled for echo-noncing, then the 1972 echoing of the nonce does not occur and the requesting side may 1973 regard the locator unreachable erroneously. An ITR SHOULD only set 1974 the E-bit in a encapsulated data packet when it knows the ETR is 1975 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 1976 Reply message. 1978 Note that other locator reachability mechanisms are being researched 1979 and can be used to compliment or even override the Echo Nonce 1980 Algorithm. See next section for an example of control-plane probing. 1982 6.3.2. RLOC Probing Algorithm 1984 RLOC Probing is a method that an ITR or PTR can use to determine the 1985 reachability status of one or more locators that it has cached in a 1986 map-cache entry. The probe-bit of the Map-Request and Map-Reply 1987 messages are used for RLOC Probing. 1989 RLOC probing is done in the control-plane on a timer basis where an 1990 ITR or PTR will originate a Map-Request destined to a locator address 1991 from one of its own locator addresses. A Map-Request used as an 1992 RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or on the 1993 ALT like one would when soliciting mapping data. The EID record 1994 encoded in the Map-Request is the EID-prefix of the map-cache entry 1995 cached by the ITR or PTR. The ITR may include a mapping data record 1996 for its own database mapping information which contains the local 1997 EID-prefixes and RLOCs for its site. 1999 When an ETR receives a Map-Request message with the probe-bit set, it 2000 returns a Map-Reply with the probe-bit set. The source address of 2001 the Map-Reply is set from the destination address of the Map-Request 2002 and the destination address of the Map-Reply is set from the source 2003 address of the Map-Request. The Map-Reply SHOULD contain mapping 2004 data for the EID-prefix contained in the Map-Request. This provides 2005 the opportunity for the ITR or PTR, which sent the RLOC-probe to get 2006 mapping updates if there were changes to the ETR's database mapping 2007 entries. 2009 There are advantages and disadvantages of RLOC Probing. The greatest 2010 benefit of RLOC Probing is that it can handle many failure scenarios 2011 allowing the ITR to determine when the path to a specific locator is 2012 reachable or has become unreachable, thus providing a robust 2013 mechanism for switching to using another locator from the cached 2014 locator. RLOC Probing can also provide rough RTT estimates between a 2015 pair of locators which can be useful for network management purposes 2016 as well as for selecting low delay paths. The major disadvantage of 2017 RLOC Probing is in the number of control messages required and the 2018 amount of bandwidth used to obtain those benefits, especially if the 2019 requirement for failure detection times are very small. 2021 Continued research and testing will attempt to characterize the 2022 tradeoffs of failure detection times versus message overhead. 2024 6.4. EID Reachability within a LISP Site 2026 A site may be multihomed using two or more ETRs. The hosts and 2027 infrastructure within a site will be addressed using one or more EID 2028 prefixes that are mapped to the RLOCs of the relevant ETRs in the 2029 mapping system. One possible failure mode is for an ETR to lose 2030 reachability to one or more of the EID prefixes within its own site. 2031 When this occurs when the ETR sends Map-Replies, it can clear the 2032 R-bit associated with its own locator. And when the ETR is also an 2033 ITR, it can clear its locator-status-bit in the encapsulation data 2034 header. 2036 6.5. Routing Locator Hashing 2038 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 2039 a requesting ITR, the locator-set for the EID-prefix may contain 2040 different priority values for each locator address. When more than 2041 one best priority locator exists, the ITR can decide how to load 2042 share traffic against the corresponding locators. 2044 The following hash algorithm may be used by an ITR to select a 2045 locator for a packet destined to an EID for the EID-to-RLOC mapping: 2047 1. Either a source and destination address hash can be used or the 2048 traditional 5-tuple hash which includes the source and 2049 destination addresses, source and destination TCP, UDP, or SCTP 2050 port numbers and the IP protocol number field or IPv6 next- 2051 protocol fields of a packet a host originates from within a LISP 2052 site. When a packet is not a TCP, UDP, or SCTP packet, the 2053 source and destination addresses only from the header are used to 2054 compute the hash. 2056 2. Take the hash value and divide it by the number of locators 2057 stored in the locator-set for the EID-to-RLOC mapping. 2059 3. The remainder will be yield a value of 0 to "number of locators 2060 minus 1". Use the remainder to select the locator in the 2061 locator-set. 2063 Note that when a packet is LISP encapsulated, the source port number 2064 in the outer UDP header needs to be set. Selecting a hashed value 2065 allows core routers which are attached to Link Aggregation Groups 2066 (LAGs) to load-split the encapsulated packets across member links of 2067 such LAGs. Otherwise, core routers would see a single flow, since 2068 packets have a source address of the ITR, for packets which are 2069 originated by different EIDs at the source site. A suggested setting 2070 for the source port number computed by an ITR is a 5-tuple hash 2071 function on the inner header, as described above. 2073 Many core router implementations use a 5-tuple hash to decide how to 2074 balance packet load across members of a LAG. The 5-tuple hash 2075 includes the source and destination addresses of the packet and the 2076 source and destination ports when the protocol number in the packet 2077 is TCP or UDP. For this reason, UDP encoding is used for LISP 2078 encapsulation. 2080 6.6. Changing the Contents of EID-to-RLOC Mappings 2082 Since the LISP architecture uses a caching scheme to retrieve and 2083 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 2084 date mapping is to re-request the mapping. However, the ITRs do not 2085 know when the mappings change and the ETRs do not keep track of which 2086 ITRs requested its mappings. For scalability reasons, we want to 2087 maintain this approach but need to provide a way for ETRs change 2088 their mappings and inform the sites that are currently communicating 2089 with the ETR site using such mappings. 2091 When a locator record is added to the end of a locator-set, it is 2092 easy to update mappings. We assume new mappings will maintain the 2093 same locator ordering as the old mapping but just have new locators 2094 appended to the end of the list. So some ITRs can have a new mapping 2095 while other ITRs have only an old mapping that is used until they 2096 time out. When an ITR has only an old mapping but detects bits set 2097 in the loc-status-bits that correspond to locators beyond the list it 2098 has cached, it simply ignores them. However, this can only happen 2099 for locator addresses that are lexicographically greater than the 2100 locator addresses in the existing locator-set. 2102 When a locator record is removed from a locator-set, ITRs that have 2103 the mapping cached will not use the removed locator because the xTRs 2104 will set the loc-status-bit to 0. So even if the locator is in the 2105 list, it will not be used. For new mapping requests, the xTRs can 2106 set the locator AFI to 0 (indicating an unspecified address), as well 2107 as setting the corresponding loc-status-bit to 0. This forces ITRs 2108 with old or new mappings to avoid using the removed locator. 2110 If many changes occur to a mapping over a long period of time, one 2111 will find empty record slots in the middle of the locator-set and new 2112 records appended to the locator-set. At some point, it would be 2113 useful to compact the locator-set so the loc-status-bit settings can 2114 be efficiently packed. 2116 We propose here three approaches for locator-set compaction, one 2117 operational and two protocol mechanisms. The operational approach 2118 uses a clock sweep method. The protocol approaches use the concept 2119 of Solicit-Map-Requests and Map-Versioning. 