idnits 2.17.1 draft-ietf-lisp-08.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 (August 13, 2010) is 4998 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reference: RFC 4634 (Obsoleted by RFC 6234) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-01) exists of draft-eubanks-chimento-6man-00 == Outdated reference: A later version (-10) exists of draft-ietf-lisp-alt-04 == Outdated reference: A later version (-04) exists of draft-meyer-lisp-cons-03 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-interworking-01 == Outdated reference: A later version (-10) exists of draft-farinacci-lisp-lcaf-00 == Outdated reference: A later version (-16) exists of draft-meyer-lisp-mn-03 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-ms-05 == Outdated reference: A later version (-14) exists of draft-ietf-lisp-multicast-03 == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-08 == Outdated reference: A later version (-05) exists of draft-narten-radir-problem-statement-00 == Outdated reference: A later version (-10) exists of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 (-02) exists of draft-iannone-lisp-mapping-versioning-01 Summary: 5 errors (**), 0 flaws (~~), 15 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: February 14, 2011 D. Lewis 6 cisco Systems 7 August 13, 2010 9 Locator/ID Separation Protocol (LISP) 10 draft-ietf-lisp-08 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 February 14, 2011. 50 Copyright Notice 52 Copyright (c) 2010 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 . . . . . . . . . . 35 87 6.1.6. Map-Register Message Format . . . . . . . . . . . . . 38 88 6.1.7. Encapsulated Control Message Format . . . . . . . . . 39 89 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 41 90 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 42 91 6.3.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . 45 92 6.3.2. RLOC Probing Algorithm . . . . . . . . . . . . . . . . 46 93 6.4. EID Reachability within a LISP Site . . . . . . . . . . . 47 94 6.5. Routing Locator Hashing . . . . . . . . . . . . . . . . . 47 95 6.6. Changing the Contents of EID-to-RLOC Mappings . . . . . . 48 96 6.6.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 49 97 6.6.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 49 98 6.6.3. Database Map Versioning . . . . . . . . . . . . . . . 51 99 7. Router Performance Considerations . . . . . . . . . . . . . . 52 100 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 53 101 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 54 102 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 54 103 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 55 104 8.4. LISP Functionality with Conventional NATs . . . . . . . . 55 105 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 56 106 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 57 107 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 57 108 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 57 109 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 59 110 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 59 111 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 59 112 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 59 113 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 61 114 10.5. LISP Mobile Node Mobility . . . . . . . . . . . . . . . . 61 115 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 63 116 12. Security Considerations . . . . . . . . . . . . . . . . . . . 64 117 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 65 118 14. References . . . . . . . . . . . . . . . . . . . . . . . . . . 66 119 14.1. Normative References . . . . . . . . . . . . . . . . . . . 66 120 14.2. Informative References . . . . . . . . . . . . . . . . . . 67 121 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 70 122 Appendix B. Document Change Log . . . . . . . . . . . . . . . . . 71 123 B.1. Changes to draft-ietf-lisp-08.txt . . . . . . . . . . . . 71 124 B.2. Changes to draft-ietf-lisp-07.txt . . . . . . . . . . . . 72 125 B.3. Changes to draft-ietf-lisp-06.txt . . . . . . . . . . . . 74 126 B.4. Changes to draft-ietf-lisp-05.txt . . . . . . . . . . . . 75 127 B.5. Changes to draft-ietf-lisp-04.txt . . . . . . . . . . . . 76 128 B.6. Changes to draft-ietf-lisp-03.txt . . . . . . . . . . . . 77 129 B.7. Changes to draft-ietf-lisp-02.txt . . . . . . . . . . . . 78 130 B.8. Changes to draft-ietf-lisp-01.txt . . . . . . . . . . . . 78 131 B.9. Changes to draft-ietf-lisp-00.txt . . . . . . . . . . . . 78 132 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 79 134 1. Requirements Notation 136 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 137 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 138 document are to be interpreted as described in [RFC2119]. 140 2. Introduction 142 This document describes the Locator/Identifier Separation Protocol 143 (LISP), which provides a set of functions for routers to exchange 144 information used to map from non-routeable Endpoint Identifiers 145 (EIDs) to routeable Routing Locators (RLOCs). It also defines a 146 mechanism for these LISP routers to encapsulate IP packets addressed 147 with EIDs for transmission across an Internet that uses RLOCs for 148 routing and forwarding. 150 Creation of LISP was initially motivated by discussions during the 151 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 152 October, 2006 (see [RFC4984]). A key conclusion of the workshop was 153 that the Internet routing and addressing system was not scaling well 154 in the face of the explosive growth of new sites; one reason for this 155 poor scaling is the increasing number of multi-homed and other sites 156 that cannot be addressed as part of topologically- or provider-based 157 aggregated prefixes. Additional work that more completely described 158 the problem statement may be found in [RADIR]. 160 A basic observation, made many years ago in early networking research 161 such as that documented in [CHIAPPA] and [RFC4984], is that using a 162 single address field for both identifying a device and for 163 determining where it is topologically located in the network requires 164 optimization along two conflicting axes: for routing to be efficient, 165 the address must be assigned topologically; for collections of 166 devices to be easily and effectively managed, without the need for 167 renumbering in response to topological change (such as that caused by 168 adding or removing attachment points to the network or by mobility 169 events), the address must explicitly not be tied to the topology. 171 The approach that LISP takes to solving the routing scalability 172 problem is to replace IP addresses with two new types of numbers: 173 Routing Locators (RLOCs), which are topologically assigned to network 174 attachment points (and are therefore amenable to aggregation) and 175 used for routing and forwarding of packets through the network; and 176 Endpoint Identifiers (EIDs), which are assigned independently from 177 the network topology, are used for numbering devices, and are 178 aggregated along administrative boundaries. LISP then defines 179 functions for mapping between the two numbering spaces and for 180 encapsulating traffic originated by devices using non-routeable EIDs 181 for transport across a network infrastructure that routes and 182 forwards using RLOCs. Both RLOCs and EIDs are syntactically- 183 identical to IP addresses; it is the semantics of how they are used 184 that differs. 186 This document describes the protocol that implements these functions. 187 The database which stores the mappings between EIDs and RLOCs is 188 explicitly a separate "module" to facilitate experimentation with a 189 variety of approaches. One database design that is being developed 190 and prototyped as part of the LISP working group work is [ALT]. 191 Others that have been described but not implemented include [CONS], 192 [EMACS], [RPMD], [NERD]. Finally, [LISP-MS], documents a general- 193 purpose service interface for accessing a mapping database; this 194 interface is intended to make the mapping database modular so that 195 different approaches can be tried without the need to modify 196 installed xTRs. 198 3. Definition of Terms 200 Provider Independent (PI) Addresses: PI addresses are an address 201 block assigned from a pool where blocks are not associated with 202 any particular location in the network (e.g. from a particular 203 service provider), and is therefore not topologically aggregatable 204 in the routing system. 206 Provider Assigned (PA) Addresses: PA addresses are an a address 207 block assigned to a site by each service provider to which a site 208 connects. Typically, each block is sub-block of a service 209 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 210 is aggregated into the larger block before being advertised into 211 the global Internet. Traditionally, IP multihoming has been 212 implemented by each multi-homed site acquiring its own, globally- 213 visible prefix. LISP uses only topologically-assigned and 214 aggregatable address blocks for RLOCs, eliminating this 215 demonstrably non-scalable practice. 217 Routing Locator (RLOC): A RLOC is an IPv4 or IPv6 address of an 218 egress tunnel router (ETR). A RLOC is the output of a EID-to-RLOC 219 mapping lookup. An EID maps to one or more RLOCs. Typically, 220 RLOCs are numbered from topologically-aggregatable blocks that are 221 assigned to a site at each point to which it attaches to the 222 global Internet; where the topology is defined by the connectivity 223 of provider networks, RLOCs can be thought of as PA addresses. 224 Multiple RLOCs can be assigned to the same ETR device or to 225 multiple ETR devices at a site. 227 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 228 IPv6) value used in the source and destination address fields of 229 the first (most inner) LISP header of a packet. The host obtains 230 a destination EID the same way it obtains an destination address 231 today, for example through a Domain Name System (DNS) [RFC1034] 232 lookup or Session Invitation Protocol (SIP) [RFC3261] exchange. 233 The source EID is obtained via existing mechanisms used to set a 234 host's "local" IP address. An EID is allocated to a host from an 235 EID-prefix block associated with the site where the host is 236 located. An EID can be used by a host to refer to other hosts. 237 EIDs MUST NOT be used as LISP RLOCs. Note that EID blocks may be 238 assigned in a hierarchical manner, independent of the network 239 topology, to facilitate scaling of the mapping database. In 240 addition, an EID block assigned to a site may have site-local 241 structure (subnetting) for routing within the site; this structure 242 is not visible to the global routing system. When used in 243 discussions with other Locator/ID separation proposals, a LISP EID 244 will be called a "LEID". Throughout this document, any references 245 to "EID" refers to an LEID. 247 EID-prefix: An EID-prefix is a power-of-two block of EIDs which are 248 allocated to a site by an address allocation authority. EID- 249 prefixes are associated with a set of RLOC addresses which make up 250 a "database mapping". EID-prefix allocations can be broken up 251 into smaller blocks when an RLOC set is to be associated with the 252 smaller EID-prefix. A globally routed address block (whether PI 253 or PA) is not an EID-prefix. However, a globally routed address 254 block may be removed from global routing and reused as an EID- 255 prefix. A site that receives an explicitly allocated EID-prefix 256 may not use that EID-prefix as a globally routed prefix assigned 257 to RLOCs. 259 End-system: An end-system is an IPv4 or IPv6 device that originates 260 packets with a single IPv4 or IPv6 header. The end-system 261 supplies an EID value for the destination address field of the IP 262 header when communicating globally (i.e. outside of its routing 263 domain). An end-system can be a host computer, a switch or router 264 device, or any network appliance. 266 Ingress Tunnel Router (ITR): An ITR is a router which accepts an IP 267 packet with a single IP header (more precisely, an IP packet that 268 does not contain a LISP header). The router treats this "inner" 269 IP destination address as an EID and performs an EID-to-RLOC 270 mapping lookup. The router then prepends an "outer" IP header 271 with one of its globally-routable RLOCs in the source address 272 field and the result of the mapping lookup in the destination 273 address field. Note that this destination RLOC may be an 274 intermediate, proxy device that has better knowledge of the EID- 275 to-RLOC mapping closer to the destination EID. In general, an ITR 276 receives IP packets from site end-systems on one side and sends 277 LISP-encapsulated IP packets toward the Internet on the other 278 side. 280 Specifically, when a service provider prepends a LISP header for 281 Traffic Engineering purposes, the router that does this is also 282 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 283 on the outer destination address (the originating ITR's supplied 284 RLOC) or the inner destination address (the originating hosts 285 supplied EID). 287 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 288 network that prepends an additional LISP header for Traffic 289 Engineering purposes. 291 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 292 packet where the destination address in the "outer" IP header is 293 one of its own RLOCs. The router strips the "outer" header and 294 forwards the packet based on the next IP header found. In 295 general, an ETR receives LISP-encapsulated IP packets from the 296 Internet on one side and sends decapsulated IP packets to site 297 end-systems on the other side. ETR functionality does not have to 298 be limited to a router device. A server host can be the endpoint 299 of a LISP tunnel as well. 301 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 302 network that strips an outer LISP header for Traffic Engineering 303 purposes. 305 xTR: A xTR is a reference to an ITR or ETR when direction of data 306 flow is not part of the context description. xTR refers to the 307 router that is the tunnel endpoint. Used synonymously with the 308 term "Tunnel Router". For example, "An xTR can be located at the 309 Customer Edge (CE) router", meaning both ITR and ETR functionality 310 is at the CE router. 312 EID-to-RLOC Cache: The EID-to-RLOC cache is a short-lived, on- 313 demand table in an ITR that stores, tracks, and is responsible for 314 timing-out and otherwise validating EID-to-RLOC mappings. This 315 cache is distinct from the full "database" of EID-to-RLOC 316 mappings, it is dynamic, local to the ITR(s), and relatively small 317 while the database is distributed, relatively static, and much 318 more global in scope. 320 EID-to-RLOC Database: The EID-to-RLOC database is a global 321 distributed database that contains all known EID-prefix to RLOC 322 mappings. Each potential ETR typically contains a small piece of 323 the database: the EID-to-RLOC mappings for the EID prefixes 324 "behind" the router. These map to one of the router's own, 325 globally-visible, IP addresses. The same database mapping entries 326 MUST be configured on all ETRs for a given site. That is, the 327 EID-prefixes for the site and locator-set for each EID-prefix MUST 328 be the same on all ETRs so they consistently send Map-Reply 329 messages with the same database mapping contents. 331 Recursive Tunneling: Recursive tunneling occurs when a packet has 332 more than one LISP IP header. Additional layers of tunneling may 333 be employed to implement traffic engineering or other re-routing 334 as needed. When this is done, an additional "outer" LISP header 335 is added and the original RLOCs are preserved in the "inner" 336 header. Any references to tunnels in this specification refers to 337 dynamic encapsulating tunnels and never are they statically 338 configured. 340 Reencapsulating Tunnels: Reencapsulating tunneling occurs when a 341 packet has no more than one LISP IP header (two IP headers total) 342 and when it needs to be diverted to new RLOC, an ETR can 343 decapsulate the packet (remove the LISP header) and prepends a new 344 tunnel header, with new RLOC, on to the packet. Doing this allows 345 a packet to be re-routed by the re-encapsulating router without 346 adding the overhead of additional tunnel headers. Any references 347 to tunnels in this specification refers to dynamic encapsulating 348 tunnels and never are they statically configured. 350 LISP Header: a term used in this document to refer to the outer 351 IPv4 or IPv6 header, a UDP header, and a LISP-specific 8-byte 352 header that follows the UDP header, an ITR prepends or an ETR 353 strips. 355 Address Family Identifier (AFI): a term used to describe an address 356 encoding in a packet. An address family currently pertains to an 357 IPv4 or IPv6 address. See [AFI] and [RFC1700] for details. An 358 AFI value of 0 used in this specification indicates an unspecified 359 encoded address where the length of the address is 0 bytes 360 following the 16-bit AFI value of 0. 362 Negative Mapping Entry: A negative mapping entry, also known as a 363 negative cache entry, is an EID-to-RLOC entry where an EID-prefix 364 is advertised or stored with no RLOCs. That is, the locator-set 365 for the EID-to-RLOC entry is empty or has an encoded locator count 366 of 0. This type of entry could be used to describe a prefix from 367 a non-LISP site, which is explicitly not in the mapping database. 368 There are a set of well defined actions that are encoded in a 369 Negative Map-Reply. 371 Data Probe: A data-probe is a LISP-encapsulated data packet where 372 the inner header destination address equals the outer header 373 destination address used to trigger a Map-Reply by a decapsulating 374 ETR. In addition, the original packet is decapsulated and 375 delivered to the destination host. A Data Probe is used in some 376 of the mapping database designs to "probe" or request a Map-Reply 377 from an ETR; in other cases, Map-Requests are used. See each 378 mapping database design for details. 380 Proxy ITR (PITR): A PITR is also known as a PTR is defined and 381 described in [INTERWORK], a PITR acts like an ITR but does so on 382 behalf of non-LISP sites which send packets to destinations at 383 LISP sites. 385 Proxy ETR (PETR): A PETR is defined and described in [INTERWORK], a 386 PETR acts like an ETR but does so on behalf of LISP sites which 387 send packets to destinations at non-LISP sites. 389 Route-returnability: is an assumption that the underlying routing 390 system will deliver packets to the destination. When combined 391 with a nonce that is provided by a sender and returned by a 392 receiver limits off-path data insertion. 394 LISP site: is a set of routers in an edge network that are under a 395 single technical administration. LISP routers which reside in the 396 edge network are the demarcation points to separate the edge 397 network from the core network. 