2121 6.6.1. Clock Sweep 2123 The clock sweep approach uses planning in advance and the use of 2124 count-down TTLs to time out mappings that have already been cached. 2125 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 2126 there is a 24 hour window to time out old mappings. The following 2127 clock sweep procedure is used: 2129 1. 24 hours before a mapping change is to take effect, a network 2130 administrator configures the ETRs at a site to start the clock 2131 sweep window. 2133 2. During the clock sweep window, ETRs continue to send Map-Reply 2134 messages with the current (unchanged) mapping records. The TTL 2135 for these mappings is set to 1 hour. 2137 3. 24 hours later, all previous cache entries will have timed out, 2138 and any active cache entries will time out within 1 hour. During 2139 this 1 hour window the ETRs continue to send Map-Reply messages 2140 with the current (unchanged) mapping records with the TTL set to 2141 1 minute. 2143 4. At the end of the 1 hour window, the ETRs will send Map-Reply 2144 messages with the new (changed) mapping records. So any active 2145 caches can get the new mapping contents right away if not cached, 2146 or in 1 minute if they had the mapping cached. The new mappings 2147 are cached with a time to live equal to the TTL in the Map-Reply. 2149 6.6.2. Solicit-Map-Request (SMR) 2151 Soliciting a Map-Request is a selective way for ETRs, at the site 2152 where mappings change, to control the rate they receive requests for 2153 Map-Reply messages. SMRs are also used to tell remote ITRs to update 2154 the mappings they have cached. 2156 Since the ETRs don't keep track of remote ITRs that have cached their 2157 mappings, they do not know which ITRs need to have their mappings 2158 updated. As a result, an ETR will solicit Map-Requests (called an 2159 SMR message) from those sites to which it has been sending 2160 encapsulated data to for the last minute. In particular, an ETR will 2161 send an SMR an ITR to which it has recently sent encapsulated data. 2163 An SMR message is simply a bit set in a Map-Request message. An ITR 2164 or PTR will send a Map-Request when they receive an SMR message. 2165 Both the SMR sender and the Map-Request responder MUST rate-limited 2166 these messages. Rate-limiting can be implemented as a global rate- 2167 limiter or one rate-limiter per SMR destination. 2169 The following procedure shows how a SMR exchange occurs when a site 2170 is doing locator-set compaction for an EID-to-RLOC mapping: 2172 1. When the database mappings in an ETR change, the ETRs at the site 2173 begin to send Map-Requests with the SMR bit set for each locator 2174 in each map-cache entry the ETR caches. 2176 2. A remote ITR which receives the SMR message will schedule sending 2177 a Map-Request message to the source locator address of the SMR 2178 message or to the mapping database system. A newly allocated 2179 random nonce is selected and the EID-prefix used is the one 2180 copied from the SMR message. If the source locator is the only 2181 locator in the cached locator-set, the remote ITR SHOULD send a 2182 Map-Request to the database mapping system just in case the 2183 single locator has changed and may no longer be reachable to 2184 accept the Map-Request. 2186 3. The remote ITR MUST rate-limit the Map-Request until it gets a 2187 Map-Reply while continuing to use the cached mapping. When Map 2188 Versioning is used, described in Section 6.6.3, an SMR sender can 2189 detect if an ITR is using the most up to date database mapping. 2191 4. The ETRs at the site with the changed mapping will reply to the 2192 Map-Request with a Map-Reply message that has a nonce from the 2193 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate 2194 limited. This is important to avoid Map-Reply implosion. 2196 5. The ETRs, at the site with the changed mapping, record the fact 2197 that the site that sent the Map-Request has received the new 2198 mapping data in the mapping cache entry for the remote site so 2199 the loc-status-bits are reflective of the new mapping for packets 2200 going to the remote site. The ETR then stops sending SMR 2201 messages. 2203 For security reasons an ITR MUST NOT process unsolicited Map-Replies. 2204 To avoid map-cache entry corruption by a third-party, a sender of an 2205 SMR-based Map-Request MUST be verified. If an ITR receives an SMR- 2206 based Map-Request and the source is not in the locator-set for the 2207 stored map-cache entry, then the responding Map-Request MUST be sent 2208 with an EID destination to the mapping database system. Since the 2209 mapping database system is more secure to reach an authoritative ETR, 2210 it will deliver the Map-Request to the authoritative source of the 2211 mapping data. 2213 When an ITR receives an SMR-based Map-Request for which it does not 2214 have a cached mapping for the EID in the SMR message, it MAY not send 2215 a SMR-invoked Map-Request. This scenario can occur when an ETR sends 2216 SMR messages to all locators in the locator-set it has stored in its 2217 map-cache but the remote ITRs that receive the SMR may not be sending 2218 packets to the site. There is no point in updating the ITRs until 2219 they need to send, in which case, they will send Map-Requests to 2220 obtain a map-cache entry. 2222 6.6.3. Database Map Versioning 2224 When there is unidirectional packet flow between an ITR and ETR, and 2225 the EID-to-RLOC mappings change on the ETR, it needs to inform the 2226 ITR so encapsulation can stop to a removed locator and start to a new 2227 locator in the locator-set. 2229 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 2230 number. This is known as the Destination Map-Version Number. ITRs 2231 include the Destination Map-Version Number in packets they 2232 encapsulate to the site. When an ETR decapsulates a packet and 2233 detects the Destination Map-Version Number is less than the current 2234 version for its mapping, the SMR procedure described in Section 6.6.2 2235 occurs. 2237 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 2238 Version number. This is known as the Source Map-Version Number. 2240 When an ETR decapsulates a packet and detects the Source Map-Version 2241 Number is greater than the last Map-Version Number sent in a Map- 2242 Reply from the ITR's site, the ETR will send a Map-Request to one of 2243 the ETRs for the source site. 2245 A Map-Version Number is used as a sequence number per EID-prefix. So 2246 values that are greater, are considered to be more recent. A value 2247 of 0 for the Source Map-Version Number or the Destination Map-Version 2248 Number conveys no versioning information and an ITR does no 2249 comparison with previously received Map-Version Numbers. 2251 A Map-Version Number can be included in Map-Register messages as 2252 well. This is a good way for the Map-Server can assure that all ETRs 2253 for a site registering to it will be Map-Version number synchronized. 2255 See [VERSIONING] for a more detailed analysis and description of 2256 Database Map Versioning. 2258 7. Router Performance Considerations 2260 LISP is designed to be very hardware-based forwarding friendly. A 2261 few implementation techniques can be used to incrementally implement 2262 LISP: 2264 o When a tunnel encapsulated packet is received by an ETR, the outer 2265 destination address may not be the address of the router. This 2266 makes it challenging for the control plane to get packets from the 2267 hardware. This may be mitigated by creating special FIB entries 2268 for the EID-prefixes of EIDs served by the ETR (those for which 2269 the router provides an RLOC translation). These FIB entries are 2270 marked with a flag indicating that control plane processing should 2271 be performed. The forwarding logic of testing for particular IP 2272 protocol number value is not necessary. No changes to existing, 2273 deployed hardware should be needed to support this. 2275 o On an ITR, prepending a new IP header consists of adding more 2276 bytes to a MAC rewrite string and prepending the string as part of 2277 the outgoing encapsulation procedure. Routers that support GRE 2278 tunneling [RFC2784] or 6to4 tunneling [RFC3056] may already 2279 support this action. 