399 Client-side: a term used in this document to indicate a connection 400 initiation attempt by an EID. The ITR(s) at the LISP site are the 401 first to get involved in obtaining database map cache entries by 402 sending Map-Request messages. 404 Server-side: a term used in this document to indicate a connection 405 initiation attempt is being accepted for a destination EID. The 406 ETR(s) at the destination LISP site are the first to send Map- 407 Replies to the source site initiating the connection. The ETR(s) 408 at this destination site can obtain mappings by gleaning 409 information from Map-Requests, Data-Probes, or encapsulated 410 packets. 412 4. Basic Overview 414 One key concept of LISP is that end-systems (hosts) operate the same 415 way they do today. The IP addresses that hosts use for tracking 416 sockets, connections, and for sending and receiving packets do not 417 change. In LISP terminology, these IP addresses are called Endpoint 418 Identifiers (EIDs). 420 Routers continue to forward packets based on IP destination 421 addresses. When a packet is LISP encapsulated, these addresses are 422 referred to as Routing Locators (RLOCs). Most routers along a path 423 between two hosts will not change; they continue to perform routing/ 424 forwarding lookups on the destination addresses. For routers between 425 the source host and the ITR as well as routers from the ETR to the 426 destination host, the destination address is an EID. For the routers 427 between the ITR and the ETR, the destination address is an RLOC. 429 Another key LISP concept is the "Tunnel Router". A tunnel router 430 prepends LISP headers on host-originated packets and strip them prior 431 to final delivery to their destination. The IP addresses in this 432 "outer header" are RLOCs. During end-to-end packet exchange between 433 two Internet hosts, an ITR prepends a new LISP header to each packet 434 and an egress tunnel router strips the new header. The ITR performs 435 EID-to-RLOC lookups to determine the routing path to the ETR, which 436 has the RLOC as one of its IP addresses. 438 Some basic rules governing LISP are: 440 o End-systems (hosts) only send to addresses which are EIDs. They 441 don't know addresses are EIDs versus RLOCs but assume packets get 442 to LISP routers, which in turn, deliver packets to the destination 443 the end-system has specified. 445 o EIDs are always IP addresses assigned to hosts. 447 o LISP routers mostly deal with Routing Locator addresses. See 448 details later in Section 4.1 to clarify what is meant by "mostly". 450 o RLOCs are always IP addresses assigned to routers; preferably, 451 topologically-oriented addresses from provider CIDR blocks. 453 o When a router originates packets it may use as a source address 454 either an EID or RLOC. When acting as a host (e.g. when 455 terminating a transport session such as SSH, TELNET, or SNMP), it 456 may use an EID that is explicitly assigned for that purpose. An 457 EID that identifies the router as a host MUST NOT be used as an 458 RLOC; an EID is only routable within the scope of a site. A 459 typical BGP configuration might demonstrate this "hybrid" EID/RLOC 460 usage where a router could use its "host-like" EID to terminate 461 iBGP sessions to other routers in a site while at the same time 462 using RLOCs to terminate eBGP sessions to routers outside the 463 site. 465 o EIDs are not expected to be usable for global end-to-end 466 communication in the absence of an EID-to-RLOC mapping operation. 467 They are expected to be used locally for intra-site communication. 469 o EID prefixes are likely to be hierarchically assigned in a manner 470 which is optimized for administrative convenience and to 471 facilitate scaling of the EID-to-RLOC mapping database. The 472 hierarchy is based on a address allocation hierarchy which is 473 independent of the network topology. 475 o EIDs may also be structured (subnetted) in a manner suitable for 476 local routing within an autonomous system. 478 An additional LISP header may be prepended to packets by a TE-ITR 479 when re-routing of the path for a packet is desired. An obvious 480 instance of this would be an ISP router that needs to perform traffic 481 engineering for packets flowing through its network. In such a 482 situation, termed Recursive Tunneling, an ISP transit acts as an 483 additional ingress tunnel router and the RLOC it uses for the new 484 prepended header would be either a TE-ETR within the ISP (along 485 intra-ISP traffic engineered path) or a TE-ETR within another ISP (an 486 inter-ISP traffic engineered path, where an agreement to build such a 487 path exists). 489 In order to avoid excessive packet overhead as well as possible 490 encapsulation loops, this document mandates that a maximum of two 491 LISP headers can be prepended to a packet. It is believed two 492 headers is sufficient, where the first prepended header is used at a 493 site for Location/Identity separation and second prepended header is 494 used inside a service provider for Traffic Engineering purposes. 496 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 497 For example, the ITR for a particular end-to-end packet exchange 498 might be the first-hop or default router within a site for the source 499 host. Similarly, the egress tunnel router might be the last-hop 500 router directly-connected to the destination host. Another example, 501 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 502 could be the site's border router at the service provider attachment 503 point. Mixing and matching of site-operated, ISP-operated, and other 504 tunnel routers is allowed for maximum flexibility. See Section 8 for 505 more details. 507 4.1. Packet Flow Sequence 509 This section provides an example of the unicast packet flow with the 510 following conditions: 512 o Source host "host1.abc.com" is sending a packet to 513 "host2.xyz.com", exactly what host1 would do if the site was not 514 using LISP. 516 o Each site is multi-homed, so each tunnel router has an address 517 (RLOC) assigned from the service provider address block for each 518 provider to which that particular tunnel router is attached. 520 o The ITR(s) and ETR(s) are directly connected to the source and 521 destination, respectively, but the source and destination can be 522 located anywhere in LISP site. 524 o Map-Requests can be sent on the underlying routing system topology 525 or over an alternative topology [ALT]. 527 o Map-Replies are sent on the underlying routing system topology. 529 Client host1.abc.com wants to communicate with server host2.xyz.com: 531 1. host1.abc.com wants to open a TCP connection to host2.xyz.com. 532 It does a DNS lookup on host2.xyz.com. An A/AAAA record is 533 returned. This address is the destination EID. The locally- 534 assigned address of host1.abc.com is used as the source EID. An 535 IPv4 or IPv6 packet is built and forwarded through the LISP site 536 as a normal IP packet until it reaches a LISP ITR. 538 2. The LISP ITR must be able to map the EID destination to an RLOC 539 of one of the ETRs at the destination site. The specific method 540 used to do this is not described in this example. See [ALT] or 541 [CONS] for possible solutions. 543 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be 544 rate-limited. 546 4. When an alternate mapping system is not in use, the Map-Request 547 packet is routed through the underlying routing system. 548 Otherwise, the Map-Request packet is routed on an alternate 549 logical topology. In either case, when the Map-Request arrives 550 at one of the ETRs at the destination site, it will process the 551 packet as a control message. 553 5. The ETR looks at the destination EID of the Map-Request and 554 matches it against the prefixes in the ETR's configured EID-to- 555 RLOC mapping database. This is the list of EID-prefixes the ETR 556 is supporting for the site it resides in. If there is no match, 557 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 558 returned to the ITR. 560 6. The ITR receives the Map-Reply message, parses the message (to 561 check for format validity) and stores the mapping information 562 from the packet. This information is stored in the ITR's EID-to- 563 RLOC mapping cache. Note that the map cache is an on-demand 564 cache. An ITR will manage its map cache in such a way that 565 optimizes for its resource constraints. 567 7. Subsequent packets from host1.abc.com to host2.xyz.com will have 568 a LISP header prepended by the ITR using the appropriate RLOC as 569 the LISP header destination address learned from the ETR. Note 570 the packet may be sent to a different ETR than the one which 571 returned the Map-Reply due to the source site's hashing policy or 572 the destination site's locator-set policy. 574 8. The ETR receives these packets directly (since the destination 575 address is one of its assigned IP addresses), strips the LISP 576 header and forwards the packets to the attached destination host. 578 In order to eliminate the need for a mapping lookup in the reverse 579 direction, an ETR MAY create a cache entry that maps the source EID 580 (inner header source IP address) to the source RLOC (outer header 581 source IP address) in a received LISP packet. Such a cache entry is 582 termed a "gleaned" mapping and only contains a single RLOC for the 583 EID in question. More complete information about additional RLOCs 584 SHOULD be verified by sending a LISP Map-Request for that EID. Both 585 ITR and the ETR may also influence the decision the other makes in 586 selecting an RLOC. See Section 6 for more details. 588 5. LISP Encapsulation Details 590 Since additional tunnel headers are prepended, the packet becomes 591 larger and can exceed the MTU of any link traversed from the ITR to 592 the ETR. It is recommended in IPv4 that packets do not get 593 fragmented as they are encapsulated by the ITR. Instead, the packet 594 is dropped and an ICMP Too Big message is returned to the source. 596 This specification recommends that implementations support for one of 597 the proposed fragmentation and reassembly schemes. These two simple 598 existing schemes are detailed in Section 5.4. 600 5.1. LISP IPv4-in-IPv4 Header Format 602 0 1 2 3 603 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 1 604 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 605 / |Version| IHL |Type of Service| Total Length | 606 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 607 | | Identification |Flags| Fragment Offset | 608 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 609 OH | Time to Live | Protocol = 17 | Header Checksum | 610 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 611 | | Source Routing Locator | 612 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 613 \ | Destination Routing Locator | 614 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 615 / | Source Port = xxxx | Dest Port = 4341 | 616 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 617 \ | UDP Length | UDP Checksum | 618 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 619 L |N|L|E|V|I|flags| Nonce/Map-Version | 620 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 621 S / | Instance ID/Locator Status Bits | 622 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 623 / |Version| IHL |Type of Service| Total Length | 624 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 625 | | Identification |Flags| Fragment Offset | 626 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 627 IH | Time to Live | Protocol | Header Checksum | 628 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 629 | | Source EID | 630 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 631 \ | Destination EID | 632 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 634 5.2. LISP IPv6-in-IPv6 Header Format 636 0 1 2 3 637 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 1 638 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 639 / |Version| Traffic Class | Flow Label | 640 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 642 | | Payload Length | Next Header=17| Hop Limit | 643 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 644 | | 645 O + + 646 u | | 647 t + Source Routing Locator + 648 e | | 649 r + + 650 | | 651 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 652 d | | 653 r + + 654 | | 655 ^ + Destination Routing Locator + 656 | | | 657 \ + + 658 \ | | 659 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 660 / | Source Port = xxxx | Dest Port = 4341 | 661 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 662 \ | UDP Length | UDP Checksum | 663 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 664 L |N|L|E|V|I|flags| Nonce/Map-Version | 665 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 666 S / | Instance ID/Locator Status Bits | 667 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 668 / |Version| Traffic Class | Flow Label | 669 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 670 / | Payload Length | Next Header | Hop Limit | 671 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 672 | | 673 I + + 674 n | | 675 n + Source EID + 676 e | | 677 r + + 678 | | 679 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 680 d | | 681 r + + 682 | | 683 ^ + Destination EID + 684 \ | | 685 \ + + 686 \ | | 687 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 689 5.3. Tunnel Header Field Descriptions 691 Inner Header: The inner header is the header on the datagram 692 received from the originating host. The source and destination IP 693 addresses are EIDs. 695 Outer Header: The outer header is a new header prepended by an ITR. 696 The address fields contain RLOCs obtained from the ingress 697 router's EID-to-RLOC cache. The IP protocol number is "UDP (17)" 698 from [RFC0768]. The DF bit of the Flags field is set to 0 when 699 the method in Section 5.4.1 is used and set to 1 when the method 700 in Section 5.4.2 is used. 702 UDP Header: The UDP header contains a ITR selected source port when 703 encapsulating a packet. See Section 6.5 for details on the hash 704 algorithm used to select a source port based on the 5-tuple of the 705 inner header. The destination port MUST be set to the well-known 706 IANA assigned port value 4341. 708 UDP Checksum: The UDP checksum field SHOULD be transmitted as zero 709 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 710 [UDP-TUNNELS]. When a packet with a zero UDP checksum is received 711 by an ETR, the ETR MUST accept the packet for decapsulation. When 712 an ITR transmits a non-zero value for the UDP checksum, it MUST 713 send a correctly computed value in this field. When an ETR 714 receives a packet with a non-zero UDP checksum, it MAY choose to 715 verify the checksum value. If it chooses to perform such 716 verification, and the verification fails, the packet MUST be 717 silently dropped. If the ETR chooses not to perform the 718 verification, or performs the verification successfully, the 719 packet MUST be accepted for decapsulation. The handling of UDP 720 checksums for all tunneling protocols, including LISP, is under 721 active discussion within the IETF. When that discussion 722 concludes, any necessary changes will be made to align LISP with 723 the outcome of the broader discussion. 725 UDP Length: The UDP length field is for an IPv4 encapsulated packet, 726 the inner header Total Length plus the UDP and LISP header lengths 727 are used. For an IPv6 encapsulated packet, the inner header 728 Payload Length plus the size of the IPv6 header (40 bytes) plus 729 the size of the UDP and LISP headers are used. The UDP header 730 length is 8 bytes. 732 N: The N bit is the nonce-present bit. When this bit is set to 1, 733 the low-order 24-bits of the first 32-bits of the LISP header 734 contains a Nonce. See Section 6.3.1 for details. Both N and V 735 bits MUST NOT be set in the same packet. If they are, a 736 decapsulating ETR MUST treat the "Nonce/Map-Version" field as 737 having a Nonce value present. 739 L: The L bit is the Locator-Status-Bits field enabled bit. When this 740 bit is set to 1, the Locator-Status-Bits in the second 32-bits of 741 the LISP header are in use. 743 x 1 x x 0 x x x 744 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 745 |N|L|E|V|I|flags| Nonce/Map-Version | 746 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 747 | Locator Status Bits | 748 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 750 E: The E bit is the echo-nonce-request bit. When this bit is set to 751 1, the N bit MUST be 1. This bit SHOULD be ignored and has no 752 meaning when the N bit is set to 0. See Section 6.3.1 for 753 details. 755 V: The B bit is the Map-Version present bit. When this bit is set to 756 1, the N bit MUST be 0. Refer to Section 6.6.3 for more details. 757 This bit indicates that the first 4 bytes of the LISP header is 758 encoded as: 760 0 x 0 1 x x x x 761 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 762 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version | 763 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 764 | Instance ID/Locator Status Bits | 765 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 767 I: The I bit is the Instance ID bit. See Section 5.5 for more 768 details. When this bit is set to 1, the Locator Status Bits field 769 is reduced to 8-bits and the high-order 24-bits are used as an 770 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 771 are transmitted as zero and ignored on receipt. The format of the 772 last 4 bytes of the LISP header would look like: 774 x x x x 1 x x x 775 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 776 |N|L|E|V|I|flags| Nonce/Map-Version | 777 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 778 | Instance ID | LSBs | 779 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 781 flags: The flags field is a 3-bit field is reserved for future flag 782 use. It is set to 0 on transmit and ignored on receipt. 784 LISP Nonce: The LISP nonce field is a 24-bit value that is randomly 785 generated by an ITR when the N-bit is set to 1. The nonce is also 786 used when the E-bit is set to request the nonce value to be echoed 787 by the other side when packets are returned. When the E-bit is 788 clear but the N-bit is set, a remote ITR is either echoing a 789 previously requested echo-nonce or providing a random nonce. See 790 Section 6.3.1 for more details. 792 LISP Locator Status Bits: The locator status bits field in the LISP 793 header is set by an ITR to indicate to an ETR the up/down status 794 of the Locators in the source site. Each RLOC in a Map-Reply is 795 assigned an ordinal value from 0 to n-1 (when there are n RLOCs in 796 a mapping entry). The Locator Status Bits are numbered from 0 to 797 n-1 from the least significant bit of field. The field is 32-bits 798 when the I-bit is set to 0 and is 8 bits when the I-bit is set to 799 1. When a Locator Status Bit is set to 1, the ITR is indicating 800 to the ETR the RLOC associated with the bit ordinal has up status. 