2281 o A packet's source address or interface the packet was received on 2282 can be used to select a VRF (Virtual Routing/Forwarding). The 2283 VRF's routing table can be used to find EID-to-RLOC mappings. 2285 8. Deployment Scenarios 2287 This section will explore how and where ITRs and ETRs can be deployed 2288 and will discuss the pros and cons of each deployment scenario. 2289 There are two basic deployment trade-offs to consider: centralized 2290 versus distributed caches and flat, recursive, or re-encapsulating 2291 tunneling. 2293 When deciding on centralized versus distributed caching, the 2294 following issues should be considered: 2296 o Are the tunnel routers spread out so that the caches are spread 2297 across all the memories of each router? 2299 o Should management "touch points" be minimized by choosing few 2300 tunnel routers, just enough for redundancy? 2302 o In general, using more ITRs doesn't increase management load, 2303 since caches are built and stored dynamically. On the other hand, 2304 more ETRs does require more management since EID-prefix-to-RLOC 2305 mappings need to be explicitly configured. 2307 When deciding on flat, recursive, or re-encapsulation tunneling, the 2308 following issues should be considered: 2310 o Flat tunneling implements a single tunnel between source site and 2311 destination site. This generally offers better paths between 2312 sources and destinations with a single tunnel path. 2314 o Recursive tunneling is when tunneled traffic is again further 2315 encapsulated in another tunnel, either to implement VPNs or to 2316 perform Traffic Engineering. When doing VPN-based tunneling, the 2317 site has some control since the site is prepending a new tunnel 2318 header. In the case of TE-based tunneling, the site may have 2319 control if it is prepending a new tunnel header, but if the site's 2320 ISP is doing the TE, then the site has no control. Recursive 2321 tunneling generally will result in suboptimal paths but at the 2322 benefit of steering traffic to resource available parts of the 2323 network. 2325 o The technique of re-encapsulation ensures that packets only 2326 require one tunnel header. So if a packet needs to be rerouted, 2327 it is first decapsulated by the ETR and then re-encapsulated with 2328 a new tunnel header using a new RLOC. 2330 The next sub-sections will survey where tunnel routers can reside in 2331 the network. 2333 8.1. First-hop/Last-hop Tunnel Routers 2335 By locating tunnel routers close to hosts, the EID-prefix set is at 2336 the granularity of an IP subnet. So at the expense of more EID- 2337 prefix-to-RLOC sets for the site, the caches in each tunnel router 2338 can remain relatively small. But caches always depend on the number 2339 of non-aggregated EID destination flows active through these tunnel 2340 routers. 2342 With more tunnel routers doing encapsulation, the increase in control 2343 traffic grows as well: since the EID-granularity is greater, more 2344 Map-Requests and Map-Replies are traveling between more routers. 2346 The advantage of placing the caches and databases at these stub 2347 routers is that the products deployed in this part of the network 2348 have better price-memory ratios then their core router counterparts. 2349 Memory is typically less expensive in these devices and fewer routes 2350 are stored (only IGP routes). These devices tend to have excess 2351 capacity, both for forwarding and routing state. 2353 LISP functionality can also be deployed in edge switches. These 2354 devices generally have layer-2 ports facing hosts and layer-3 ports 2355 facing the Internet. Spare capacity is also often available in these 2356 devices as well. 2358 8.2. Border/Edge Tunnel Routers 2360 Using customer-edge (CE) routers for tunnel endpoints allows the EID 2361 space associated with a site to be reachable via a small set of RLOCs 2362 assigned to the CE routers for that site. This is the default 2363 behavior envisioned in the rest of this specification. 2365 This offers the opposite benefit of the first-hop/last-hop tunnel 2366 router scenario: the number of mapping entries and network management 2367 touch points are reduced, allowing better scaling. 2369 One disadvantage is that less of the network's resources are used to 2370 reach host endpoints thereby centralizing the point-of-failure domain 2371 and creating network choke points at the CE router. 2373 Note that more than one CE router at a site can be configured with 2374 the same IP address. In this case an RLOC is an anycast address. 2375 This allows resilience between the CE routers. That is, if a CE 2376 router fails, traffic is automatically routed to the other routers 2377 using the same anycast address. However, this comes with the 2378 disadvantage where the site cannot control the entrance point when 2379 the anycast route is advertised out from all border routers. Another 2380 disadvantage of using anycast locators is the limited advertisement 2381 scope of /32 (or /128 for IPv6) routes. 2383 8.3. ISP Provider-Edge (PE) Tunnel Routers 2385 Use of ISP PE routers as tunnel endpoint routers is not the typical 2386 deployment scenario envisioned in the specification. This section 2387 attempts to capture some of reasoning behind this preference of 2388 implementing LISP on CE routers. 2390 Use of ISP PE routers as tunnel endpoint routers gives an ISP, rather 2391 than a site, control over the location of the egress tunnel 2392 endpoints. That is, the ISP can decide if the tunnel endpoints are 2393 in the destination site (in either CE routers or last-hop routers 2394 within a site) or at other PE edges. The advantage of this case is 2395 that two tunnel headers can be avoided. By having the PE be the 2396 first router on the path to encapsulate, it can choose a TE path 2397 first, and the ETR can decapsulate and re-encapsulate for a tunnel to 2398 the destination end site. 2400 An obvious disadvantage is that the end site has no control over 2401 where its packets flow or the RLOCs used. Other disadvantages 2402 include the difficulty in synchronizing path liveness updates between 2403 CE and PE routers. 2405 As mentioned in earlier sections a combination of these scenarios is 2406 possible at the expense of extra packet header overhead, if both site 2407 and provider want control, then recursive or re-encapsulating tunnels 2408 are used. 2410 8.4. LISP Functionality with Conventional NATs 2412 LISP routers can be deployed behind Network Address Translator (NAT) 2413 devices to provide the same set of packet services hosts have today 2414 when they are addressed out of private address space. 2416 It is important to note that a locator address in any LISP control 2417 message MUST be a globally routable address and therefore SHOULD NOT 2418 contain [RFC1918] addresses. If a LISP router is configured with 2419 private addresses, they MUST be used only in the outer IP header so 2420 the NAT device can translate properly. Otherwise, EID addresses MUST 2421 be translated before encapsulation is performed. Both NAT 2422 translation and LISP encapsulation functions could be co-located in 2423 the same device. 2425 More details on LISP address translation can be found in [INTERWORK]. 2427 9. Traceroute Considerations 2429 When a source host in a LISP site initiates a traceroute to a 2430 destination host in another LISP site, it is highly desirable for it 2431 to see the entire path. Since packets are encapsulated from ITR to 2432 ETR, the hop across the tunnel could be viewed as a single hop. 2433 However, LISP traceroute will provide the entire path so the user can 2434 see 3 distinct segments of the path from a source LISP host to a 2435 destination LISP host: 2437 Segment 1 (in source LISP site based on EIDs): 2439 source-host ---> first-hop ... next-hop ---> ITR 2441 Segment 2 (in the core network based on RLOCs): 2443 ITR ---> next-hop ... next-hop ---> ETR 2445 Segment 3 (in the destination LISP site based on EIDs): 2447 ETR ---> next-hop ... last-hop ---> destination-host 2449 For segment 1 of the path, ICMP Time Exceeded messages are returned 2450 in the normal matter as they are today. The ITR performs a TTL 2451 decrement and test for 0 before encapsulating. So the ITR hop is 2452 seen by the traceroute source has an EID address (the address of 2453 site-facing interface). 