801 See Section 6.3 for details on how an ITR can determine the status 802 of other ITRs at the same site. When a site has multiple EID- 803 prefixes which result in multiple mappings (where each could have 804 a different locator-set), the Locator Status Bits setting in an 805 encapsulated packet MUST reflect the mapping for the EID-prefix 806 that the inner-header source EID address matches. 808 When doing ITR/PITR encapsulation: 810 o The outer header Time to Live field (or Hop Limit field, in case 811 of IPv6) SHOULD be copied from the inner header Time to Live 812 field. 814 o The outer header Type of Service field (or the Traffic Class 815 field, in the case of IPv6) SHOULD be copied from the inner header 816 Type of Service field (with one caveat, see below). 818 When doing ETR/PETR decapsulation: 820 o The inner header Time to Live field (or Hop Limit field, in case 821 of IPv6) SHOULD be copied from the outer header Time to Live 822 field, when the Time to Live field of the outer header is less 823 than the Time to Live of the inner header. Failing to perform 824 this check can cause the Time to Live of the inner header to 825 increment across encapsulation/decapsulation cycle. This check is 826 also performed when doing initial encapsulation when a packet 827 comes to an ITR or PITR destined for a LISP site. 829 o The inner header Type of Service field (or the Traffic Class 830 field, in the case of IPv6) SHOULD be copied from the outer header 831 Type of Service field (with one caveat, see below). 833 Note if an ETR/PETR is also an ITR/PITR and choose to reencapsulate 834 after decapsulating, the net effect of this is that the new outer 835 header will carry the same Time to Live as the old outer header. 837 Copying the TTL serves two purposes: first, it preserves the distance 838 the host intended the packet to travel; second, and more importantly, 839 it provides for suppression of looping packets in the event there is 840 a loop of concatenated tunnels due to misconfiguration. See 841 Section 9.3 for TTL exception handling for traceroute packets. 843 The ECN field occupies bits 6 and 7 of both the IPv4 Type of Service 844 field and the IPv6 Traffic Class field [RFC3168]. The ECN field 845 requires special treatment in order to avoid discarding indications 846 of congestion [RFC3168]. ITR encapsulation MUST copy the 2-bit ECN 847 field from the inner header to the outer header. Re-encapsulation 848 MUST copy the 2-bit ECN field from the stripped outer header to the 849 new outer header. If the ECN field contains a congestion indication 850 codepoint (the value is '11', the Congestion Experienced (CE) 851 codepoint), then ETR decapsulation MUST copy the 2-bit ECN field from 852 the stripped outer header to the surviving inner header that is used 853 to forward the packet beyond the ETR. These requirements preserve 854 Congestion Experienced (CE) indications when a packet that uses ECN 855 traverses a LISP tunnel and becomes marked with a CE indication due 856 to congestion between the tunnel endpoints. 858 5.4. Dealing with Large Encapsulated Packets 860 This section proposes two simple mechanisms to deal with packets that 861 exceed the path MTU between the ITR and ETR. 863 It is left to the implementor to decide if the stateless or stateful 864 mechanism should be implemented. Both or neither can be used since 865 it is a local decision in the ITR regarding how to deal with MTU 866 issues, and sites can interoperate with differing mechanisms. 868 Both stateless and stateful mechanisms also apply to Reencapsulating 869 and Recursive Tunneling. So any actions below referring to an ITR 870 also apply to an TE-ITR. 872 5.4.1. A Stateless Solution to MTU Handling 874 An ITR stateless solution to handle MTU issues is described as 875 follows: 877 1. Define an architectural constant S for the maximum size of a 878 packet, in bytes, an ITR would like to receive from a source 879 inside of its site. 881 2. Define L to be the maximum size, in bytes, a packet of size S 882 would be after the ITR prepends the LISP header, UDP header, and 883 outer network layer header of size H. 885 3. Calculate: S + H = L. 887 When an ITR receives a packet from a site-facing interface and adds H 888 bytes worth of encapsulation to yield a packet size greater than L 889 bytes, it resolves the MTU issue by first splitting the original 890 packet into 2 equal-sized fragments. A LISP header is then prepended 891 to each fragment. The size of the encapsulated fragments is then 892 (S/2 + H), which is less than the ITR's estimate of the path MTU 893 between the ITR and its correspondent ETR. 895 When an ETR receives encapsulated fragments, it treats them as two 896 individually encapsulated packets. It strips the LISP headers then 897 forwards each fragment to the destination host of the destination 898 site. The two fragments are reassembled at the destination host into 899 the single IP datagram that was originated by the source host. 901 This behavior is performed by the ITR when the source host originates 902 a packet with the DF field of the IP header is set to 0. When the DF 903 field of the IP header is set to 1, or the packet is an IPv6 packet 904 originated by the source host, the ITR will drop the packet when the 905 size is greater than L, and sends an ICMP Too Big message to the 906 source with a value of S, where S is (L - H). 908 When the outer header encapsulation uses an IPv4 header, an 909 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 910 can be avoided. An implementation MAY set the DF bit in such headers 911 to 0 if it has good reason to believe there are unresolvable path MTU 912 issues between the sending ITR and the receiving ETR. 914 This specification recommends that L be defined as 1500. 916 5.4.2. A Stateful Solution to MTU Handling 918 An ITR stateful solution to handle MTU issues is described as follows 919 and was first introduced in [OPENLISP]: 921 1. The ITR will keep state of the effective MTU for each locator per 922 mapping cache entry. The effective MTU is what the core network 923 can deliver along the path between ITR and ETR. 925 2. When an IPv6 encapsulated packet or an IPv4 encapsulated packet 926 with DF bit set to 1, exceeds what the core network can deliver, 927 one of the intermediate routers on the path will send an ICMP Too 928 Big message to the ITR. The ITR will parse the ICMP message to 929 determine which locator is affected by the effective MTU change 930 and then record the new effective MTU value in the mapping cache 931 entry. 933 3. When a packet is received by the ITR from a source inside of the 934 site and the size of the packet is greater than the effective MTU 935 stored with the mapping cache entry associated with the 936 destination EID the packet is for, the ITR will send an ICMP Too 937 Big message back to the source. The packet size advertised by 938 the ITR in the ICMP Too Big message is the effective MTU minus 939 the LISP encapsulation length. 941 Even though this mechanism is stateful, it has advantages over the 942 stateless IP fragmentation mechanism, by not involving the 943 destination host with reassembly of ITR fragmented packets. 945 5.5. Using Virtualization and Segmentation with LISP 947 When multiple organizations inside of a LISP site are using private 948 addresses [RFC1918] as EID-prefixes, their address spaces MUST remain 949 segregated due to possible address duplication. An Instance ID in 950 the address encoding can aid in making the entire AFI based address 951 unique. See [LCAF] for details for a possible address encoding. The 952 LCAF encoding is an area for further study. 954 An Instance ID can be carried in a LISP encapsulated packet. An ITR 955 that prepends a LISP header, will copy a 24-bit value, used by the 956 LISP router to uniquely identify the address space. The value is 957 copied to the Instance ID field of the LISP header and the I-bit is 958 set to 1. 960 When an ETR decapsulates a packet, the Instance ID from the LISP 961 header is used as a table identifier to locate the forwarding table 962 to use for the inner destination EID lookup. 964 For example, a 802.1Q VLAN tag or VPN identifier could be used as a 965 24-bit Instance ID. 967 6. EID-to-RLOC Mapping 969 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats 971 The following new UDP packet types are used to retrieve EID-to-RLOC 972 mappings: 974 0 1 2 3 975 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 1 976 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 977 |Version| IHL |Type of Service| Total Length | 978 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 979 | Identification |Flags| Fragment Offset | 980 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 981 | Time to Live | Protocol = 17 | Header Checksum | 982 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 983 | Source Routing Locator | 984 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 985 | Destination Routing Locator | 986 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 987 / | Source Port | Dest Port | 988 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 989 \ | UDP Length | UDP Checksum | 990 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 991 | | 992 | LISP Message | 993 | | 994 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 996 0 1 2 3 997 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 1 998 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 999 |Version| Traffic Class | Flow Label | 1000 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1001 | Payload Length | Next Header=17| Hop Limit | 1002 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1003 | | 1004 + + 1005 | | 1006 + Source Routing Locator + 1007 | | 1008 + + 1009 | | 1010 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1011 | | 1012 + + 1013 | | 1014 + Destination Routing Locator + 1015 | | 1016 + + 1017 | | 1018 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1019 / | Source Port | Dest Port | 1020 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1021 \ | UDP Length | UDP Checksum | 1022 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1023 | | 1024 | LISP Message | 1025 | | 1026 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1028 The LISP UDP-based messages are the Map-Request and Map-Reply 1029 messages. When a UDP Map-Request is sent, the UDP source port is 1030 chosen by the sender and the destination UDP port number is set to 1031 4342. When a UDP Map-Reply is sent, the source UDP port number is 1032 set to 4342 and the destination UDP port number is copied from the 1033 source port of either the Map-Request or the invoking data packet. 1035 The UDP Length field will reflect the length of the UDP header and 1036 the LISP Message payload. 1038 The UDP Checksum is computed and set to non-zero for Map-Request, 1039 Map-Reply, Map-Register and ECM control messages. It MUST be checked 1040 on receipt and if the checksum fails, the packet MUST be dropped. 1042 LISP-CONS [CONS] uses TCP to send LISP control messages. The format 1043 of control messages includes the UDP header so the checksum and 1044 length fields can be used to protect and delimit message boundaries. 1046 This main LISP specification is the authoritative source for message 1047 format definitions for the Map-Request and Map-Reply messages. 1049 6.1.1. LISP Packet Type Allocations 1051 This section will be the authoritative source for allocating LISP 1052 Type values. Current allocations are: 1054 Reserved: 0 b'0000' 1055 LISP Map-Request: 1 b'0001' 1056 LISP Map-Reply: 2 b'0010' 1057 LISP Map-Register: 3 b'0011' 1058 LISP Encapsulated Control Message: 8 b'1000' 1060 6.1.2. Map-Request Message Format 1062 0 1 2 3 1063 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 1064 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1065 |Type=1 |A|M|P|S| Reserved | IRC | Record Count | 1066 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1067 | Nonce . . . | 1068 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1069 | . . . Nonce | 1070 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1071 | Source-EID-AFI | Source EID Address ... | 1072 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1073 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... | 1074 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1075 | ... | 1076 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1077 | ITR-RLOC-AFI n | ITR-RLOC Address n ... | 1078 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1079 / | Reserved | EID mask-len | EID-prefix-AFI | 1080 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1081 \ | EID-prefix ... | 1082 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1083 | Map-Reply Record ... | 1084 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1085 | Mapping Protocol Data | 1086 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1088 Packet field descriptions: 1090 Type: 1 (Map-Request) 1092 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 1093 Requests sent by an ITR. 1095 M: When set, it indicates a Map-Reply Record segment is included in 1096 the Map-Request. 1098 P: This is the probe-bit which indicates that a Map-Request SHOULD be 1099 treated as a locator reachability probe. The receiver SHOULD 1100 respond with a Map-Reply with the probe-bit set, indicating the 1101 Map-Reply is a locator reachability probe reply, with the nonce 1102 copied from the Map-Request. See Section 6.3.2 for more details. 1104 S: This is the SMR bit. See Section 6.6.2 for details. 1106 Reserved: Set to 0 on transmission and ignored on receipt. 1108 IRC: This 5-bit field is the ITR-RLOC Count which encodes the 1109 additional number of (ITR-RLOC-AFI, ITR-RLOC Address) fields 1110 present in this message. At least one (ITR-RLOC-AFI, ITR-RLOC- 1111 Address) pair must always be encoded. Multiple ITR-RLOC Address 1112 fields are used so a Map-Replier can select which destination 1113 address to use for a Map-Reply. The IRC value ranges from 0 to 1114 31, and for a value of 1, there are 2 ITR-RLOC addresses encoded 1115 and so on up to 31 which encodes a total of 32 ITR-RLOC addresses. 1117 Record Count: The number of records in this Map-Request message. A 1118 record is comprised of the portion of the packet that is labeled 1119 'Rec' above and occurs the number of times equal to Record Count. 1120 For this version of the protocol, a receiver MUST accept and 1121 process Map-Requests that contain one or more records, but a 1122 sender MUST only send Map-Requests containing one record. Support 1123 for requesting multiple EIDs in a single Map-Request message will 1124 be specified in a future version of the protocol. 1126 Nonce: An 8-byte random value created by the sender of the Map- 1127 Request. This nonce will be returned in the Map-Reply. The 1128 security of the LISP mapping protocol depends critically on the 1129 strength of the nonce in the Map-Request message. The nonce 1130 SHOULD be generated by a properly seeded pseudo-random (or strong 1131 random) source. See [RFC4086] for advice on generating security- 1132 sensitive random data. 1134 Source-EID-AFI: Address family of the "Source EID Address" field. 1136 Source EID Address: This is the EID of the source host which 1137 originated the packet which is invoking this Map-Request. When 1138 Map-Requests are used for refreshing a map-cache entry or for 1139 RLOC-probing, an AFI value 0 is used and this field is of zero 1140 length. 1142 ITR-RLOC-AFI: Address family of the "ITR-RLOC Address" field that 1143 follows this field. 1145 ITR-RLOC Address: Used to give the ETR the option of selecting the 1146 destination address from any address family for the Map-Reply 1147 message. This address MUST be a routable RLOC address of the 1148 sender of the Map-Request message. 1150 EID mask-len: Mask length for EID prefix. 1152 EID-prefix-AFI: Address family of EID-prefix according to [RFC5226] 1154 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1155 address-family. When a Map-Request is sent by an ITR because a 1156 data packet is received for a destination where there is no 1157 mapping entry, the EID-prefix is set to the destination IP address 1158 of the data packet. And the 'EID mask-len' is set to 32 or 128 1159 for IPv4 or IPv6, respectively. When an xTR wants to query a site 1160 about the status of a mapping it already has cached, the EID- 1161 prefix used in the Map-Request has the same mask-length as the 1162 EID-prefix returned from the site when it sent a Map-Reply 1163 message. 1165 Map-Reply Record: When the M bit is set, this field is the size of a 1166 single "Record" in the Map-Reply format. This Map-Reply record 1167 contains the EID-to-RLOC mapping entry associated with the Source 1168 EID. This allows the ETR which will receive this Map-Request to 1169 cache the data if it chooses to do so. 1171 Mapping Protocol Data: See [CONS] for details. This field is 1172 optional and present when the UDP length indicates there is enough 1173 space in the packet to include it. 1175 6.1.3. EID-to-RLOC UDP Map-Request Message 1177 A Map-Request is sent from an ITR when it needs a mapping for an EID, 1178 wants to test an RLOC for reachability, or wants to refresh a mapping 1179 before TTL expiration. For the initial case, the destination IP 1180 address used for the Map-Request is the destination-EID from the 1181 packet which had a mapping cache lookup failure. For the latter 2 1182 cases, the destination IP address used for the Map-Request is one of 1183 the RLOC addresses from the locator-set of the map cache entry. The 1184 source address is either an IPv4 or IPv6 RLOC address depending if 1185 the Map-Request is using an IPv4 versus IPv6 header, respectively. 1186 In all cases, the UDP source port number for the Map-Request message 1187 is an ITR/PITR selected 16-bit value and the UDP destination port 1188 number is set to the well-known destination port number 4342. A 1189 successful Map-Reply updates the cached set of RLOCs associated with 1190 the EID prefix range. 1192 One or more Map-Request (ITR-RLOC-AFI, ITR-RLOC-Address) fields MUST 1193 be filled in by the ITR. The number of fields (minus 1) encoded MUST 1194 be placed in the IRC field. The ITR MAY include all locally 1195 configured locators in this list or just provide one locator address 1196 from each address family it supports. If the ITR erroneously 1197 provides no ITR-RLOC addresses, the Map-Replier MUST drop the Map- 1198 Request. 1200 Map-Requests can also be LISP encapsulated using UDP destination port 1201 4342 with a LISP type value set to "Encapsulated Control Message", 1202 when sent from an ITR to a Map-Resolver. Likewise, Map-Requests are 1203 LISP encapsulated the same way from a Map-Server to an ETR. Details 1204 on encapsulated Map-Requests and Map-Resolvers can be found in 1205 [LISP-MS]. 1207 Map-Requests MUST be rate-limited. It is recommended that a Map- 1208 Request for the same EID-prefix be sent no more than once per second. 1210 An ITR that is configured with mapping database information (i.e. it 1211 is also an ETR) may optionally include those mappings in a Map- 1212 Request. When an ETR configured to accept and verify such 1213 "piggybacked" mapping data receives such a Map-Request and it does 1214 not have this mapping in the map-cache, it may originate a "verifying 1215 Map-Request", addressed to the map-requesting ITR. If the ETR has a 1216 map-cache entry that matches the "piggybacked" EID and the RLOC is in 1217 the locator-set for the entry, then it may send the "verifying Map- 1218 Request" directly to the originating Map-Request source. If the RLOC 1219 is not in the locator-set, then the ETR MUST send the "verifying Map- 1220 Request" to the "piggybacked" EID. Doing this forces the "verifying 1221 Map-Request" to go through the mapping database system to reach the 1222 authoritative source of information about that EID, guarding against 1223 RLOC-spoofing in in the "piggybacked" mapping data. 1225 6.1.4. Map-Reply Message Format 1227 0 1 2 3 1228 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 1229 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1230 |Type=2 |P|E| Reserved | Record Count | 1231 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1232 | Nonce . . . | 1233 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1234 | . . . Nonce | 1235 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1236 | | Record TTL | 1237 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1238 R | Locator Count | EID mask-len | ACT |A| Reserved | 1239 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1240 c | Rsvd | Map-Version Number | EID-AFI | 1241 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1242 r | EID-prefix | 1243 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1244 | /| Priority | Weight | M Priority | M Weight | 1245 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1246 | o | Unused Flags |L|p|R| Loc-AFI | 1247 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1248 | \| Locator | 1249 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1250 | Mapping Protocol Data | 1251 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1253 Packet field descriptions: 1255 Type: 2 (Map-Reply) 1257 P: This is the probe-bit which indicates that the Map-Reply is in 1258 response to a locator reachability probe Map-Request. The nonce 1259 field MUST contain a copy of the nonce value from the original 1260 Map-Request. See Section 6.3.2 for more details. 1262 E: Indicates that the ETR which sends this Map-Reply message is 1263 advertising that the site is enabled for the Echo-Nonce locator 1264 reachability algorithm. See Section 6.3.1 for more details. 1266 Reserved: Set to 0 on transmission and ignored on receipt. 1268 Record Count: The number of records in this reply message. A record 1269 is comprised of that portion of the packet labeled 'Record' above 1270 and occurs the number of times equal to Record count. 1272 Nonce: A 24-bit value set in a Data-Probe packet or a 64-bit value 1273 from the Map-Request is echoed in this Nonce field of the Map- 1274 Reply. 1276 Record TTL: The time in minutes the recipient of the Map-Reply will 1277 store the mapping. If the TTL is 0, the entry SHOULD be removed 1278 from the cache immediately. If the value is 0xffffffff, the 1279 recipient can decide locally how long to store the mapping. 1281 Locator Count: The number of Locator entries. A locator entry 1282 comprises what is labeled above as 'Loc'. The locator count can 1283 be 0 indicating there are no locators for the EID-prefix. 1285 EID mask-len: Mask length for EID prefix. 1287 ACT: This 3-bit field describes negative Map-Reply actions. These 1288 bits are used only when the 'Locator Count' field is set to 0. 1289 The action bits are encoded only in Map-Reply messages. The 1290 actions defined are used by an ITR or PTR when a destination EID 1291 matches a negative mapping cache entry. Unassigned values should 1292 cause a map-cache entry to be created and, when packets match this 1293 negative cache entry, they will be dropped. The current assigned 1294 values are: 1296 (0) No-Action: The map-cache is kept alive and packet 1297 encapsulation occurs. 1299 (1) Natively-Forward: The packet is not encapsulated or dropped 1300 but natively forwarded. 1302 (2) Send-Map-Request: The packet invokes sending a Map-Request. 1304 (3) Drop: A packet that matches this map-cache entry is dropped. 1306 A: The Authoritative bit, when sent by a UDP-based message is always 1307 set to 1 by an ETR. See [CONS] for TCP-based Map-Replies. When a 1308 Map-Server is proxy Map-Replying [LISP-MS] for a LISP site, the 1309 Authoritative bit is set to 0. This indicates to requesting ITRs 1310 that the Map-Reply was not originated by a LISP node managed at 1311 the site that owns the EID-prefix. 1313 Map-Version Number: When this 12-bit value is non-zero the Map-Reply 1314 sender is informing the ITR what the version number is for the 1315 EID-record contained in the Map-Reply. The ETR can allocate this 1316 number internally but MUST coordinate this value with other ETRs 1317 for the site. When this value is 0, there is no versioning 1318 information conveyed. The Map-Version Number can be included in 1319 Map-Request and Map-Register messages. See Section 6.6.3 for more 1320 details. 1322 EID-AFI: Address family of EID-prefix according to [RFC5226]. 1324 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1325 address-family. 1327 Priority: each RLOC is assigned a unicast priority. Lower values 1328 are more preferable. When multiple RLOCs have the same priority, 1329 they may be used in a load-split fashion. A value of 255 means 1330 the RLOC MUST NOT be used for unicast forwarding. 1332 Weight: when priorities are the same for multiple RLOCs, the weight 1333 indicates how to balance unicast traffic between them. Weight is 1334 encoded as a relative weight of total unicast packets that match 1335 the mapping entry. If a non-zero weight value is used for any 1336 RLOC, then all RLOCs MUST use a non-zero weight value and then the 1337 sum of all weight values MUST equal 100. If a zero value is used 1338 for any RLOC weight, then all weights MUST be zero and the 1339 receiver of the Map-Reply will decide how to load-split traffic. 1340 See Section 6.5 for a suggested hash algorithm to distribute load 1341 across locators with same priority and equal weight values. 1343 M Priority: each RLOC is assigned a multicast priority used by an 1344 ETR in a receiver multicast site to select an ITR in a source 1345 multicast site for building multicast distribution trees. A value 1346 of 255 means the RLOC MUST NOT be used for joining a multicast 1347 distribution tree. 1349 M Weight: when priorities are the same for multiple RLOCs, the 1350 weight indicates how to balance building multicast distribution 1351 trees across multiple ITRs. The weight is encoded as a relative 1352 weight of total number of trees built to the source site 1353 identified by the EID-prefix. If a non-zero weight value is used 1354 for any RLOC, then all RLOCs MUST use a non-zero weight value and 1355 then the sum of all weight values MUST equal 100. If a zero value 1356 is used for any RLOC weight, then all weights MUST be zero and the 1357 receiver of the Map-Reply will decide how to distribute multicast 1358 state across ITRs. 1360 Unused Flags: set to 0 when sending and ignored on receipt. 1362 L: when this bit is set, the locator is flagged as a local locator to 1363 the ETR that is sending the Map-Reply. When a Map-Server is doing 1364 proxy Map-Replying [LISP-MS] for a LISP site, the L bit is set to 1365 0 for all locators in this locator-set. 1367 p: when this bit is set, an ETR informs the RLOC-probing ITR that the 1368 locator address, for which this bit is set, is the one being RLOC- 1369 probed and may be different from the source address of the Map- 1370 Reply. An ITR that RLOC-probes a particular locator, MUST use 1371 this locator for retrieving the data structure used to store the 1372 fact that the locator is reachable. The "p" bit is set for a 1373 single locator in the same locator set. If an implementation sets 1374 more than one "p" bit erroneously, the receiver of the Map-Reply 1375 MUST select the first locator. The "p" bit MUST NOT be set for 1376 locator-set records sent in Map-Request and Map-Register messages. 1378 R: set when the sender of a Map-Reply has a route to the locator in 1379 the locator data record. This receiver may find this useful to 1380 know when determining if the locator is reachable from the 1381 receiver. See also Section 6.4 for another way the R-bit may be 1382 used. 1384 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1385 assigned to an ETR. Note that the destination RLOC address MAY be 1386 an anycast address. A source RLOC can be an anycast address as 1387 well. The source or destination RLOC MUST NOT be the broadcast 1388 address (255.255.255.255 or any subnet broadcast address known to 1389 the router), and MUST NOT be a link-local multicast address. The 1390 source RLOC MUST NOT be a multicast address. The destination RLOC 1391 SHOULD be a multicast address if it is being mapped from a 1392 multicast destination EID. 1394 Mapping Protocol Data: See [CONS] or [ALT] for details. This field 1395 is optional and present when the UDP length indicates there is 1396 enough space in the packet to include it. 1398 6.1.5. EID-to-RLOC UDP Map-Reply Message 1400 A Map-Reply returns an EID-prefix with a prefix length that is less 1401 than or equal to the EID being requested. The EID being requested is 1402 either from the destination field of an IP header of a Data-Probe or 1403 the EID record of a Map-Request. The RLOCs in the Map-Reply are 1404 globally-routable IP addresses of all ETRs for the LISP site. Each 1405 RLOC conveys status reachability but does not convey path 1406 reachability from a requesters perspective. Separate testing of path 1407 reachability is required, See Section 6.3 for details. 1409 Note that a Map-Reply may contain different EID-prefix granularity 1410 (prefix + length) than the Map-Request which triggers it. This might 1411 occur if a Map-Request were for a prefix that had been returned by an 1412 earlier Map-Reply. In such a case, the requester updates its cache 1413 with the new prefix information and granularity. For example, a 1414 requester with two cached EID-prefixes that are covered by a Map- 1415 Reply containing one, less-specific prefix, replaces the entry with 1416 the less-specific EID-prefix. Note that the reverse, replacement of 1417 one less-specific prefix with multiple more-specific prefixes, can 1418 also occur but not by removing the less-specific prefix rather by 1419 adding the more-specific prefixes which during a lookup will override 1420 the less-specific prefix. 1422 When an ETR is configured with overlapping EID-prefixes, a Map- 1423 Request with an EID that longest matches any EID-prefix MUST be 1424 returned in a single Map-Reply message. For instance, if an ETR had 1425 database mapping entries for EID-prefixes: 1427 10.0.0.0/8 1428 10.1.0.0/16 1429 10.1.1.0/24 1430 10.1.2.0/24 1432 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record 1433 count of 1 to be returned with a mapping record EID-prefix of 1434 10.1.1.0/24. 1436 A Map-Request for EID 10.1.5.5, would cause a Map-Reply with a record 1437 count of 3 to be returned with mapping records for EID-prefixes 1438 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24. 1440 Note that not all overlapping EID-prefixes need to be returned, only 1441 the more specifics (note in the second example above 10.0.0.0/8 was 1442 not returned for requesting EID 10.1.5.5) entries for the matching 1443 EID-prefix of the requesting EID. When more than one EID-prefix is 1444 returned, all SHOULD use the same Time-to-Live value so they can all 1445 time out at the same time. When a more specific EID-prefix is 1446 received later, its Time-to-Live value in the Map-Reply record can be 1447 stored even when other less specifics exist. When a less specific 1448 EID-prefix is received later, its map-cache expiration time SHOULD be 1449 set to the minimum expiration time of any more specific EID-prefix in 1450 the map-cache. 1452 Map-Replies SHOULD be sent for an EID-prefix no more often than once 1453 per second to the same requesting router. For scalability, it is 1454 expected that aggregation of EID addresses into EID-prefixes will 1455 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1456 prefix range thereby reducing the number of Map-Request messages. 1458 Map-Reply records can have an empty locator-set. A negative Map- 1459 Reply is a Map-Reply with an empty locator-set. Negative Map-Replies 1460 convey special actions by the sender to the ITR or PTR which have 1461 solicited the Map-Reply. There are two primary applications for 1462 Negative Map-Replies. The first is for a Map-Resolver to instruct an 1463 ITR or PTR when a destination is for a LISP site versus a non-LISP 1464 site. And the other is to source quench Map-Requests which are sent 1465 for non-allocated EIDs. 1467 For each Map-Reply record, the list of locators in a locator-set MUST 1468 appear in the same order for each ETR that originates a Map-Reply 1469 message. The locator-set MUST be sorted in order of ascending IP 1470 address where an IPv4 locator address is considered numerically 'less 1471 than' an IPv6 locator address. 1473 When sending a Map-Reply message, the destination address is copied 1474 from the one of the ITR-RLOC fields from the Map-Request. The ETR 1475 can choose a locator address from one of the address families it 1476 supports. For Data-Probes, the destination address of the Map-Reply 1477 is copied from the source address of the Data-Probe message which is 1478 invoking the reply. The source address of the Map-Reply is one of 1479 the local locator addresses listed in the locator-set of any mapping 1480 record in the message and SHOULD be chosen to allow uRPF checks to 1481 succeed in the upstream service provider. The destination port of a 1482 Map-Reply message is copied from the source port of the Map-Request 1483 or Data-Probe and the source port of the Map-Reply message is set to 1484 the well-known UDP port 4342. 1486 6.1.5.1. Traffic Redirection with Coarse EID-Prefixes 1488 When an ETR is misconfigured or compromised, it could return coarse 1489 EID-prefixes in Map-Reply messages it sends. The EID-prefix could 1490 cover EID-prefixes which are allocated to other sites redirecting 1491 their traffic to the locators of the compromised site. 1493 To solve this problem, there are two basic solutions that could be 1494 used. The first is to have Map-Servers proxy-map-reply on behalf of 1495 ETRs so their registered EID-prefixes are the ones returned in Map- 1496 Replies. Since the interaction between an ETR and Map-Server is 1497 secured with shared-keys, it is more difficult for an ETR to 1498 misbehave. The second solution is to have ITRs and PTRs cache EID- 1499 prefixes with mask-lengths that are greater than or equal to a 1500 configured prefix length. This limits the damage to a specific width 1501 of any EID-prefix advertised, but needs to be coordinated with the 1502 allocation of site prefixes. These solutions can be used 1503 independently or at the same time. 1505 At the time of this writing, other approaches are being considered 1506 and researched. 1508 6.1.6. Map-Register Message Format 1510 The usage details of the Map-Register message can be found in 1511 specification [LISP-MS]. This section solely defines the message 1512 format. 1514 The message is sent in UDP with a destination UDP port of 4342 and a 1515 randomly selected UDP source port number. 1517 The Map-Register message format is: 1519 0 1 2 3 1520 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 1521 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1522 |Type=3 |P| Reserved | Record Count | 1523 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1524 | Nonce . . . | 1525 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1526 | . . . Nonce | 1527 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1528 | Key ID | Authentication Data Length | 1529 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1530 ~ Authentication Data ~ 1531 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1532 | | Record TTL | 1533 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1534 R | Locator Count | EID mask-len | ACT |A| Reserved | 1535 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1536 c | Rsvd | Map-Version Number | EID-AFI | 1537 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1538 r | EID-prefix | 1539 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1540 | /| Priority | Weight | M Priority | M Weight | 1541 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1542 | o | Unused Flags |L|p|R| Loc-AFI | 1543 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1544 | \| Locator | 1545 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1547 Packet field descriptions: 1549 Type: 3 (Map-Register) 1551 P: This is the proxy-map-reply bit, when set to 1 an ETR sends a Map- 1552 Register message requesting for the Map-Server to proxy Map-Reply. 1553 The Map-Server will send non-authoritative Map-Replies on behalf 1554 of the ETR. Details on this usage will be provided in a future 1555 version of this draft. 1557 Reserved: Set to 0 on transmission and ignored on receipt. 1559 Record Count: The number of records in this Map-Register message. A 1560 record is comprised of that portion of the packet labeled 'Record' 1561 above and occurs the number of times equal to Record count. 1563 Nonce: This 8-byte Nonce field is set to 0 in Map-Register messages. 1565 Key ID: A configured ID to find the configured Message 1566 Authentication Code (MAC) algorithm and key value used for the 1567 authentication function. 1569 Authentication Data Length: The length in bytes of the 1570 Authentication Data field that follows this field. The length of 1571 the Authentication Data field is dependent on the Message 1572 Authentication Code (MAC) algorithm used. The length field allows 1573 a device that doesn't know the MAC algorithm to correctly parse 1574 the packet. 1576 Authentication Data: The message digest used from the output of the 1577 Message Authentication Code (MAC) algorithm. The entire Map- 1578 Register payload is authenticated with this field preset to 0. 1579 After the MAC is computed, it is placed in this field. 1580 Implementations of this specification MUST include support for 1581 HMAC-SHA-1-96 [RFC2404] and support for HMAC-SHA-128-256 [RFC4634] 1582 is recommended. 1584 The definition of the rest of the Map-Register can be found in the 1585 Map-Reply section. 1587 6.1.7. Encapsulated Control Message Format 1589 An Encapsulated Control Message is used to encapsulate control 1590 packets sent between xTRs and the mapping database system described 1591 in [LISP-MS]. 