2455 For segment 2 of the path, ICMP Time Exceeded messages are returned 2456 to the ITR because the TTL decrement to 0 is done on the outer 2457 header, so the destination of the ICMP messages are to the ITR RLOC 2458 address, the source RLOC address of the encapsulated traceroute 2459 packet. The ITR looks inside of the ICMP payload to inspect the 2460 traceroute source so it can return the ICMP message to the address of 2461 the traceroute client as well as retaining the core router IP address 2462 in the ICMP message. This is so the traceroute client can display 2463 the core router address (the RLOC address) in the traceroute output. 2464 The ETR returns its RLOC address and responds to the TTL decrement to 2465 0 like the previous core routers did. 2467 For segment 3, the next-hop router downstream from the ETR will be 2468 decrementing the TTL for the packet that was encapsulated, sent into 2469 the core, decapsulated by the ETR, and forwarded because it isn't the 2470 final destination. If the TTL is decremented to 0, any router on the 2471 path to the destination of the traceroute, including the next-hop 2472 router or destination, will send an ICMP Time Exceeded message to the 2473 source EID of the traceroute client. The ICMP message will be 2474 encapsulated by the local ITR and sent back to the ETR in the 2475 originated traceroute source site, where the packet will be delivered 2476 to the host. 2478 9.1. IPv6 Traceroute 2480 IPv6 traceroute follows the procedure described above since the 2481 entire traceroute data packet is included in ICMP Time Exceeded 2482 message payload. Therefore, only the ITR needs to pay special 2483 attention for forwarding ICMP messages back to the traceroute source. 2485 9.2. IPv4 Traceroute 2487 For IPv4 traceroute, we cannot follow the above procedure since IPv4 2488 ICMP Time Exceeded messages only include the invoking IP header and 8 2489 bytes that follow the IP header. Therefore, when a core router sends 2490 an IPv4 Time Exceeded message to an ITR, all the ITR has in the ICMP 2491 payload is the encapsulated header it prepended followed by a UDP 2492 header. The original invoking IP header, and therefore the identity 2493 of the traceroute source is lost. 2495 The solution we propose to solve this problem is to cache traceroute 2496 IPv4 headers in the ITR and to match them up with corresponding IPv4 2497 Time Exceeded messages received from core routers and the ETR. The 2498 ITR will use a circular buffer for caching the IPv4 and UDP headers 2499 of traceroute packets. It will select a 16-bit number as a key to 2500 find them later when the IPv4 Time Exceeded messages are received. 2501 When an ITR encapsulates an IPv4 traceroute packet, it will use the 2502 16-bit number as the UDP source port in the encapsulating header. 2503 When the ICMP Time Exceeded message is returned to the ITR, the UDP 2504 header of the encapsulating header is present in the ICMP payload 2505 thereby allowing the ITR to find the cached headers for the 2506 traceroute source. The ITR puts the cached headers in the payload 2507 and sends the ICMP Time Exceeded message to the traceroute source 2508 retaining the source address of the original ICMP Time Exceeded 2509 message (a core router or the ETR of the site of the traceroute 2510 destination). 2512 The signature of a traceroute packet comes in two forms. The first 2513 form is encoded as a UDP message where the destination port is 2514 inspected for a range of values. The second form is encoded as an 2515 ICMP message where the IP identification field is inspected for a 2516 well-known value. 2518 9.3. Traceroute using Mixed Locators 2520 When either an IPv4 traceroute or IPv6 traceroute is originated and 2521 the ITR encapsulates it in the other address family header, you 2522 cannot get all 3 segments of the traceroute. Segment 2 of the 2523 traceroute can not be conveyed to the traceroute source since it is 2524 expecting addresses from intermediate hops in the same address format 2525 for the type of traceroute it originated. Therefore, in this case, 2526 segment 2 will make the tunnel look like one hop. All the ITR has to 2527 do to make this work is to not copy the inner TTL to the outer, 2528 encapsulating header's TTL when a traceroute packet is encapsulated 2529 using an RLOC from a different address family. This will cause no 2530 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2532 10. Mobility Considerations 2534 There are several kinds of mobility of which only some might be of 2535 concern to LISP. Essentially they are as follows. 2537 10.1. Site Mobility 2539 A site wishes to change its attachment points to the Internet, and 2540 its LISP Tunnel Routers will have new RLOCs when it changes upstream 2541 providers. Changes in EID-RLOC mappings for sites are expected to be 2542 handled by configuration, outside of the LISP protocol. 2544 10.2. Slow Endpoint Mobility 2546 An individual endpoint wishes to move, but is not concerned about 2547 maintaining session continuity. Renumbering is involved. LISP can 2548 help with the issues surrounding renumbering [RFC4192] [LISA96] by 2549 decoupling the address space used by a site from the address spaces 2550 used by its ISPs. [RFC4984] 2552 10.3. Fast Endpoint Mobility 2554 Fast endpoint mobility occurs when an endpoint moves relatively 2555 rapidly, changing its IP layer network attachment point. Maintenance 2556 of session continuity is a goal. This is where the Mobile IPv4 2557 [RFC3344bis] and Mobile IPv6 [RFC3775] [RFC4866] mechanisms are used, 2558 and primarily where interactions with LISP need to be explored. 2560 The problem is that as an endpoint moves, it may require changes to 2561 the mapping between its EID and a set of RLOCs for its new network 2562 location. When this is added to the overhead of mobile IP binding 2563 updates, some packets might be delayed or dropped. 2565 In IPv4 mobility, when an endpoint is away from home, packets to it 2566 are encapsulated and forwarded via a home agent which resides in the 2567 home area the endpoint's address belongs to. The home agent will 2568 encapsulate and forward packets either directly to the endpoint or to 2569 a foreign agent which resides where the endpoint has moved to. 2570 Packets from the endpoint may be sent directly to the correspondent 2571 node, may be sent via the foreign agent, or may be reverse-tunneled 2572 back to the home agent for delivery to the mobile node. As the 2573 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 2574 mappings are required for communication between the mobile node and 2575 the home agent, whether via foreign agent or not. As a mobile 2576 endpoint changes networks, up to three LISP mapping changes may be 2577 required: 2579 o The mobile node moves from an old location to a new visited 2580 network location and notifies its home agent that it has done so. 2581 The Mobile IPv4 control packets the mobile node sends pass through 2582 one of the new visited network's ITRs, which needs a EID-RLOC 2583 mapping for the home agent. 2585 o The home agent might not have the EID-RLOC mappings for the mobile 2586 node's "care-of" address or its foreign agent in the new visited 2587 network, in which case it will need to acquire them. 2589 o When packets are sent directly to the correspondent node, it may 2590 be that no traffic has been sent from the new visited network to 2591 the correspondent node's network, and the new visited network's 2592 ITR will need to obtain an EID-RLOC mapping for the correspondent 2593 node's site. 2595 In addition, if the IPv4 endpoint is sending packets from the new 2596 visited network using its original EID, then LISP will need to 2597 perform a route-returnability check on the new EID-RLOC mapping for 2598 that EID. 2600 In IPv6 mobility, packets can flow directly between the mobile node 2601 and the correspondent node in either direction. The mobile node uses 2602 its "care-of" address (EID). In this case, the route-returnability 2603 check would not be needed but one more LISP mapping lookup may be 2604 required instead: 2606 o As above, three mapping changes may be needed for the mobile node 2607 to communicate with its home agent and to send packets to the 2608 correspondent node. 2610 o In addition, another mapping will be needed in the correspondent 2611 node's ITR, in order for the correspondent node to send packets to 2612 the mobile node's "care-of" address (EID) at the new network 2613 location. 