1593 0 1 2 3 1594 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 1595 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1596 / | IPv4 or IPv6 Header | 1597 OH | (uses RLOC addresses) | 1598 \ | | 1599 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1600 / | Source Port = xxxx | Dest Port = 4342 | 1601 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1602 \ | UDP Length | UDP Checksum | 1603 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1604 LH |Type=8 | Reserved | 1605 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1606 / | IPv4 or IPv6 Header | 1607 IH | (uses RLOC or EID addresses) | 1608 \ | | 1609 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1610 / | Source Port = xxxx | Dest Port = yyyy | 1611 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1612 \ | UDP Length | UDP Checksum | 1613 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1614 LCM | LISP Control Message | 1615 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1617 Packet header descriptions: 1619 OH: The outer IPv4 or IPv6 header which uses RLOC addresses in the 1620 source and destination header address fields. 1622 UDP: The outer UDP header with destination port 4342. The source 1623 port is randomly allocated. The checksum field MUST be non-zero. 1625 LH: Type 8 is defined to be a "LISP Encapsulated Control Message" 1626 and what follows is either an IPv4 or IPv6 header as encoded by 1627 the first 4 bits after the reserved field. 1629 IH: The inner IPv4 or IPv6 header which can use either RLOC or EID 1630 addresses in the header address fields. When a Map-Request is 1631 encapsulated in this packet format the destination address in this 1632 header is an EID. 1634 UDP: The inner UDP header where the port assignments depends on the 1635 control packet being encapsulated. When the control packet is a 1636 Map-Request or Map-Register, the source port is ITR/PITR selected 1637 and the destination port is 4342. When the control packet is a 1638 Map-Reply, the source port is 4342 and the destination port is 1639 assigned from the source port of the invoking Map-Request. Port 1640 number 4341 MUST NOT be assigned to either port. The checksum 1641 field MUST be non-zero. 1643 LCM: The format is one of the control message formats described in 1644 this section. At this time, only Map-Request messages and PIM 1645 Join-Prune messages [MLISP] are allowed to be encapsulated. 1646 Encapsulating other types of LISP control messages are for further 1647 study. When Map-Requests are sent for RLOC-probing purposes (i.e 1648 the probe-bit is set), they MUST NOT be sent inside Encapsulated 1649 Control Messages. 1651 6.2. Routing Locator Selection 1653 Both client-side and server-side may need control over the selection 1654 of RLOCs for conversations between them. This control is achieved by 1655 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1656 messages. Alternatively, RLOC information may be gleaned from 1657 received tunneled packets or EID-to-RLOC Map-Request messages. 1659 The following enumerates different scenarios for choosing RLOCs and 1660 the controls that are available: 1662 o Server-side returns one RLOC. Client-side can only use one RLOC. 1663 Server-side has complete control of the selection. 1665 o Server-side returns a list of RLOC where a subset of the list has 1666 the same best priority. Client can only use the subset list 1667 according to the weighting assigned by the server-side. In this 1668 case, the server-side controls both the subset list and load- 1669 splitting across its members. The client-side can use RLOCs 1670 outside of the subset list if it determines that the subset list 1671 is unreachable (unless RLOCs are set to a Priority of 255). Some 1672 sharing of control exists: the server-side determines the 1673 destination RLOC list and load distribution while the client-side 1674 has the option of using alternatives to this list if RLOCs in the 1675 list are unreachable. 1677 o Server-side sets weight of 0 for the RLOC subset list. In this 1678 case, the client-side can choose how the traffic load is spread 1679 across the subset list. Control is shared by the server-side 1680 determining the list and the client determining load distribution. 1681 Again, the client can use alternative RLOCs if the server-provided 1682 list of RLOCs are unreachable. 1684 o Either side (more likely on the server-side ETR) decides not to 1685 send a Map-Request. For example, if the server-side ETR does not 1686 send Map-Requests, it gleans RLOCs from the client-side ITR, 1687 giving the client-side ITR responsibility for bidirectional RLOC 1688 reachability and preferability. Server-side ETR gleaning of the 1689 client-side ITR RLOC is done by caching the inner header source 1690 EID and the outer header source RLOC of received packets. The 1691 client-side ITR controls how traffic is returned and can alternate 1692 using an outer header source RLOC, which then can be added to the 1693 list the server-side ETR uses to return traffic. Since no 1694 Priority or Weights are provided using this method, the server- 1695 side ETR MUST assume each client-side ITR RLOC uses the same best 1696 Priority with a Weight of zero. In addition, since EID-prefix 1697 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1698 on tunnel routers can grow to be very large. 1700 o A "gleaned" map-cache entry, one learned from the source RLOC of a 1701 received encapsulated packet, is only stored and used for a few 1702 seconds, pending verification. Verification is performed by 1703 sending a Map-Request to the source EID (the inner header IP 1704 source address) of the received encapsulated packet. A reply to 1705 this "verifying Map-Request" is used to fully populate the map- 1706 cache entry for the "gleaned" EID and is stored and used for the 1707 time indicated from the TTL field of a received Map-Reply. When a 1708 verified map-cache entry is stored, data gleaning no longer occurs 1709 for subsequent packets which have a source EID that matches the 1710 EID-prefix of the verified entry. 1712 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1713 reachable when the R-bit for the locator record is set to 1. Neither 1714 the information contained in a Map-Reply or that stored in the 1715 mapping database system provides reachability information for RLOCs. 1716 Note that reachability is not part of the mapping system and is 1717 determined using one or more of the Routing Locator Reachability 1718 Algorithms described in the next section. 1720 6.3. Routing Locator Reachability 1722 Several mechanisms for determining RLOC reachability are currently 1723 defined: 1725 1. An ETR may examine the Loc-Status-Bits in the LISP header of an 1726 encapsulated data packet received from an ITR. If the ETR is 1727 also acting as an ITR and has traffic to return to the original 1728 ITR site, it can use this status information to help select an 1729 RLOC. 1731 2. An ITR may receive an ICMP Network or ICMP Host Unreachable 1732 message for an RLOC it is using. This indicates that the RLOC is 1733 likely down. 1735 3. An ITR which participates in the global routing system can 1736 determine that an RLOC is down if no BGP RIB route exists that 1737 matches the RLOC IP address. 1739 4. An ITR may receive an ICMP Port Unreachable message from a 1740 destination host. This occurs if an ITR attempts to use 1741 interworking [INTERWORK] and LISP-encapsulated data is sent to a 1742 non-LISP-capable site. 1744 5. An ITR may receive a Map-Reply from a ETR in response to a 1745 previously sent Map-Request. The RLOC source of the Map-Reply is 1746 likely up since the ETR was able to send the Map-Reply to the 1747 ITR. 1749 6. When an ETR receives an encapsulated packet from an ITR, the 1750 source RLOC from the outer header of the packet is likely up. 1752 7. An ITR/ETR pair can use the Locator Reachability Algorithms 1753 described in this section, namely Echo-Noncing or RLOC-Probing. 1755 When determining Locator up/down reachability by examining the Loc- 1756 Status-Bits from the LISP encapsulated data packet, an ETR will 1757 receive up to date status from an encapsulating ITR about 1758 reachability for all ETRs at the site. CE-based ITRs at the source 1759 site can determine reachability relative to each other using the site 1760 IGP as follows: 1762 o Under normal circumstances, each ITR will advertise a default 1763 route into the site IGP. 1765 o If an ITR fails or if the upstream link to its PE fails, its 1766 default route will either time-out or be withdrawn. 1768 Each ITR can thus observe the presence or lack of a default route 1769 originated by the others to determine the Locator Status Bits it sets 1770 for them. 1772 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1773 Loc-Status-Bits in a LISP encapsulated packet are numbered from 0 to 1774 n-1 starting with the least significant bit. For example, if an RLOC 1775 listed in the 3rd position of the Map-Reply goes down (ordinal value 1776 2), then all ITRs at the site will clear the 3rd least significant 1777 bit (xxxx x0xx) of the Loc-Status-Bits field for the packets they 1778 encapsulate. 1780 When an ETR decapsulates a packet, it will check for any change in 1781 the Loc-Status-Bits field. When a bit goes from 1 to 0, the ETR will 1782 refrain from encapsulating packets to an RLOC that is indicated as 1783 down. It will only resume using that RLOC if the corresponding Loc- 1784 Status-Bit returns to a value of 1. Loc-Status-Bits are associated 1785 with a locator-set per EID-prefix. Therefore, when a locator becomes 1786 unreachable, the Loc-Status-Bit that corresponds to that locator's 1787 position in the list returned by the last Map-Reply will be set to 1788 zero for that particular EID-prefix. 1790 When ITRs at the site are not deployed in CE routers, the IGP can 1791 still be used to determine the reachability of Locators provided they 1792 are injected into the IGP. This is typically done when a /32 address 1793 is configured on a loopback interface. 1795 When ITRs receive ICMP Network or Host Unreachable messages as a 1796 method to determine unreachability, they will refrain from using 1797 Locators which are described in Locator lists of Map-Replies. 1798 However, using this approach is unreliable because many network 1799 operators turn off generation of ICMP Unreachable messages. 1801 If an ITR does receive an ICMP Network or Host Unreachable message, 1802 it MAY originate its own ICMP Unreachable message destined for the 1803 host that originated the data packet the ITR encapsulated. 1805 Also, BGP-enabled ITRs can unilaterally examine the BGP RIB to see if 1806 a locator address from a locator-set in a mapping entry matches a 1807 prefix. If it does not find one and BGP is running in the Default 1808 Free Zone (DFZ), it can decide to not use the locator even though the 1809 Loc-Status-Bits indicate the locator is up. In this case, the path 1810 from the ITR to the ETR that is assigned the locator is not 1811 available. More details are in [LOC-ID-ARCH]. 1813 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 1814 Reply is returned, reachability of the Locator has been determined. 1815 Obviously, sending such probes increases the number of control 1816 messages originated by tunnel routers for active flows, so Locators 1817 are assumed to be reachable when they are advertised. 1819 This assumption does create a dependency: Locator unreachability is 1820 detected by the receipt of ICMP Host Unreachable messages. When an 1821 Locator has been determined to be unreachable, it is not used for 1822 active traffic; this is the same as if it were listed in a Map-Reply 1823 with priority 255. 1825 The ITR can test the reachability of the unreachable Locator by 1826 sending periodic Requests. Both Requests and Replies MUST be rate- 1827 limited. Locator reachability testing is never done with data 1828 packets since that increases the risk of packet loss for end-to-end 1829 sessions. 1831 When an ETR decapsulates a packet, it knows that it is reachable from 1832 the encapsulating ITR because that is how the packet arrived. In 1833 most cases, the ETR can also reach the ITR but cannot assume this to 1834 be true due to the possibility of path asymmetry. In the presence of 1835 unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD NOT 1836 use the lack of return traffic as an indication that the ETR is 1837 unreachable. Instead, it MUST use an alternate mechanisms to 1838 determine reachability. 1840 6.3.1. Echo Nonce Algorithm 1842 When data flows bidirectionally between locators from different 1843 sites, a simple mechanism called "nonce echoing" can be used to 1844 determine reachability between an ITR and ETR. When an ITR wants to 1845 solicit a nonce echo, it sets the N and E bits and places a 24-bit 1846 nonce in the LISP header of the next encapsulated data packet. 1848 When this packet is received by the ETR, the encapsulated packet is 1849 forwarded as normal. When the ETR next sends a data packet to the 1850 ITR, it includes the nonce received earlier with the N bit set and E 1851 bit cleared. The ITR sees this "echoed nonce" and knows the path to 1852 and from the ETR is up. 1854 The ITR will set the E-bit and N-bit for every packet it sends while 1855 in echo-nonce-request state. The time the ITR waits to process the 1856 echoed nonce before it determines the path is unreachable is variable 1857 and a choice left for the implementation. 1859 If the ITR is receiving packets from the ETR but does not see the 1860 nonce echoed while being in echo-nonce-request state, then the path 1861 to the ETR is unreachable. This decision may be overridden by other 1862 locator reachability algorithms. Once the ITR determines the path to 1863 the ETR is down it can switch to another locator for that EID-prefix. 1865 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 1866 be implementing both ITR and ETR functionality for the echo nonce 1867 mechanism to operate. 1869 The ITR and ETR may both go into echo-nonce-request state at the same 1870 time. The number of packets sent or the time during which echo nonce 1871 requests are sent is an implementation specific setting. However, 1872 when an ITR is in echo-nonce-request state, it can echo the ETR's 1873 nonce in the next set of packets that it encapsulates and then 1874 subsequently, continue sending echo-nonce-request packets. 1876 This mechanism does not completely solve the forward path 1877 reachability problem as traffic may be unidirectional. That is, the 1878 ETR receiving traffic at a site may not be the same device as an ITR 1879 which transmits traffic from that site or the site to site traffic is 1880 unidirectional so there is no ITR returning traffic. 1882 The echo-nonce algorithm is bilateral. That is, if one side sets the 1883 E-bit and the other side is not enabled for echo-noncing, then the 1884 echoing of the nonce does not occur and the requesting side may 1885 regard the locator unreachable erroneously. An ITR SHOULD only set 1886 the E-bit in a encapsulated data packet when it knows the ETR is 1887 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 1888 Reply message. 1890 Note that other locator reachability mechanisms are being researched 1891 and can be used to compliment or even override the Echo Nonce 1892 Algorithm. See next section for an example of control-plane probing. 1894 6.3.2. RLOC Probing Algorithm 1896 RLOC Probing is a method that an ITR or PTR can use to determine the 1897 reachability status of one or more locators that it has cached in a 1898 map-cache entry. The probe-bit of the Map-Request and Map-Reply 1899 messages are used for RLOC Probing. 1901 RLOC probing is done in the control-plane on a timer basis where an 1902 ITR or PTR will originate a Map-Request destined to a locator address 1903 from one of its own locator addresses. A Map-Request used as an 1904 RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or on the 1905 ALT like one would when soliciting mapping data. The EID record 1906 encoded in the Map-Request is the EID-prefix of the map-cache entry 1907 cached by the ITR or PTR. The ITR may include a mapping data record 1908 for its own database mapping information which contains the local 1909 EID-prefixes and RLOCs for its site. 1911 When an ETR receives a Map-Request message with the probe-bit set, it 1912 returns a Map-Reply with the probe-bit set. The source address of 1913 the Map-Reply is set from the destination address of the Map-Request 1914 and the destination address of the Map-Reply is set from the source 1915 address of the Map-Request. The Map-Reply SHOULD contain mapping 1916 data for the EID-prefix contained in the Map-Request. This provides 1917 the opportunity for the ITR or PTR, which sent the RLOC-probe to get 1918 mapping updates if there were changes to the ETR's database mapping 1919 entries. 1921 There are advantages and disadvantages of RLOC Probing. The greatest 1922 benefit of RLOC Probing is that it can handle many failure scenarios 1923 allowing the ITR to determine when the path to a specific locator is 1924 reachable or has become unreachable, thus providing a robust 1925 mechanism for switching to using another locator from the cached 1926 locator. RLOC Probing can also provide rough RTT estimates between a 1927 pair of locators which can be useful for network management purposes 1928 as well as for selecting low delay paths. The major disadvantage of 1929 RLOC Probing is in the number of control messages required and the 1930 amount of bandwidth used to obtain those benefits, especially if the 1931 requirement for failure detection times are very small. 1933 Continued research and testing will attempt to characterize the 1934 tradeoffs of failure detection times versus message overhead. 1936 6.4. EID Reachability within a LISP Site 1938 A site may be multihomed using two or more ETRs. The hosts and 1939 infrastructure within a site will be addressed using one or more EID 1940 prefixes that are mapped to the RLOCs of the relevant ETRs in the 1941 mapping system. One possible failure mode is for an ETR to lose 1942 reachability to one or more of the EID prefixes within its own site. 1943 When this occurs when the ETR sends Map-Replies, it can clear the 1944 R-bit associated with its own locator. And when the ETR is also an 1945 ITR, it can clear its locator-status-bit in the encapsulation data 1946 header. 1948 6.5. Routing Locator Hashing 1950 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 1951 a requesting ITR, the locator-set for the EID-prefix may contain 1952 different priority values for each locator address. When more than 1953 one best priority locator exists, the ITR can decide how to load 1954 share traffic against the corresponding locators. 1956 The following hash algorithm may be used by an ITR to select a 1957 locator for a packet destined to an EID for the EID-to-RLOC mapping: 1959 1. Either a source and destination address hash can be used or the 1960 traditional 5-tuple hash which includes the source and 1961 destination addresses, source and destination TCP, UDP, or SCTP 1962 port numbers and the IP protocol number field or IPv6 next- 1963 protocol fields of a packet a host originates from within a LISP 1964 site. When a packet is not a TCP, UDP, or SCTP packet, the 1965 source and destination addresses only from the header are used to 1966 compute the hash. 1968 2. Take the hash value and divide it by the number of locators 1969 stored in the locator-set for the EID-to-RLOC mapping. 1971 3. The remainder will be yield a value of 0 to "number of locators 1972 minus 1". Use the remainder to select the locator in the 1973 locator-set. 1975 Note that when a packet is LISP encapsulated, the source port number 1976 in the outer UDP header needs to be set. Selecting a hashed value 1977 allows core routers which are attached to Link Aggregation Groups 1978 (LAGs) to load-split the encapsulated packets across member links of 1979 such LAGs. Otherwise, core routers would see a single flow, since 1980 packets have a source address of the ITR, for packets which are 1981 originated by different EIDs at the source site. A suggested setting 1982 for the source port number computed by an ITR is a 5-tuple hash 1983 function on the inner header, as described above. 1985 Many core router implementations use a 5-tuple hash to decide how to 1986 balance packet load across members of a LAG. The 5-tuple hash 1987 includes the source and destination addresses of the packet and the 1988 source and destination ports when the protocol number in the packet 1989 is TCP or UDP. For this reason, UDP encoding is used for LISP 1990 encapsulation. 1992 6.6. Changing the Contents of EID-to-RLOC Mappings 1994 Since the LISP architecture uses a caching scheme to retrieve and 1995 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 1996 date mapping is to re-request the mapping. However, the ITRs do not 1997 know when the mappings change and the ETRs do not keep track of which 1998 ITRs requested its mappings. For scalability reasons, we want to 1999 maintain this approach but need to provide a way for ETRs change 2000 their mappings and inform the sites that are currently communicating 2001 with the ETR site using such mappings. 2003 When a locator record is added to the end of a locator-set, it is 2004 easy to update mappings. We assume new mappings will maintain the 2005 same locator ordering as the old mapping but just have new locators 2006 appended to the end of the list. So some ITRs can have a new mapping 2007 while other ITRs have only an old mapping that is used until they 2008 time out. When an ITR has only an old mapping but detects bits set 2009 in the loc-status-bits that correspond to locators beyond the list it 2010 has cached, it simply ignores them. However, this can only happen 2011 for locator addresses that are lexicographically greater than the 2012 locator addresses in the existing locator-set. 2014 When a locator record is removed from a locator-set, ITRs that have 2015 the mapping cached will not use the removed locator because the xTRs 2016 will set the loc-status-bit to 0. So even if the locator is in the 2017 list, it will not be used. For new mapping requests, the xTRs can 2018 set the locator AFI to 0 (indicating an unspecified address), as well 2019 as setting the corresponding loc-status-bit to 0. This forces ITRs 2020 with old or new mappings to avoid using the removed locator. 2022 If many changes occur to a mapping over a long period of time, one 2023 will find empty record slots in the middle of the locator-set and new 2024 records appended to the locator-set. At some point, it would be 2025 useful to compact the locator-set so the loc-status-bit settings can 2026 be efficiently packed. 2028 We propose here three approaches for locator-set compaction, one 2029 operational and two protocol mechanisms. The operational approach 2030 uses a clock sweep method. The protocol approaches use the concept 2031 of Solicit-Map-Requests and Map-Versioning. 2033 6.6.1. Clock Sweep 2035 The clock sweep approach uses planning in advance and the use of 2036 count-down TTLs to time out mappings that have already been cached. 2037 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 2038 there is a 24 hour window to time out old mappings. The following 2039 clock sweep procedure is used: 2041 1. 24 hours before a mapping change is to take effect, a network 2042 administrator configures the ETRs at a site to start the clock 2043 sweep window. 2045 2. During the clock sweep window, ETRs continue to send Map-Reply 2046 messages with the current (unchanged) mapping records. The TTL 2047 for these mappings is set to 1 hour. 2049 3. 24 hours later, all previous cache entries will have timed out, 2050 and any active cache entries will time out within 1 hour. During 2051 this 1 hour window the ETRs continue to send Map-Reply messages 2052 with the current (unchanged) mapping records with the TTL set to 2053 1 minute. 2055 4. At the end of the 1 hour window, the ETRs will send Map-Reply 2056 messages with the new (changed) mapping records. So any active 2057 caches can get the new mapping contents right away if not cached, 2058 or in 1 minute if they had the mapping cached. The new mappings 2059 are cached with a time to live equal to the TTL in the Map-Reply. 2061 6.6.2. Solicit-Map-Request (SMR) 2063 Soliciting a Map-Request is a selective way for ETRs, at the site 2064 where mappings change, to control the rate they receive requests for 2065 Map-Reply messages. SMRs are also used to tell remote ITRs to update 2066 the mappings they have cached. 2068 Since the ETRs don't keep track of remote ITRs that have cached their 2069 mappings, they do not know which ITRs need to have their mappings 2070 updated. As a result, an ETR will solicit Map-Requests (called an 2071 SMR message) from those sites to which it has been sending 2072 encapsulated data to for the last minute. In particular, an ETR will 2073 send an SMR an ITR to which it has recently sent encapsulated data. 2075 An SMR message is simply a bit set in a Map-Request message. An ITR 2076 or PTR will send a Map-Request when they receive an SMR message. 2077 Both the SMR sender and the Map-Request responder MUST rate-limited 2078 these messages. Rate-limiting can be implemented as a global rate- 2079 limiter or one rate-limiter per SMR destination. 2081 The following procedure shows how a SMR exchange occurs when a site 2082 is doing locator-set compaction for an EID-to-RLOC mapping: 2084 1. When the database mappings in an ETR change, the ETRs at the site 2085 begin to send Map-Requests with the SMR bit set for each locator 2086 in each map-cache entry the ETR caches. 2088 2. A remote ITR which receives the SMR message will schedule sending 2089 a Map-Request message to the source locator address of the SMR 2090 message. A newly allocated random nonce is selected and the EID- 2091 prefix used is the one copied from the SMR message. If the 2092 source locator is the only locator in the cached locator-set, the 2093 remote ITR SHOULD send a Map-Request to the database mapping 2094 system just in case the single locator has changed and may no 2095 longer be reachable to accept the Map-Request. 2097 3. The remote ITR MUST rate-limit the Map-Request until it gets a 2098 Map-Reply while continuing to use the cached mapping. When Map 2099 Versioning is used, described in Section 6.6.3, an SMR sender can 2100 detect if an ITR is using the most up to date database mapping. 2102 4. The ETRs at the site with the changed mapping will reply to the 2103 Map-Request with a Map-Reply message that has a nonce from the 2104 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate 2105 limited. This is important to avoid Map-Reply implosion. 2107 5. The ETRs, at the site with the changed mapping, record the fact 2108 that the site that sent the Map-Request has received the new 2109 mapping data in the mapping cache entry for the remote site so 2110 the loc-status-bits are reflective of the new mapping for packets 2111 going to the remote site. The ETR then stops sending SMR 2112 messages. 2114 For security reasons an ITR MUST NOT process unsolicited Map-Replies. 2115 To avoid map-cache entry corruption by a third-party, a sender of an 2116 SMR-based Map-Request MUST be verified. If an ITR receives an SMR- 2117 based Map-Request and the source is not in the locator-set for the 2118 stored map-cache entry, then the responding Map-Request MUST be sent 2119 with an EID destination to the mapping database system. Since the 2120 mapping database system is more secure to reach an authoritative ETR, 2121 it will deliver the Map-Request to the authoritative source of the 2122 mapping data. 2124 6.6.3. Database Map Versioning 2126 When there is unidirectional packet flow between an ITR and ETR, and 2127 the EID-to-RLOC mappings change on the ETR, it needs to inform the 2128 ITR so encapsulation can stop to a removed locator and start to a new 2129 locator in the locator-set. 2131 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 2132 number. This is known as the Destination Map-Version Number. ITRs 2133 include the Destination Map-Version Number in packets they 2134 encapsulate to the site. When an ETR decapsulates a packet and 2135 detects the Destination Map-Version Number is less than the current 2136 version for its mapping, the SMR procedure described in Section 6.6.2 2137 occurs. 2139 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 2140 Version number. This is known as the Source Map-Version Number. 2141 When an ETR decapsulates a packet and detects the Source Map-Version 2142 Number is greater than the last Map-Version Number sent in a Map- 2143 Reply from the ITR's site, the ETR will send a Map-Request to one of 2144 the ETRs for the source site. 2146 A Map-Version Number is used as a sequence number per EID-prefix. So 2147 values that are greater, are considered to be more recent. A value 2148 of 0 for the Source Map-Version Number or the Destination Map-Version 2149 Number conveys no versioning information and an ITR does no 2150 comparison with previously received Map-Version Numbers. 2152 A Map-Version Number can be included in Map-Register messages as 2153 well. This is a good way for the Map-Server can assure that all ETRs 2154 for a site registering to it will be Map-Version number synchronized. 2156 See [VERSIONING] for a more detailed analysis and description of 2157 Database Map Versioning. 2159 7. Router Performance Considerations 2161 LISP is designed to be very hardware-based forwarding friendly. A 2162 few implementation techniques can be used to incrementally implement 2163 LISP: 2165 o When a tunnel encapsulated packet is received by an ETR, the outer 2166 destination address may not be the address of the router. This 2167 makes it challenging for the control plane to get packets from the 2168 hardware. This may be mitigated by creating special FIB entries 2169 for the EID-prefixes of EIDs served by the ETR (those for which 2170 the router provides an RLOC translation). These FIB entries are 2171 marked with a flag indicating that control plane processing should 2172 be performed. The forwarding logic of testing for particular IP 2173 protocol number value is not necessary. No changes to existing, 2174 deployed hardware should be needed to support this. 2176 o On an ITR, prepending a new IP header consists of adding more 2177 bytes to a MAC rewrite string and prepending the string as part of 2178 the outgoing encapsulation procedure. Routers that support GRE 2179 tunneling [RFC2784] or 6to4 tunneling [RFC3056] may already 2180 support this action. 2182 o A packet's source address or interface the packet was received on 2183 can be used to select a VRF (Virtual Routing/Forwarding). The 2184 VRF's routing table can be used to find EID-to-RLOC mappings. 2186 8. Deployment Scenarios 2188 This section will explore how and where ITRs and ETRs can be deployed 2189 and will discuss the pros and cons of each deployment scenario. 2190 There are two basic deployment trade-offs to consider: centralized 2191 versus distributed caches and flat, recursive, or re-encapsulating 2192 tunneling. 2194 When deciding on centralized versus distributed caching, the 2195 following issues should be considered: 2197 o Are the tunnel routers spread out so that the caches are spread 2198 across all the memories of each router? 2200 o Should management "touch points" be minimized by choosing few 2201 tunnel routers, just enough for redundancy? 2203 o In general, using more ITRs doesn't increase management load, 2204 since caches are built and stored dynamically. On the other hand, 2205 more ETRs does require more management since EID-prefix-to-RLOC 2206 mappings need to be explicitly configured. 2208 When deciding on flat, recursive, or re-encapsulation tunneling, the 2209 following issues should be considered: 2211 o Flat tunneling implements a single tunnel between source site and 2212 destination site. This generally offers better paths between 2213 sources and destinations with a single tunnel path. 2215 o Recursive tunneling is when tunneled traffic is again further 2216 encapsulated in another tunnel, either to implement VPNs or to 2217 perform Traffic Engineering. When doing VPN-based tunneling, the 2218 site has some control since the site is prepending a new tunnel 2219 header. In the case of TE-based tunneling, the site may have 2220 control if it is prepending a new tunnel header, but if the site's 2221 ISP is doing the TE, then the site has no control. Recursive 2222 tunneling generally will result in suboptimal paths but at the 2223 benefit of steering traffic to resource available parts of the 2224 network. 2226 o The technique of re-encapsulation ensures that packets only 2227 require one tunnel header. So if a packet needs to be rerouted, 2228 it is first decapsulated by the ETR and then re-encapsulated with 2229 a new tunnel header using a new RLOC. 2231 The next sub-sections will survey where tunnel routers can reside in 2232 the network. 2234 8.1. First-hop/Last-hop Tunnel Routers 2236 By locating tunnel routers close to hosts, the EID-prefix set is at 2237 the granularity of an IP subnet. So at the expense of more EID- 2238 prefix-to-RLOC sets for the site, the caches in each tunnel router 2239 can remain relatively small. But caches always depend on the number 2240 of non-aggregated EID destination flows active through these tunnel 2241 routers. 2243 With more tunnel routers doing encapsulation, the increase in control 2244 traffic grows as well: since the EID-granularity is greater, more 2245 Map-Requests and Map-Replies are traveling between more routers. 2247 The advantage of placing the caches and databases at these stub 2248 routers is that the products deployed in this part of the network 2249 have better price-memory ratios then their core router counterparts. 2250 Memory is typically less expensive in these devices and fewer routes 2251 are stored (only IGP routes). These devices tend to have excess 2252 capacity, both for forwarding and routing state. 2254 LISP functionality can also be deployed in edge switches. These 2255 devices generally have layer-2 ports facing hosts and layer-3 ports 2256 facing the Internet. Spare capacity is also often available in these 2257 devices as well. 2259 8.2. Border/Edge Tunnel Routers 2261 Using customer-edge (CE) routers for tunnel endpoints allows the EID 2262 space associated with a site to be reachable via a small set of RLOCs 2263 assigned to the CE routers for that site. This is the default 2264 behavior envisioned in the rest of this specification. 2266 This offers the opposite benefit of the first-hop/last-hop tunnel 2267 router scenario: the number of mapping entries and network management 2268 touch points are reduced, allowing better scaling. 2270 One disadvantage is that less of the network's resources are used to 2271 reach host endpoints thereby centralizing the point-of-failure domain 2272 and creating network choke points at the CE router. 2274 Note that more than one CE router at a site can be configured with 2275 the same IP address. In this case an RLOC is an anycast address. 2276 This allows resilience between the CE routers. That is, if a CE 2277 router fails, traffic is automatically routed to the other routers 2278 using the same anycast address. However, this comes with the 2279 disadvantage where the site cannot control the entrance point when 2280 the anycast route is advertised out from all border routers. Another 2281 disadvantage of using anycast locators is the limited advertisement 2282 scope of /32 (or /128 for IPv6) routes. 2284 8.3. ISP Provider-Edge (PE) Tunnel Routers 2286 Use of ISP PE routers as tunnel endpoint routers is not the typical 2287 deployment scenario envisioned in the specification. This section 2288 attempts to capture some of reasoning behind this preference of 2289 implementing LISP on CE routers. 2291 Use of ISP PE routers as tunnel endpoint routers gives an ISP, rather 2292 than a site, control over the location of the egress tunnel 2293 endpoints. That is, the ISP can decide if the tunnel endpoints are 2294 in the destination site (in either CE routers or last-hop routers 2295 within a site) or at other PE edges. The advantage of this case is 2296 that two tunnel headers can be avoided. By having the PE be the 2297 first router on the path to encapsulate, it can choose a TE path 2298 first, and the ETR can decapsulate and re-encapsulate for a tunnel to 2299 the destination end site. 2301 An obvious disadvantage is that the end site has no control over 2302 where its packets flow or the RLOCs used. Other disadvantages 2303 include the difficulty in synchronizing path liveness updates between 2304 CE and PE routers. 2306 As mentioned in earlier sections a combination of these scenarios is 2307 possible at the expense of extra packet header overhead, if both site 2308 and provider want control, then recursive or re-encapsulating tunnels 2309 are used. 2311 8.4. LISP Functionality with Conventional NATs 2313 LISP routers can be deployed behind Network Address Translator (NAT) 2314 devices to provide the same set of packet services hosts have today 2315 when they are addressed out of private address space. 2317 It is important to note that a locator address in any LISP control 2318 message MUST be a globally routable address and therefore SHOULD NOT 2319 contain [RFC1918] addresses. If a LISP router is configured with 2320 private addresses, they MUST be used only in the outer IP header so 2321 the NAT device can translate properly. Otherwise, EID addresses MUST 2322 be translated before encapsulation is performed. Both NAT 2323 translation and LISP encapsulation functions could be co-located in 2324 the same device. 2326 More details on LISP address translation can be found in [INTERWORK]. 