2615 When both endpoints are mobile the number of potential mapping 2616 lookups increases accordingly. 2618 As a mobile node moves there are not only mobility state changes in 2619 the mobile node, correspondent node, and home agent, but also state 2620 changes in the ITRs and ETRs for at least some EID-prefixes. 2622 The goal is to support rapid adaptation, with little delay or packet 2623 loss for the entire system. Also IP mobility can be modified to 2624 require fewer mapping changes. In order to increase overall system 2625 performance, there may be a need to reduce the optimization of one 2626 area in order to place fewer demands on another. 2628 In LISP, one possibility is to "glean" information. When a packet 2629 arrives, the ETR could examine the EID-RLOC mapping and use that 2630 mapping for all outgoing traffic to that EID. It can do this after 2631 performing a route-returnability check, to ensure that the new 2632 network location does have a internal route to that endpoint. 2633 However, this does not cover the case where an ITR (the node assigned 2634 the RLOC) at the mobile-node location has been compromised. 2636 Mobile IP packet exchange is designed for an environment in which all 2637 routing information is disseminated before packets can be forwarded. 2638 In order to allow the Internet to grow to support expected future 2639 use, we are moving to an environment where some information may have 2640 to be obtained after packets are in flight. Modifications to IP 2641 mobility should be considered in order to optimize the behavior of 2642 the overall system. Anything which decreases the number of new EID- 2643 RLOC mappings needed when a node moves, or maintains the validity of 2644 an EID-RLOC mapping for a longer time, is useful. 2646 10.4. Fast Network Mobility 2648 In addition to endpoints, a network can be mobile, possibly changing 2649 xTRs. A "network" can be as small as a single router and as large as 2650 a whole site. This is different from site mobility in that it is 2651 fast and possibly short-lived, but different from endpoint mobility 2652 in that a whole prefix is changing RLOCs. However, the mechanisms 2653 are the same and there is no new overhead in LISP. A map request for 2654 any endpoint will return a binding for the entire mobile prefix. 2656 If mobile networks become a more common occurrence, it may be useful 2657 to revisit the design of the mapping service and allow for dynamic 2658 updates of the database. 2660 The issue of interactions between mobility and LISP needs to be 2661 explored further. Specific improvements to the entire system will 2662 depend on the details of mapping mechanisms. Mapping mechanisms 2663 should be evaluated on how well they support session continuity for 2664 mobile nodes. 2666 10.5. LISP Mobile Node Mobility 2668 A mobile device can use the LISP infrastructure to achieve mobility 2669 by implementing the LISP encapsulation and decapsulation functions 2670 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 2671 node" can use topologically-independent EID IP addresses that are not 2672 advertised into and do not impose a cost on the global routing 2673 system. These EIDs are maintained at the edges of the mapping system 2674 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 2675 only the correspondents of the LISP mobile node. 2677 Refer to the LISP Mobility Architecture specification [LISP-MN] for 2678 more details. 2680 11. Multicast Considerations 2682 A multicast group address, as defined in the original Internet 2683 architecture is an identifier of a grouping of topologically 2684 independent receiver host locations. The address encoding itself 2685 does not determine the location of the receiver(s). The multicast 2686 routing protocol, and the network-based state the protocol creates, 2687 determines where the receivers are located. 2689 In the context of LISP, a multicast group address is both an EID and 2690 a Routing Locator. Therefore, no specific semantic or action needs 2691 to be taken for a destination address, as it would appear in an IP 2692 header. Therefore, a group address that appears in an inner IP 2693 header built by a source host will be used as the destination EID. 2694 The outer IP header (the destination Routing Locator address), 2695 prepended by a LISP router, will use the same group address as the 2696 destination Routing Locator. 2698 Having said that, only the source EID and source Routing Locator 2699 needs to be dealt with. Therefore, an ITR merely needs to put its 2700 own IP address in the source Routing Locator field when prepending 2701 the outer IP header. This source Routing Locator address, like any 2702 other Routing Locator address MUST be globally routable. 2704 Therefore, an EID-to-RLOC mapping does not need to be performed by an 2705 ITR when a received data packet is a multicast data packet or when 2706 processing a source-specific Join (either by IGMPv3 or PIM). But the 2707 source Routing Locator is decided by the multicast routing protocol 2708 in a receiver site. That is, an EID to Routing Locator translation 2709 is done at control-time. 2711 Another approach is to have the ITR not encapsulate a multicast 2712 packet and allow the host built packet to flow into the core even if 2713 the source address is allocated out of the EID namespace. If the 2714 RPF-Vector TLV [RFC5496] is used by PIM in the core, then core 2715 routers can RPF to the ITR (the Locator address which is injected 2716 into core routing) rather than the host source address (the EID 2717 address which is not injected into core routing). 2719 To avoid any EID-based multicast state in the network core, the first 2720 approach is chosen for LISP-Multicast. Details for LISP-Multicast 2721 and Interworking with non-LISP sites is described in specification 2722 [MLISP]. 2724 12. Security Considerations 2726 It is believed that most of the security mechanisms will be part of 2727 the mapping database service when using control plane procedures for 2728 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 2729 as described in this specification, protection is provided against 2730 ETR spoofing by using Return- Routability mechanisms evidenced by the 2731 use of a 24-bit Nonce field in the LISP encapsulation header and a 2732 64-bit Nonce field in the LISP control message. The nonce, coupled 2733 with the ITR accepting only solicited Map-Replies goes a long way 2734 toward providing decent authentication. 2736 LISP does not rely on a PKI infrastructure or a more heavy weight 2737 authentication system. These systems challenge the scalability of 2738 LISP which was a primary design goal. 2740 DoS attack prevention will depend on implementations rate-limiting 2741 Map-Requests and Map-Replies to the control plane as well as rate- 2742 limiting the number of data-triggered Map-Replies. 2744 To deal with map-cache exhaustion attempts in an ITR/PTR, the 2745 implementation should consider putting a maximum cap on the number of 2746 entries stored with a reserve list for special or frequently accessed 2747 sites. This should be a configuration policy control set by the 2748 network administrator who manages ITRs and PTRs. 2750 13. Network Management Considerations 2752 Considerations for Network Management tools exist so the LISP 2753 protocol suite can be operationally managed. The mechanisms can be 2754 found in [LISP-MIB] and [LISP-LIG]. 2756 14. IANA Considerations 2758 This section provides guidance to the Internet Assigned Numbers 2759 Authority (IANA) regarding registration of values related to the LISP 2760 specification, in accordance with BCP 26 and RFC 5226 [RFC5226]. 