2328 9. Traceroute Considerations 2330 When a source host in a LISP site initiates a traceroute to a 2331 destination host in another LISP site, it is highly desirable for it 2332 to see the entire path. Since packets are encapsulated from ITR to 2333 ETR, the hop across the tunnel could be viewed as a single hop. 2334 However, LISP traceroute will provide the entire path so the user can 2335 see 3 distinct segments of the path from a source LISP host to a 2336 destination LISP host: 2338 Segment 1 (in source LISP site based on EIDs): 2340 source-host ---> first-hop ... next-hop ---> ITR 2342 Segment 2 (in the core network based on RLOCs): 2344 ITR ---> next-hop ... next-hop ---> ETR 2346 Segment 3 (in the destination LISP site based on EIDs): 2348 ETR ---> next-hop ... last-hop ---> destination-host 2350 For segment 1 of the path, ICMP Time Exceeded messages are returned 2351 in the normal matter as they are today. The ITR performs a TTL 2352 decrement and test for 0 before encapsulating. So the ITR hop is 2353 seen by the traceroute source has an EID address (the address of 2354 site-facing interface). 2356 For segment 2 of the path, ICMP Time Exceeded messages are returned 2357 to the ITR because the TTL decrement to 0 is done on the outer 2358 header, so the destination of the ICMP messages are to the ITR RLOC 2359 address, the source RLOC address of the encapsulated traceroute 2360 packet. The ITR looks inside of the ICMP payload to inspect the 2361 traceroute source so it can return the ICMP message to the address of 2362 the traceroute client as well as retaining the core router IP address 2363 in the ICMP message. This is so the traceroute client can display 2364 the core router address (the RLOC address) in the traceroute output. 2365 The ETR returns its RLOC address and responds to the TTL decrement to 2366 0 like the previous core routers did. 2368 For segment 3, the next-hop router downstream from the ETR will be 2369 decrementing the TTL for the packet that was encapsulated, sent into 2370 the core, decapsulated by the ETR, and forwarded because it isn't the 2371 final destination. If the TTL is decremented to 0, any router on the 2372 path to the destination of the traceroute, including the next-hop 2373 router or destination, will send an ICMP Time Exceeded message to the 2374 source EID of the traceroute client. The ICMP message will be 2375 encapsulated by the local ITR and sent back to the ETR in the 2376 originated traceroute source site, where the packet will be delivered 2377 to the host. 2379 9.1. IPv6 Traceroute 2381 IPv6 traceroute follows the procedure described above since the 2382 entire traceroute data packet is included in ICMP Time Exceeded 2383 message payload. Therefore, only the ITR needs to pay special 2384 attention for forwarding ICMP messages back to the traceroute source. 2386 9.2. IPv4 Traceroute 2388 For IPv4 traceroute, we cannot follow the above procedure since IPv4 2389 ICMP Time Exceeded messages only include the invoking IP header and 8 2390 bytes that follow the IP header. Therefore, when a core router sends 2391 an IPv4 Time Exceeded message to an ITR, all the ITR has in the ICMP 2392 payload is the encapsulated header it prepended followed by a UDP 2393 header. The original invoking IP header, and therefore the identity 2394 of the traceroute source is lost. 2396 The solution we propose to solve this problem is to cache traceroute 2397 IPv4 headers in the ITR and to match them up with corresponding IPv4 2398 Time Exceeded messages received from core routers and the ETR. The 2399 ITR will use a circular buffer for caching the IPv4 and UDP headers 2400 of traceroute packets. It will select a 16-bit number as a key to 2401 find them later when the IPv4 Time Exceeded messages are received. 2402 When an ITR encapsulates an IPv4 traceroute packet, it will use the 2403 16-bit number as the UDP source port in the encapsulating header. 2404 When the ICMP Time Exceeded message is returned to the ITR, the UDP 2405 header of the encapsulating header is present in the ICMP payload 2406 thereby allowing the ITR to find the cached headers for the 2407 traceroute source. The ITR puts the cached headers in the payload 2408 and sends the ICMP Time Exceeded message to the traceroute source 2409 retaining the source address of the original ICMP Time Exceeded 2410 message (a core router or the ETR of the site of the traceroute 2411 destination). 2413 The signature of a traceroute packet comes in two forms. The first 2414 form is encoded as a UDP message where the destination port is 2415 inspected for a range of values. The second form is encoded as an 2416 ICMP message where the IP identification field is inspected for a 2417 well-known value. 2419 9.3. Traceroute using Mixed Locators 2421 When either an IPv4 traceroute or IPv6 traceroute is originated and 2422 the ITR encapsulates it in the other address family header, you 2423 cannot get all 3 segments of the traceroute. Segment 2 of the 2424 traceroute can not be conveyed to the traceroute source since it is 2425 expecting addresses from intermediate hops in the same address format 2426 for the type of traceroute it originated. Therefore, in this case, 2427 segment 2 will make the tunnel look like one hop. All the ITR has to 2428 do to make this work is to not copy the inner TTL to the outer, 2429 encapsulating header's TTL when a traceroute packet is encapsulated 2430 using an RLOC from a different address family. This will cause no 2431 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2433 10. Mobility Considerations 2435 There are several kinds of mobility of which only some might be of 2436 concern to LISP. Essentially they are as follows. 2438 10.1. Site Mobility 2440 A site wishes to change its attachment points to the Internet, and 2441 its LISP Tunnel Routers will have new RLOCs when it changes upstream 2442 providers. Changes in EID-RLOC mappings for sites are expected to be 2443 handled by configuration, outside of the LISP protocol. 2445 10.2. Slow Endpoint Mobility 2447 An individual endpoint wishes to move, but is not concerned about 2448 maintaining session continuity. Renumbering is involved. LISP can 2449 help with the issues surrounding renumbering [RFC4192] [LISA96] by 2450 decoupling the address space used by a site from the address spaces 2451 used by its ISPs. [RFC4984] 2453 10.3. Fast Endpoint Mobility 2455 Fast endpoint mobility occurs when an endpoint moves relatively 2456 rapidly, changing its IP layer network attachment point. Maintenance 2457 of session continuity is a goal. This is where the Mobile IPv4 2458 [RFC3344bis] and Mobile IPv6 [RFC3775] [RFC4866] mechanisms are used, 2459 and primarily where interactions with LISP need to be explored. 2461 The problem is that as an endpoint moves, it may require changes to 2462 the mapping between its EID and a set of RLOCs for its new network 2463 location. When this is added to the overhead of mobile IP binding 2464 updates, some packets might be delayed or dropped. 2466 In IPv4 mobility, when an endpoint is away from home, packets to it 2467 are encapsulated and forwarded via a home agent which resides in the 2468 home area the endpoint's address belongs to. The home agent will 2469 encapsulate and forward packets either directly to the endpoint or to 2470 a foreign agent which resides where the endpoint has moved to. 2471 Packets from the endpoint may be sent directly to the correspondent 2472 node, may be sent via the foreign agent, or may be reverse-tunneled 2473 back to the home agent for delivery to the mobile node. As the 2474 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 2475 mappings are required for communication between the mobile node and 2476 the home agent, whether via foreign agent or not. As a mobile 2477 endpoint changes networks, up to three LISP mapping changes may be 2478 required: 2480 o The mobile node moves from an old location to a new visited 2481 network location and notifies its home agent that it has done so. 2482 The Mobile IPv4 control packets the mobile node sends pass through 2483 one of the new visited network's ITRs, which needs a EID-RLOC 2484 mapping for the home agent. 2486 o The home agent might not have the EID-RLOC mappings for the mobile 2487 node's "care-of" address or its foreign agent in the new visited 2488 network, in which case it will need to acquire them. 2490 o When packets are sent directly to the correspondent node, it may 2491 be that no traffic has been sent from the new visited network to 2492 the correspondent node's network, and the new visited network's 2493 ITR will need to obtain an EID-RLOC mapping for the correspondent 2494 node's site. 2496 In addition, if the IPv4 endpoint is sending packets from the new 2497 visited network using its original EID, then LISP will need to 2498 perform a route-returnability check on the new EID-RLOC mapping for 2499 that EID. 2501 In IPv6 mobility, packets can flow directly between the mobile node 2502 and the correspondent node in either direction. The mobile node uses 2503 its "care-of" address (EID). In this case, the route-returnability 2504 check would not be needed but one more LISP mapping lookup may be 2505 required instead: 2507 o As above, three mapping changes may be needed for the mobile node 2508 to communicate with its home agent and to send packets to the 2509 correspondent node. 2511 o In addition, another mapping will be needed in the correspondent 2512 node's ITR, in order for the correspondent node to send packets to 2513 the mobile node's "care-of" address (EID) at the new network 2514 location. 2516 When both endpoints are mobile the number of potential mapping 2517 lookups increases accordingly. 2519 As a mobile node moves there are not only mobility state changes in 2520 the mobile node, correspondent node, and home agent, but also state 2521 changes in the ITRs and ETRs for at least some EID-prefixes. 2523 The goal is to support rapid adaptation, with little delay or packet 2524 loss for the entire system. Also IP mobility can be modified to 2525 require fewer mapping changes. In order to increase overall system 2526 performance, there may be a need to reduce the optimization of one 2527 area in order to place fewer demands on another. 2529 In LISP, one possibility is to "glean" information. When a packet 2530 arrives, the ETR could examine the EID-RLOC mapping and use that 2531 mapping for all outgoing traffic to that EID. It can do this after 2532 performing a route-returnability check, to ensure that the new 2533 network location does have a internal route to that endpoint. 2534 However, this does not cover the case where an ITR (the node assigned 2535 the RLOC) at the mobile-node location has been compromised. 2537 Mobile IP packet exchange is designed for an environment in which all 2538 routing information is disseminated before packets can be forwarded. 2539 In order to allow the Internet to grow to support expected future 2540 use, we are moving to an environment where some information may have 2541 to be obtained after packets are in flight. Modifications to IP 2542 mobility should be considered in order to optimize the behavior of 2543 the overall system. Anything which decreases the number of new EID- 2544 RLOC mappings needed when a node moves, or maintains the validity of 2545 an EID-RLOC mapping for a longer time, is useful. 2547 10.4. Fast Network Mobility 2549 In addition to endpoints, a network can be mobile, possibly changing 2550 xTRs. A "network" can be as small as a single router and as large as 2551 a whole site. This is different from site mobility in that it is 2552 fast and possibly short-lived, but different from endpoint mobility 2553 in that a whole prefix is changing RLOCs. However, the mechanisms 2554 are the same and there is no new overhead in LISP. A map request for 2555 any endpoint will return a binding for the entire mobile prefix. 2557 If mobile networks become a more common occurrence, it may be useful 2558 to revisit the design of the mapping service and allow for dynamic 2559 updates of the database. 2561 The issue of interactions between mobility and LISP needs to be 2562 explored further. Specific improvements to the entire system will 2563 depend on the details of mapping mechanisms. Mapping mechanisms 2564 should be evaluated on how well they support session continuity for 2565 mobile nodes. 2567 10.5. LISP Mobile Node Mobility 2569 A mobile device can use the LISP infrastructure to achieve mobility 2570 by implementing the LISP encapsulation and decapsulation functions 2571 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 2572 node" can use topologically-independent EID IP addresses that are not 2573 advertised into and do not impose a cost on the global routing 2574 system. These EIDs are maintained at the edges of the mapping system 2575 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 2576 only the correspondents of the LISP mobile node. 2578 Refer to the LISP Mobility Architecture specification [LISP-MN] for 2579 more details. 2581 11. Multicast Considerations 2583 A multicast group address, as defined in the original Internet 2584 architecture is an identifier of a grouping of topologically 2585 independent receiver host locations. The address encoding itself 2586 does not determine the location of the receiver(s). The multicast 2587 routing protocol, and the network-based state the protocol creates, 2588 determines where the receivers are located. 2590 In the context of LISP, a multicast group address is both an EID and 2591 a Routing Locator. Therefore, no specific semantic or action needs 2592 to be taken for a destination address, as it would appear in an IP 2593 header. Therefore, a group address that appears in an inner IP 2594 header built by a source host will be used as the destination EID. 2595 The outer IP header (the destination Routing Locator address), 2596 prepended by a LISP router, will use the same group address as the 2597 destination Routing Locator. 2599 Having said that, only the source EID and source Routing Locator 2600 needs to be dealt with. Therefore, an ITR merely needs to put its 2601 own IP address in the source Routing Locator field when prepending 2602 the outer IP header. This source Routing Locator address, like any 2603 other Routing Locator address MUST be globally routable. 2605 Therefore, an EID-to-RLOC mapping does not need to be performed by an 2606 ITR when a received data packet is a multicast data packet or when 2607 processing a source-specific Join (either by IGMPv3 or PIM). But the 2608 source Routing Locator is decided by the multicast routing protocol 2609 in a receiver site. That is, an EID to Routing Locator translation 2610 is done at control-time. 2612 Another approach is to have the ITR not encapsulate a multicast 2613 packet and allow the host built packet to flow into the core even if 2614 the source address is allocated out of the EID namespace. If the 2615 RPF-Vector TLV [RFC5496] is used by PIM in the core, then core 2616 routers can RPF to the ITR (the Locator address which is injected 2617 into core routing) rather than the host source address (the EID 2618 address which is not injected into core routing). 2620 To avoid any EID-based multicast state in the network core, the first 2621 approach is chosen for LISP-Multicast. Details for LISP-Multicast 2622 and Interworking with non-LISP sites is described in specification 2623 [MLISP]. 2625 12. Security Considerations 2627 It is believed that most of the security mechanisms will be part of 2628 the mapping database service when using control plane procedures for 2629 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 2630 as described in this specification, protection is provided against 2631 ETR spoofing by using Return- Routability mechanisms evidenced by the 2632 use of a 24-bit Nonce field in the LISP encapsulation header and a 2633 64-bit Nonce field in the LISP control message. The nonce, coupled 2634 with the ITR accepting only solicited Map-Replies goes a long way 2635 toward providing decent authentication. 2637 LISP does not rely on a PKI infrastructure or a more heavy weight 2638 authentication system. These systems challenge the scalability of 2639 LISP which was a primary design goal. 2641 DoS attack prevention will depend on implementations rate-limiting 2642 Map-Requests and Map-Replies to the control plane as well as rate- 2643 limiting the number of data-triggered Map-Replies. 2645 To deal with map-cache exhaustion attempts in an ITR/PTR, the 2646 implementation should consider putting a maximum cap on the number of 2647 entries stored with a reserve list for special or frequently accessed 2648 sites. This should be a configuration policy control set by the 2649 network administrator who manages ITRs and PTRs. 2651 13. IANA Considerations 2653 This specification has already allocated UDP port numbers 4341 and 2654 4342 assigned from the IANA registry. 2656 14. References 2658 14.1. Normative References 2660 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2661 August 1980. 2663 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2664 STD 13, RFC 1034, November 1987. 2666 [RFC1700] Reynolds, J. and J. Postel, "Assigned Numbers", RFC 1700, 2667 October 1994. 2669 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 2670 E. Lear, "Address Allocation for Private Internets", 2671 BCP 5, RFC 1918, February 1996. 2673 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2674 Requirement Levels", BCP 14, RFC 2119, March 1997. 2676 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 2677 ESP and AH", RFC 2404, November 1998. 2679 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2680 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2681 March 2000. 2683 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 2684 via IPv4 Clouds", RFC 3056, February 2001. 2686 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2687 of Explicit Congestion Notification (ECN) to IP", 2688 RFC 3168, September 2001. 2690 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2691 A., Peterson, J., Sparks, R., Handley, M., and E. 2692 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2693 June 2002. 2695 [RFC3775] Johnson, D., Perkins, C., and J. Arkko, "Mobility Support 2696 in IPv6", RFC 3775, June 2004. 2698 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 2699 Requirements for Security", BCP 106, RFC 4086, June 2005. 2701 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 2702 (CIDR): The Internet Address Assignment and Aggregation 2703 Plan", BCP 122, RFC 4632, August 2006. 2705 [RFC4634] Eastlake, D. and T. Hansen, "US Secure Hash Algorithms 2706 (SHA and HMAC-SHA)", RFC 4634, July 2006. 2708 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2709 Optimization for Mobile IPv6", RFC 4866, May 2007. 2711 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 2712 Workshop on Routing and Addressing", RFC 4984, 2713 September 2007. 2715 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2716 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2717 May 2008. 2719 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 2720 Forwarding (RPF) Vector TLV", RFC 5496, March 2009. 2722 [UDP-TUNNELS] 2723 Eubanks, M. and P. Chimento, "UDP Checksums for Tunneled 2724 Packets"", draft-eubanks-chimento-6man-00.txt (work in 2725 progress), February 2009. 2727 14.2. Informative References 2729 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 2730 NUMBERS http://www.iana.org/numbers.html, Febuary 2007. 2732 [ALT] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "LISP 2733 Alternative Topology (LISP-ALT)", 2734 draft-ietf-lisp-alt-04.txt (work in progress), April 2010. 2736 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 2737 Enhancement to the Internet Architecture", Internet- 2738 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt, 2739 1999. 2741 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 2742 Content distribution Overlay Network Service for LISP", 2743 draft-meyer-lisp-cons-03.txt (work in progress), 2744 November 2007. 2746 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 2747 Mappings Multicast Across Cooperating Systems for LISP", 2748 draft-curran-lisp-emacs-00.txt (work in progress), 2749 November 2007. 2751 [INTERWORK] 2752 Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 2753 "Interworking LISP with IPv4 and IPv6", 2754 draft-ietf-lisp-interworking-01.txt (work in progress), 2755 March 2010. 2757 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 2758 Address Format", draft-farinacci-lisp-lcaf-00.txt (work in 2759 progress), April 2010. 2761 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 2762 "Renumbering: Threat or Menace?", Usenix , September 1996. 2764 [LISP-MAIN] 2765 Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 2766 "Locator/ID Separation Protocol (LISP)", 2767 draft-farinacci-lisp-12.txt (work in progress), 2768 March 2009. 2770 [LISP-MN] Farinacci, D., Fuller, V., Lewis, D., and D. Meyer, "LISP 2771 Mobility Architecture", draft-meyer-lisp-mn-03.txt (work 2772 in progress), August 2010. 2774 [LISP-MS] Farinacci, D. and V. Fuller, "LISP Map Server", 2775 draft-ietf-lisp-ms-05.txt (work in progress), April 2010. 2777 [LOC-ID-ARCH] 2778 Meyer, D. and D. Lewis, "Architectural Implications of 2779 Locator/ID Separation", 2780 draft-meyer-loc-id-implications-01.txt (work in progress), 2781 Januaryr 2009. 2783 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 2784 "LISP for Multicast Environments", 2785 draft-ietf-lisp-multicast-03.txt (work in progress), 2786 April 2010. 2788 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 2789 draft-lear-lisp-nerd-08.txt (work in progress), 2790 March 2010. 2792 [OPENLISP] 2793 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 2794 Report", draft-iannone-openlisp-implementation-01.txt 2795 (work in progress), July 2008. 2797 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 2798 draft-narten-radir-problem-statement-00.txt (work in 2799 progress), July 2007. 2801 [RFC3344bis] 2802 Perkins, C., "IP Mobility Support for IPv4, revised", 2803 draft-ietf-mip4-rfc3344bis-05 (work in progress), 2804 July 2007. 2806 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2807 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2808 September 2005. 2810 [RPMD] Handley, M., Huici, F., and A. Greenhalgh, "RPMD: Protocol 2811 for Routing Protocol Meta-data Dissemination", 2812 draft-handley-p2ppush-unpublished-2007726.txt (work in 2813 progress), July 2007. 2815 [VERSIONING] 2816 Iannone, L., Saucez, D., and O. Bonaventure, "LISP Mapping 2817 Versioning", draft-iannone-lisp-mapping-versioning-01.txt 2818 (work in progress), March 2010. 2820 Appendix A. Acknowledgments 2822 An initial thank you goes to Dave Oran for planting the seeds for the 2823 initial ideas for LISP. His consultation continues to provide value 2824 to the LISP authors. 2826 A special and appreciative thank you goes to Noel Chiappa for 2827 providing architectural impetus over the past decades on separation 2828 of location and identity, as well as detailed review of the LISP 2829 architecture and documents, coupled with enthusiasm for making LISP a 2830 practical and incremental transition for the Internet. 2832 The authors would like to gratefully acknowledge many people who have 2833 contributed discussion and ideas to the making of this proposal. 2834 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 2835 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 2836 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 2837 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 2838 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 2839 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Roger 2840 Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van Beijnum, Roland 2841 Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien Saucez, Damian 2842 Lezama, Attilla De Groot, Parantap Lahiri, David Black, Roque 2843 Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, Margaret 2844 Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari Arkko, 2845 Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, Dhirendra 2846 Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 2847 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, and Vina 2848 Ermagan. 2850 This work originated in the Routing Research Group (RRG) of the IRTF. 2851 The individual submission [LISP-MAIN] was converted into this IETF 2852 LISP working group draft. 2854 Appendix B. Document Change Log 2856 B.1. Changes to draft-ietf-lisp-08.txt 2858 o Posted August 2010. 2860 o In section 6.1.6, remove statement about setting TTL to 0 in Map- 2861 Register messages. 2863 o Clarify language in section 6.1.5 about Map-Replying to Data- 2864 Probes or Map-Requests. 2866 o Indicate that outer TTL should only be copied to inner TTL when it 2867 is less than inner TTL. 2869 o Indicate a source-EID for RLOC-probes are encoded with an AFI 2870 value of 0. 2872 o Indicate that SMRs can have a global or per SMR destination rate- 2873 limiter. 2875 o Add clarifications to the SMR procedures. 2877 o Add definitions for "client-side" and 'server-side" terms used in 2878 this specification. 2880 o Clear up language in section 6.4, last paragraph. 2882 o Change ACT of value 0 to "no-action". This is so we can RLOC- 2883 probe a PETR and have it return a Map-Reply with a locator-set of 2884 size 0. The way it is spec'ed the map-cache entry has action 2885 "dropped". Drop-action is set to 3. 2887 o Add statement about normalizing locator weights. 2889 o Clarify R-bit definition in the Map-Reply locator record. 2891 o Add section on EID Reachability within a LISP site. 2893 o Clarify another disadvantage of using anycast locators. 2895 o Reworded Abstract. 2897 o Change section 2.0 Introduction to remove obsolete information 2898 such as the LISP variant definitions. 2900 o Change section 5 title from "Tunneling Details" to "LISP 2901 Encapsulation Details". 2903 o Changes to section 5 to include results of network deployment 2904 experience with MTU. Recommend that implementations use either 2905 the stateful or stateless handling. 2907 o Make clarification wordsmithing to Section 7 and 8. 2909 o Identify that if there is one locator in the locator-set of a map- 2910 cache entry, that an SMR from that locator should be responded to 2911 by sending the the SMR-invoked Map-Request to the database mapping 2912 system rather than to the RLOC itself (which may be unreachable). 2914 o When describing Unicast and Multicast Weights indicate the the 2915 values are relative weights rather than percentages. So it 2916 doesn't imply the sum of all locator weights in the locator-set 2917 need to be 100. 2919 o Do some wordsmithing on copying TTL and TOS fields. 2921 o Numerous wordsmithing changes from Dave Meyer. He fine toothed 2922 combed the spec. 2924 o Removed Section 14 "Prototype Plans and Status". We felt this 2925 type of section is no longer appropriate for a protocol 2926 specification. 2928 o Add clarification text for the IRC description per Damien's 2929 commentary. 2931 o Remove text on copying nonce from SMR to SMR-invoked Map- Request 2932 per Vina's comment about a possible DoS vector. 2934 o Clarify (S/2 + H) in the stateless MTU section. 2936 o Add text to reflect Damien's comment about the description of the 2937 "ITR-RLOC Address" field in the Map-Request. that the list of RLOC 2938 addresses are local addresses of the Map-Requester. 2940 B.2. Changes to draft-ietf-lisp-07.txt 2942 o Posted April 2010. 2944 o Added I-bit to data header so LSB field can also be used as an 2945 Instance ID field. When this occurs, the LSB field is reduced to 2946 8-bits (from 32-bits). 2948 o Added V-bit to the data header so the 24-bit nonce field can also 2949 be used for source and destination version numbers. 2951 o Added Map-Version 12-bit value to the EID-record to be used in all 2952 of Map-Request, Map-Reply, and Map-Register messages. 2954 o Added multiple ITR-RLOC fields to the Map-Request packet so an ETR 2955 can decide what address to select for the destination of a Map- 2956 Reply. 2958 o Added L-bit (Local RLOC bit) and p-bit (Probe-Reply RLOC bit) to 2959 the Locator-Set record of an EID-record for a Map-Reply message. 2960 The L-bit indicates which RLOCs in the locator-set are local to 2961 the sender of the message. The P-bit indicates which RLOC is the 2962 source of a RLOC-probe Reply (Map-Reply) message. 2964 o Add reference to the LISP Canonical Address Format [LCAF] draft. 2966 o Made editorial and clarification changes based on comments from 2967 Dhirendra Trivedi. 2969 o Added wordsmithing comments from Joel Halpern on DF=1 setting. 2971 o Add John Zwiebel clarification to Echo Nonce Algorithm section 2972 6.3.1. 2974 o Add John Zwiebel comment about expanding on proxy-map-reply bit 2975 for Map-Register messages. 2977 o Add NAT section per Ron Bonica comments. 2979 o Fix IDnits issues per Ron Bonica. 2981 o Added section on Virtualization and Segmentation to explain the 2982 use if the Instance ID field in the data header. 2984 o There are too many P-bits, keep their scope to the packet format 2985 description and refer to them by name every where else in the 2986 spec. 2988 o Scanned all occurrences of "should", "should not", "must" and 2989 "must not" and uppercased them. 2991 o John Zwiebel offered text for section 4.1 to modernize the 2992 example. Thanks Z! 2994 o Make it more clear in the definition of "EID-to-RLOC Database" 2995 that all ETRs need to have the same database mapping. This 2996 reflects a comment from John Scudder. 2998 o Add a definition "Route-returnability" to the Definition of Terms 2999 section. 3001 o In section 9.2, add text to describe what the signature of 3002 traceroute packets can look like. 3004 o Removed references to Data Probe for introductory example. Data- 3005 probes are still part of the LISP design but not encouraged. 3007 o Added the definition for "LISP site" to the Definition of Terms" 3008 section. 3010 B.3. Changes to draft-ietf-lisp-06.txt 3012 Editorial based changes: 3014 o Posted December 2009. 3016 o Fix typo for flags in LISP data header. Changed from "4" to "5". 3018 o Add text to indicate that Map-Register messages must contain a 3019 computed UDP checksum. 3021 o Add definitions for PITR and PETR. 3023 o Indicate an AFI value of 0 is an unspecified address. 3025 o Indicate that the TTL field of a Map-Register is not used and set 3026 to 0 by the sender. This change makes this spec consistent with 3027 [LISP-MS]. 3029 o Change "... yield a packet size of L bytes" to "... yield a packet 3030 size greater than L bytes". 3032 o Clarify section 6.1.5 on what addresses and ports are used in Map- 3033 Reply messages. 3035 o Clarify that LSBs that go beyond the number of locators do not to 3036 be SMRed when the locator addresses are greater lexicographically 3037 than the locator in the existing locator-set. 3039 o Add Gregg, Srini, and Amit to acknowledgment section. 3041 o Clarify in the definition of a LISP header what is following the 3042 UDP header. 3044 o Clarify "verifying Map-Request" text in section 6.1.3. 3046 o Add Xu Xiaohu to the acknowledgment section for introducing the 3047 problem of overlapping EID-prefixes among multiple sites in an RRG 3048 email message. 3050 Design based changes: 3052 o Use stronger language to have the outer IPv4 header set DF=1 so we 3053 can avoid fragment reassembly in an ETR or PETR. This will also 3054 make IPv4 and IPv6 encapsulation have consistent behavior. 3056 o Map-Requests should not be sent in ECM with the Probe bit is set. 3057 These type of Map-Requests are used as RLOC-probes and are sent 3058 directly to locator addresses in the underlying network. 3060 o Add text in section 6.1.5 about returning all EID-prefixes in a 3061 Map-Reply sent by an ETR when there are overlapping EID-prefixes 3062 configure. 3064 o Add text in a new subsection of section 6.1.5 about dealing with 3065 Map-Replies with coarse EID-prefixes. 3067 B.4. Changes to draft-ietf-lisp-05.txt 3069 o Posted September 2009. 3071 o Added this Document Change Log appendix. 3073 o Added section indicating that encapsulated Map-Requests must use 3074 destination UDP port 4342. 3076 o Don't use AH in Map-Registers. Put key-id, auth-length, and auth- 3077 data in Map-Register payload. 3079 o Added Jari to acknowledgment section. 3081 o State the source-EID is set to 0 when using Map-Requests to 3082 refresh or RLOC-probe. 3084 o Make more clear what source-RLOC should be for a Map-Request. 3086 o The LISP-CONS authors thought that the Type definitions for CONS 3087 should be removed from this specification. 3089 o Removed nonce from Map-Register message, it wasn't used so no need 3090 for it. 3092 o Clarify what to do for unspecified Action bits for negative Map- 3093 Replies. Since No Action is a drop, make value 0 Drop. 3095 B.5. Changes to draft-ietf-lisp-04.txt 3097 o Posted September 2009. 3099 o How do deal with record count greater than 1 for a Map-Request. 3100 Damien and Joel comment. Joel suggests: 1) Specify that senders 3101 compliant with the current document will always set the count to 3102 1, and note that the count is included for future extensibility. 3103 2) Specify what a receiver compliant with the draft should do if 3104 it receives a request with a count greater than 1. Presumably, it 3105 should send some error back? 3107 o Add Fred Templin in acknowledgment section. 3109 o Add Margaret and Sam to the acknowledgment section for their great 3110 comments. 3112 o Say more about LAGs in the UDP section per Sam Hartman's comment. 3114 o Sam wants to use MAY instead of SHOULD for ignoring checksums on 3115 ETR. From the mailing list: "You'd need to word it as an ITR MAY 3116 send a zero checksum, an ETR MUST accept a 0 checksum and MAY 3117 ignore the checksum completely. And of course we'd need to 3118 confirm that can actually be implemented. In particular, hardware 3119 that verifies UDP checksums on receive needs to be checked to make 3120 sure it permits 0 checksums." 3122 o Margaret wants a reference to 3123 http://www.ietf.org/id/draft-eubanks-chimento-6man-00.txt. 3125 o Fix description in Map-Request section. Where we describe Map- 3126 Reply Record, change "R-bit" to "M-bit". 3128 o Add the mobility bit to Map-Replies. So PTRs don't probe so often 3129 for MNs but often enough to get mapping updates. 3131 o Indicate SHA1 can be used as well for Map-Registers. 3133 o More Fred comments on MTU handling. 3135 o Isidor comment about spec'ing better periodic Map-Registers. Will 3136 be fixed in draft-ietf-lisp-ms-02.txt. 3138 o Margaret's comment on gleaning: "The current specification does 3139 not make it clear how long gleaned map entries should be retained 3140 in the cache, nor does it make it clear how/ when they will be 3141 validated. The LISP spec should, at the very least, include a 3142 (short) default lifetime for gleaned entries, require that they be 3143 validated within a short period of time, and state that a new 3144 gleaned entry should never overwrite an entry that was obtained 3145 from the mapping system. The security implications of storing 3146 "gleaned" entries should also be explored in detail." 3148 o Add section on RLOC-probing per working group feedback. 3150 o Change "loc-reach-bits" to "loc-status-bits" per comment from 3151 Noel. 3153 o Remove SMR-bit from data-plane. Dino prefers to have it in the 3154 control plane only. 3156 o Change LISP header to allow a "Research Bit" so the Nonce and LSB 3157 fields can be turned off and used for another future purpose. For 3158 Luigi et al versioning convergence. 3160 o Add a N-bit to the data header suggested by Noel. Then the nonce 3161 field could be used when N is not 1. 3163 o Clarify that when E-bit is 0, the nonce field can be an echoed 3164 nonce or a random nonce. Comment from Jesper. 3166 o Indicate when doing data-gleaning that a verifying Map-Request is 3167 sent to the source-EID of the gleaned data packet so we can avoid 3168 map-cache corruption by a 3rd party. Comment from Pedro. 3170 o Indicate that a verifying Map-Request, for accepting mapping data, 3171 should be sent over the ALT (or to the EID). 3173 o Reference IPsec RFC 4302. Comment from Sam and Brian Weis. 3175 o Put E-bit in Map-Reply to tell ITRs that the ETR supports echo- 3176 noncing. Comment by Pedro and Dino. 3178 o Jesper made a comment to loosen the language about requiring the 3179 copy of inner TTL to outer TTL since the text to get mixed-AF 3180 traceroute to work would violate the "MUST" clause. Changed from 3181 MUST to SHOULD in section 5.3. 3183 B.6. Changes to draft-ietf-lisp-03.txt 3185 o Posted July 2009. 3187 o Removed loc-reach-bits longword from control packets per Damien 3188 comment. 3190 o Clarifications in MTU text from Roque. 3192 o Added text to indicate that the locator-set be sorted by locator 3193 address from Isidor. 3195 o Clarification text from John Zwiebel in Echo-Nonce section. 3197 B.7. Changes to draft-ietf-lisp-02.txt 3199 o Posted July 2009. 3201 o Encapsulation packet format change to add E-bit and make loc- 3202 reach-bits 32-bits in length. 3204 o Added Echo-Nonce Algorithm section. 3206 o Clarification how ECN bits are copied. 3208 o Moved S-bit in Map-Request. 3210 o Added P-bit in Map-Request and Map-Reply messages to anticipate 3211 RLOC-Probe Algorithm. 3213 o Added to Mobility section to reference [LISP-MN]. 3215 B.8. Changes to draft-ietf-lisp-01.txt 3217 o Posted 2 days after draft-ietf-lisp-00.txt in May 2009. 3219 o Defined LEID to be a "LISP EID". 3221 o Indicate encapsulation use IPv4 DF=0. 3223 o Added negative Map-Reply messages with drop, native-forward, and 3224 send-map-request actions. 3226 o Added Proxy-Map-Reply bit to Map-Register. 3228 B.9. Changes to draft-ietf-lisp-00.txt 3230 o Posted May 2009. 3232 o Rename of draft-farinacci-lisp-12.txt. 3234 o Acknowledgment to RRG. 3236 Authors' Addresses 3238 Dino Farinacci 3239 cisco Systems 3240 Tasman Drive 3241 San Jose, CA 95134 3242 USA 3244 Email: dino@cisco.com 3246 Vince Fuller 3247 cisco Systems 3248 Tasman Drive 3249 San Jose, CA 95134 3250 USA 3252 Email: vaf@cisco.com 3254 Dave Meyer 3255 cisco Systems 3256 170 Tasman Drive 3257 San Jose, CA 3258 USA 3260 Email: dmm@cisco.com 3262 Darrel Lewis 3263 cisco Systems 3264 170 Tasman Drive 3265 San Jose, CA 3266 USA 3268 Email: darlewis@cisco.com