2762 There are two name spaces in LISP that require registration: 2764 o LISP IANA registry allocations should not be made for purposes 2765 unrelated to LISP routing or transport protocols. 2767 o The following policies are used here with the meanings defined in 2768 BCP 26: "Specification Required", "IETF Consensus", "Experimental 2769 Use", "First Come First Served". 2771 14.1. LISP Address Type Codes 2773 Instance ID type codes have a range from 0 to 15, of which 0 and 1 2774 have been allocated [LCAF]. New Type Codes MUST be allocated 2775 starting at 2. Type Codes 2 - 10 are to be assigned by IETF Review. 2776 Type Codes 11 - 15 are available on a First Come First Served policy. 2778 The following codes have been allocated: 2780 Type 0: Null Body Type 2782 Type 1: AFI List Type 2784 See [LCAF] for details for other possible unapproved address 2785 encodings. The unapproved LCAF encodings are an area for further 2786 study and experimentation. 2788 14.2. LISP UDP Port Numbers 2790 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2791 LISP data-plane and control-plane operation, respectively. 2793 15. References 2795 15.1. Normative References 2797 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2798 August 1980. 2800 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2801 STD 13, RFC 1034, November 1987. 2803 [RFC1700] Reynolds, J. and J. Postel, "Assigned Numbers", RFC 1700, 2804 October 1994. 2806 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 2807 E. Lear, "Address Allocation for Private Internets", 2808 BCP 5, RFC 1918, February 1996. 2810 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2811 Requirement Levels", BCP 14, RFC 2119, March 1997. 2813 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 2814 ESP and AH", RFC 2404, November 1998. 2816 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2817 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2818 March 2000. 2820 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 2821 via IPv4 Clouds", RFC 3056, February 2001. 2823 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2824 of Explicit Congestion Notification (ECN) to IP", 2825 RFC 3168, September 2001. 2827 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2828 A., Peterson, J., Sparks, R., Handley, M., and E. 2829 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2830 June 2002. 2832 [RFC3775] Johnson, D., Perkins, C., and J. Arkko, "Mobility Support 2833 in IPv6", RFC 3775, June 2004. 2835 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 2836 Requirements for Security", BCP 106, RFC 4086, June 2005. 2838 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 2839 (CIDR): The Internet Address Assignment and Aggregation 2840 Plan", BCP 122, RFC 4632, August 2006. 2842 [RFC4634] Eastlake, D. and T. Hansen, "US Secure Hash Algorithms 2843 (SHA and HMAC-SHA)", RFC 4634, July 2006. 2845 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2846 Optimization for Mobile IPv6", RFC 4866, May 2007. 2848 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 2849 Workshop on Routing and Addressing", RFC 4984, 2850 September 2007. 2852 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2853 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2854 May 2008. 2856 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 2857 Forwarding (RPF) Vector TLV", RFC 5496, March 2009. 2859 [UDP-TUNNELS] 2860 Eubanks, M. and P. Chimento, "UDP Checksums for Tunneled 2861 Packets"", draft-eubanks-chimento-6man-00.txt (work in 2862 progress), February 2009. 2864 15.2. Informative References 2866 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 2867 NUMBERS http://www.iana.org/numbers.html, Febuary 2007. 2869 [ALT] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "LISP 2870 Alternative Topology (LISP-ALT)", 2871 draft-ietf-lisp-alt-06.txt (work in progress), March 2011. 2873 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 2874 Enhancement to the Internet Architecture", Internet- 2875 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt, 2876 1999. 2878 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 2879 Content distribution Overlay Network Service for LISP", 2880 draft-meyer-lisp-cons-03.txt (work in progress), 2881 November 2007. 2883 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 2884 Mappings Multicast Across Cooperating Systems for LISP", 2885 draft-curran-lisp-emacs-00.txt (work in progress), 2886 November 2007. 2888 [INTERWORK] 2889 Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 2890 "Interworking LISP with IPv4 and IPv6", 2891 draft-ietf-lisp-interworking-02.txt (work in progress), 2892 March 2011. 2894 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 2895 Address Format", draft-farinacci-lisp-lcaf-04.txt (work in 2896 progress), October 2010. 2898 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 2899 "Renumbering: Threat or Menace?", Usenix , September 1996. 2901 [LISP-LIG] 2902 Farinacci, D. and D. Meyer, "LISP Internet Groper (LIG)", 2903 draft-ietf-lisp-lig-01.txt (work in progress), 2904 October 2010. 2906 [LISP-MAIN] 2907 Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 2908 "Locator/ID Separation Protocol (LISP)", 2909 draft-farinacci-lisp-12.txt (work in progress), 2910 March 2009. 2912 [LISP-MIB] 2913 Schudel, G., Jain, A., and V. Moreno, "LISP MIB", 2914 draft-ietf-lisp-mib-01.txt (work in progress), March 2011. 2916 [LISP-MN] Farinacci, D., Fuller, V., Lewis, D., and D. Meyer, "LISP 2917 Mobility Architecture", draft-meyer-lisp-mn-04.txt (work 2918 in progress), October 2010. 2920 [LISP-MS] Farinacci, D. and V. Fuller, "LISP Map Server", 2921 draft-ietf-lisp-ms-07.txt (work in progress), March 2011. 2923 [LISP-SEC] 2924 Maino, F., Ermagon, V., Cabellos, A., Sausez, D., and O. 2925 Bonaventure, "LISP-Security (LISP-SEC)", 2926 draft-maino-lisp-sec-00.txt (work in progress), 2927 February 2011. 2929 [LOC-ID-ARCH] 2930 Meyer, D. and D. Lewis, "Architectural Implications of 2931 Locator/ID Separation", 2932 draft-meyer-loc-id-implications-01.txt (work in progress), 2933 Januaryr 2009. 2935 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 2936 "LISP for Multicast Environments", 2937 draft-ietf-lisp-multicast-04.txt (work in progress), 2938 October 2010. 2940 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 2941 draft-lear-lisp-nerd-08.txt (work in progress), 2942 March 2010. 2944 [OPENLISP] 2945 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 2946 Report", draft-iannone-openlisp-implementation-01.txt 2947 (work in progress), July 2008. 2949 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 2950 draft-narten-radir-problem-statement-00.txt (work in 2951 progress), July 2007. 2953 [RFC3344bis] 2954 Perkins, C., "IP Mobility Support for IPv4, revised", 2955 draft-ietf-mip4-rfc3344bis-05 (work in progress), 2956 July 2007. 2958 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2959 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2960 September 2005. 2962 [RPMD] Handley, M., Huici, F., and A. Greenhalgh, "RPMD: Protocol 2963 for Routing Protocol Meta-data Dissemination", 2964 draft-handley-p2ppush-unpublished-2007726.txt (work in 2965 progress), July 2007. 2967 [VERSIONING] 2968 Iannone, L., Saucez, D., and O. Bonaventure, "LISP Mapping 2969 Versioning", draft-ietf-lisp-map-versioning-01.txt (work 2970 in progress), March 2011. 2972 Appendix A. Acknowledgments 2974 An initial thank you goes to Dave Oran for planting the seeds for the 2975 initial ideas for LISP. His consultation continues to provide value 2976 to the LISP authors. 2978 A special and appreciative thank you goes to Noel Chiappa for 2979 providing architectural impetus over the past decades on separation 2980 of location and identity, as well as detailed review of the LISP 2981 architecture and documents, coupled with enthusiasm for making LISP a 2982 practical and incremental transition for the Internet. 2984 The authors would like to gratefully acknowledge many people who have 2985 contributed discussion and ideas to the making of this proposal. 2986 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 2987 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 2988 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 2989 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 2990 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 2991 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 2992 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 2993 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 2994 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 2995 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 2996 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 2997 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 2998 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 2999 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 3000 Ermagan, Albert Cabellos, Fabio Maino, Victor Moreno, and Chris 3001 White. 3003 This work originated in the Routing Research Group (RRG) of the IRTF. 3004 The individual submission [LISP-MAIN] was converted into this IETF 3005 LISP working group draft. 3007 Appendix B. Document Change Log 3009 B.1. Changes to draft-ietf-lisp-10.txt 3011 o Posted March 2011. 3013 o Add p-bit to Map-Request so there is documentary reasons to know 3014 when a PITR has sent a Map-Request to an ETR. 3016 o Add Map-Notify message which is used to acknowledge a Map-Register 3017 message sent to a Map-Server. 3019 o Add M-bit to the Map-Register message so an ETR that wants an 3020 acknowledgment for the Map-Register can request one. 3022 o Add S-bit to the ECM and Map-Reply messages to describe security 3023 data that can be present in each message. Then refer to 3024 [LISP-SEC] for expansive details. 3026 o Add Network Management Considerations section and point to the MIB 3027 and LIG drafts. 3029 o Remove the word "simple" per Yakov's comments. 3031 B.2. Changes to draft-ietf-lisp-09.txt 3033 o Posted October 2010. 3035 o Add to IANA Consideration section about the use of LCAF Type 3036 values that accepted and maintained by the IANA registry and not 3037 the LCAF specification. 3039 o Indicate that implementations should be able to receive LISP 3040 control messages when either UDP port is 4342, so they can be 3041 robust in the face of intervening NAT boxes. 3043 o Add paragraph to SMR section to indicate that an ITR does not need 3044 to respond to an SMR-based Map-Request when it has no map-cache 3045 entry for the SMR source's EID-prefix. 3047 B.3. Changes to draft-ietf-lisp-08.txt 3049 o Posted August 2010. 3051 o In section 6.1.6, remove statement about setting TTL to 0 in Map- 3052 Register messages. 3054 o Clarify language in section 6.1.5 about Map-Replying to Data- 3055 Probes or Map-Requests. 3057 o Indicate that outer TTL should only be copied to inner TTL when it 3058 is less than inner TTL. 3060 o Indicate a source-EID for RLOC-probes are encoded with an AFI 3061 value of 0. 3063 o Indicate that SMRs can have a global or per SMR destination rate- 3064 limiter. 3066 o Add clarifications to the SMR procedures. 3068 o Add definitions for "client-side" and 'server-side" terms used in 3069 this specification. 3071 o Clear up language in section 6.4, last paragraph. 3073 o Change ACT of value 0 to "no-action". This is so we can RLOC- 3074 probe a PETR and have it return a Map-Reply with a locator-set of 3075 size 0. The way it is spec'ed the map-cache entry has action 3076 "dropped". Drop-action is set to 3. 3078 o Add statement about normalizing locator weights. 3080 o Clarify R-bit definition in the Map-Reply locator record. 3082 o Add section on EID Reachability within a LISP site. 3084 o Clarify another disadvantage of using anycast locators. 3086 o Reworded Abstract. 3088 o Change section 2.0 Introduction to remove obsolete information 3089 such as the LISP variant definitions. 3091 o Change section 5 title from "Tunneling Details" to "LISP 3092 Encapsulation Details". 3094 o Changes to section 5 to include results of network deployment 3095 experience with MTU. Recommend that implementations use either 3096 the stateful or stateless handling. 3098 o Make clarification wordsmithing to Section 7 and 8. 3100 o Identify that if there is one locator in the locator-set of a map- 3101 cache entry, that an SMR from that locator should be responded to 3102 by sending the the SMR-invoked Map-Request to the database mapping 3103 system rather than to the RLOC itself (which may be unreachable). 3105 o When describing Unicast and Multicast Weights indicate the the 3106 values are relative weights rather than percentages. So it 3107 doesn't imply the sum of all locator weights in the locator-set 3108 need to be 100. 3110 o Do some wordsmithing on copying TTL and TOS fields. 3112 o Numerous wordsmithing changes from Dave Meyer. He fine toothed 3113 combed the spec. 3115 o Removed Section 14 "Prototype Plans and Status". We felt this 3116 type of section is no longer appropriate for a protocol 3117 specification. 3119 o Add clarification text for the IRC description per Damien's 3120 commentary. 3122 o Remove text on copying nonce from SMR to SMR-invoked Map- Request 3123 per Vina's comment about a possible DoS vector. 3125 o Clarify (S/2 + H) in the stateless MTU section. 3127 o Add text to reflect Damien's comment about the description of the 3128 "ITR-RLOC Address" field in the Map-Request. that the list of RLOC 3129 addresses are local addresses of the Map-Requester. 3131 B.4. Changes to draft-ietf-lisp-07.txt 3133 o Posted April 2010. 3135 o Added I-bit to data header so LSB field can also be used as an 3136 Instance ID field. When this occurs, the LSB field is reduced to 3137 8-bits (from 32-bits). 3139 o Added V-bit to the data header so the 24-bit nonce field can also 3140 be used for source and destination version numbers. 3142 o Added Map-Version 12-bit value to the EID-record to be used in all 3143 of Map-Request, Map-Reply, and Map-Register messages. 3145 o Added multiple ITR-RLOC fields to the Map-Request packet so an ETR 3146 can decide what address to select for the destination of a Map- 3147 Reply. 3149 o Added L-bit (Local RLOC bit) and p-bit (Probe-Reply RLOC bit) to 3150 the Locator-Set record of an EID-record for a Map-Reply message. 3151 The L-bit indicates which RLOCs in the locator-set are local to 3152 the sender of the message. The P-bit indicates which RLOC is the 3153 source of a RLOC-probe Reply (Map-Reply) message. 3155 o Add reference to the LISP Canonical Address Format [LCAF] draft. 3157 o Made editorial and clarification changes based on comments from 3158 Dhirendra Trivedi. 3160 o Added wordsmithing comments from Joel Halpern on DF=1 setting. 3162 o Add John Zwiebel clarification to Echo Nonce Algorithm section 3163 6.3.1. 3165 o Add John Zwiebel comment about expanding on proxy-map-reply bit 3166 for Map-Register messages. 3168 o Add NAT section per Ron Bonica comments. 3170 o Fix IDnits issues per Ron Bonica. 3172 o Added section on Virtualization and Segmentation to explain the 3173 use if the Instance ID field in the data header. 3175 o There are too many P-bits, keep their scope to the packet format 3176 description and refer to them by name every where else in the 3177 spec. 3179 o Scanned all occurrences of "should", "should not", "must" and 3180 "must not" and uppercased them. 3182 o John Zwiebel offered text for section 4.1 to modernize the 3183 example. Thanks Z! 3185 o Make it more clear in the definition of "EID-to-RLOC Database" 3186 that all ETRs need to have the same database mapping. This 3187 reflects a comment from John Scudder. 3189 o Add a definition "Route-returnability" to the Definition of Terms 3190 section. 3192 o In section 9.2, add text to describe what the signature of 3193 traceroute packets can look like. 3195 o Removed references to Data Probe for introductory example. Data- 3196 probes are still part of the LISP design but not encouraged. 3198 o Added the definition for "LISP site" to the Definition of Terms" 3199 section. 3201 B.5. Changes to draft-ietf-lisp-06.txt 3203 Editorial based changes: 3205 o Posted December 2009. 3207 o Fix typo for flags in LISP data header. Changed from "4" to "5". 3209 o Add text to indicate that Map-Register messages must contain a 3210 computed UDP checksum. 3212 o Add definitions for PITR and PETR. 3214 o Indicate an AFI value of 0 is an unspecified address. 3216 o Indicate that the TTL field of a Map-Register is not used and set 3217 to 0 by the sender. This change makes this spec consistent with 3218 [LISP-MS]. 3220 o Change "... yield a packet size of L bytes" to "... yield a packet 3221 size greater than L bytes". 3223 o Clarify section 6.1.5 on what addresses and ports are used in Map- 3224 Reply messages. 3226 o Clarify that LSBs that go beyond the number of locators do not to 3227 be SMRed when the locator addresses are greater lexicographically 3228 than the locator in the existing locator-set. 3230 o Add Gregg, Srini, and Amit to acknowledgment section. 3232 o Clarify in the definition of a LISP header what is following the 3233 UDP header. 3235 o Clarify "verifying Map-Request" text in section 6.1.3. 3237 o Add Xu Xiaohu to the acknowledgment section for introducing the 3238 problem of overlapping EID-prefixes among multiple sites in an RRG 3239 email message. 3241 Design based changes: 3243 o Use stronger language to have the outer IPv4 header set DF=1 so we 3244 can avoid fragment reassembly in an ETR or PETR. This will also 3245 make IPv4 and IPv6 encapsulation have consistent behavior. 3247 o Map-Requests should not be sent in ECM with the Probe bit is set. 3248 These type of Map-Requests are used as RLOC-probes and are sent 3249 directly to locator addresses in the underlying network. 3251 o Add text in section 6.1.5 about returning all EID-prefixes in a 3252 Map-Reply sent by an ETR when there are overlapping EID-prefixes 3253 configure. 3255 o Add text in a new subsection of section 6.1.5 about dealing with 3256 Map-Replies with coarse EID-prefixes. 3258 B.6. Changes to draft-ietf-lisp-05.txt 3260 o Posted September 2009. 3262 o Added this Document Change Log appendix. 3264 o Added section indicating that encapsulated Map-Requests must use 3265 destination UDP port 4342. 3267 o Don't use AH in Map-Registers. Put key-id, auth-length, and auth- 3268 data in Map-Register payload. 3270 o Added Jari to acknowledgment section. 3272 o State the source-EID is set to 0 when using Map-Requests to 3273 refresh or RLOC-probe. 3275 o Make more clear what source-RLOC should be for a Map-Request. 3277 o The LISP-CONS authors thought that the Type definitions for CONS 3278 should be removed from this specification. 3280 o Removed nonce from Map-Register message, it wasn't used so no need 3281 for it. 3283 o Clarify what to do for unspecified Action bits for negative Map- 3284 Replies. Since No Action is a drop, make value 0 Drop. 3286 B.7. Changes to draft-ietf-lisp-04.txt 3288 o Posted September 2009. 3290 o How do deal with record count greater than 1 for a Map-Request. 3291 Damien and Joel comment. Joel suggests: 1) Specify that senders 3292 compliant with the current document will always set the count to 3293 1, and note that the count is included for future extensibility. 3294 2) Specify what a receiver compliant with the draft should do if 3295 it receives a request with a count greater than 1. Presumably, it 3296 should send some error back? 3298 o Add Fred Templin in acknowledgment section. 3300 o Add Margaret and Sam to the acknowledgment section for their great 3301 comments. 3303 o Say more about LAGs in the UDP section per Sam Hartman's comment. 3305 o Sam wants to use MAY instead of SHOULD for ignoring checksums on 3306 ETR. From the mailing list: "You'd need to word it as an ITR MAY 3307 send a zero checksum, an ETR MUST accept a 0 checksum and MAY 3308 ignore the checksum completely. And of course we'd need to 3309 confirm that can actually be implemented. In particular, hardware 3310 that verifies UDP checksums on receive needs to be checked to make 3311 sure it permits 0 checksums." 3313 o Margaret wants a reference to 3314 http://www.ietf.org/id/draft-eubanks-chimento-6man-00.txt. 3316 o Fix description in Map-Request section. Where we describe Map- 3317 Reply Record, change "R-bit" to "M-bit". 3319 o Add the mobility bit to Map-Replies. So PTRs don't probe so often 3320 for MNs but often enough to get mapping updates. 3322 o Indicate SHA1 can be used as well for Map-Registers. 3324 o More Fred comments on MTU handling. 3326 o Isidor comment about spec'ing better periodic Map-Registers. Will 3327 be fixed in draft-ietf-lisp-ms-02.txt. 3329 o Margaret's comment on gleaning: "The current specification does 3330 not make it clear how long gleaned map entries should be retained 3331 in the cache, nor does it make it clear how/ when they will be 3332 validated. The LISP spec should, at the very least, include a 3333 (short) default lifetime for gleaned entries, require that they be 3334 validated within a short period of time, and state that a new 3335 gleaned entry should never overwrite an entry that was obtained 3336 from the mapping system. The security implications of storing 3337 "gleaned" entries should also be explored in detail." 3339 o Add section on RLOC-probing per working group feedback. 3341 o Change "loc-reach-bits" to "loc-status-bits" per comment from 3342 Noel. 3344 o Remove SMR-bit from data-plane. Dino prefers to have it in the 3345 control plane only. 3347 o Change LISP header to allow a "Research Bit" so the Nonce and LSB 3348 fields can be turned off and used for another future purpose. For 3349 Luigi et al versioning convergence. 3351 o Add a N-bit to the data header suggested by Noel. Then the nonce 3352 field could be used when N is not 1. 3354 o Clarify that when E-bit is 0, the nonce field can be an echoed 3355 nonce or a random nonce. Comment from Jesper. 3357 o Indicate when doing data-gleaning that a verifying Map-Request is 3358 sent to the source-EID of the gleaned data packet so we can avoid 3359 map-cache corruption by a 3rd party. Comment from Pedro. 3361 o Indicate that a verifying Map-Request, for accepting mapping data, 3362 should be sent over the ALT (or to the EID). 3364 o Reference IPsec RFC 4302. Comment from Sam and Brian Weis. 3366 o Put E-bit in Map-Reply to tell ITRs that the ETR supports echo- 3367 noncing. Comment by Pedro and Dino. 3369 o Jesper made a comment to loosen the language about requiring the 3370 copy of inner TTL to outer TTL since the text to get mixed-AF 3371 traceroute to work would violate the "MUST" clause. Changed from 3372 MUST to SHOULD in section 5.3. 3374 B.8. Changes to draft-ietf-lisp-03.txt 3376 o Posted July 2009. 3378 o Removed loc-reach-bits longword from control packets per Damien 3379 comment. 3381 o Clarifications in MTU text from Roque. 3383 o Added text to indicate that the locator-set be sorted by locator 3384 address from Isidor. 3386 o Clarification text from John Zwiebel in Echo-Nonce section. 3388 B.9. Changes to draft-ietf-lisp-02.txt 3390 o Posted July 2009. 3392 o Encapsulation packet format change to add E-bit and make loc- 3393 reach-bits 32-bits in length. 3395 o Added Echo-Nonce Algorithm section. 3397 o Clarification how ECN bits are copied. 3399 o Moved S-bit in Map-Request. 3401 o Added P-bit in Map-Request and Map-Reply messages to anticipate 3402 RLOC-Probe Algorithm. 3404 o Added to Mobility section to reference [LISP-MN]. 3406 B.10. Changes to draft-ietf-lisp-01.txt 3408 o Posted 2 days after draft-ietf-lisp-00.txt in May 2009. 3410 o Defined LEID to be a "LISP EID". 3412 o Indicate encapsulation use IPv4 DF=0. 3414 o Added negative Map-Reply messages with drop, native-forward, and 3415 send-map-request actions. 3417 o Added Proxy-Map-Reply bit to Map-Register. 3419 B.11. Changes to draft-ietf-lisp-00.txt 3421 o Posted May 2009. 3423 o Rename of draft-farinacci-lisp-12.txt. 3425 o Acknowledgment to RRG. 3427 Authors' Addresses 3429 Dino Farinacci 3430 cisco Systems 3431 Tasman Drive 3432 San Jose, CA 95134 3433 USA 3435 Email: dino@cisco.com 3437 Vince Fuller 3438 cisco Systems 3439 Tasman Drive 3440 San Jose, CA 95134 3441 USA 3443 Email: vaf@cisco.com 3445 Dave Meyer 3446 cisco Systems 3447 170 Tasman Drive 3448 San Jose, CA 3449 USA 3451 Email: dmm@cisco.com 3453 Darrel Lewis 3454 cisco Systems 3455 170 Tasman Drive 3456 San Jose, CA 3457 USA 3459 Email: darlewis@cisco.com