idnits 2.17.1 draft-ietf-lisp-16.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == There are 10 instances of lines with private range IPv4 addresses in the document. If these are generic example addresses, they should be changed to use any of the ranges defined in RFC 6890 (or successor): 192.0.2.x, 198.51.100.x or 203.0.113.x. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 31, 2011) is 4566 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- == Outdated reference: A later version (-10) exists of draft-ietf-lisp-alt-09 == Outdated reference: A later version (-16) exists of draft-ietf-lisp-ms-12 ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) == Outdated reference: A later version (-09) exists of draft-ietf-lisp-map-versioning-05 == Outdated reference: A later version (-06) exists of draft-ietf-lisp-interworking-02 == Outdated reference: A later version (-10) exists of draft-ietf-karp-design-guide-06 == Outdated reference: A later version (-10) exists of draft-farinacci-lisp-lcaf-06 == Outdated reference: A later version (-12) exists of draft-ietf-lisp-deployment-02 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-mib-02 == Outdated reference: A later version (-16) exists of draft-meyer-lisp-mn-06 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-00 == Outdated reference: A later version (-14) exists of draft-ietf-lisp-multicast-10 == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-08 -- No information found for draft-handley-p2ppush-unpublished-2007726 - is the name correct? Summary: 1 error (**), 0 flaws (~~), 14 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Farinacci 3 Internet-Draft V. Fuller 4 Intended status: Experimental D. Meyer 5 Expires: May 3, 2012 D. Lewis 6 cisco Systems 7 October 31, 2011 9 Locator/ID Separation Protocol (LISP) 10 draft-ietf-lisp-16 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 to early adopters, even 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 in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at http://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on May 3, 2012. 44 Copyright Notice 46 Copyright (c) 2011 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (http://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 62 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 63 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 7 64 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . . 12 65 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . . 14 66 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . . 16 67 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 17 68 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 17 69 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . . 19 70 5.4. Dealing with Large Encapsulated Packets . . . . . . . . . 23 71 5.4.1. A Stateless Solution to MTU Handling . . . . . . . . . 23 72 5.4.2. A Stateful Solution to MTU Handling . . . . . . . . . 24 73 5.5. Using Virtualization and Segmentation with LISP . . . . . 24 74 6. EID-to-RLOC Mapping . . . . . . . . . . . . . . . . . . . . . 26 75 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats . . . . . 26 76 6.1.1. LISP Packet Type Allocations . . . . . . . . . . . . . 28 77 6.1.2. Map-Request Message Format . . . . . . . . . . . . . . 28 78 6.1.3. EID-to-RLOC UDP Map-Request Message . . . . . . . . . 31 79 6.1.4. Map-Reply Message Format . . . . . . . . . . . . . . . 32 80 6.1.5. EID-to-RLOC UDP Map-Reply Message . . . . . . . . . . 36 81 6.1.6. Map-Register Message Format . . . . . . . . . . . . . 38 82 6.1.7. Map-Notify Message Format . . . . . . . . . . . . . . 40 83 6.1.8. Encapsulated Control Message Format . . . . . . . . . 41 84 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 43 85 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 45 86 6.3.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . 47 87 6.3.2. RLOC Probing Algorithm . . . . . . . . . . . . . . . . 48 88 6.4. EID Reachability within a LISP Site . . . . . . . . . . . 49 89 6.5. Routing Locator Hashing . . . . . . . . . . . . . . . . . 50 90 6.6. Changing the Contents of EID-to-RLOC Mappings . . . . . . 51 91 6.6.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 52 92 6.6.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 52 93 6.6.3. Database Map Versioning . . . . . . . . . . . . . . . 54 94 7. Router Performance Considerations . . . . . . . . . . . . . . 55 95 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 56 96 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 57 97 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 57 98 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 58 99 8.4. LISP Functionality with Conventional NATs . . . . . . . . 58 100 8.5. Packets Egressing a LISP Site . . . . . . . . . . . . . . 59 101 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 60 102 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 61 103 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 61 104 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 61 105 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 63 106 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 63 107 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 63 108 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 63 109 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 65 110 10.5. LISP Mobile Node Mobility . . . . . . . . . . . . . . . . 65 111 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 67 112 12. Security Considerations . . . . . . . . . . . . . . . . . . . 68 113 13. Network Management Considerations . . . . . . . . . . . . . . 70 114 14. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 71 115 14.1. LISP ACT and Flag Fields . . . . . . . . . . . . . . . . . 71 116 14.2. LISP Address Type Codes . . . . . . . . . . . . . . . . . 71 117 14.3. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . . 71 118 14.4. LISP Key ID Numbers . . . . . . . . . . . . . . . . . . . 72 119 15. Known Open Issues and Areas of Future Work . . . . . . . . . . 73 120 16. References . . . . . . . . . . . . . . . . . . . . . . . . . . 74 121 16.1. Normative References . . . . . . . . . . . . . . . . . . . 74 122 16.2. Informative References . . . . . . . . . . . . . . . . . . 75 123 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 78 124 Appendix B. Document Change Log . . . . . . . . . . . . . . . . . 79 125 B.1. Changes to draft-ietf-lisp-16.txt . . . . . . . . . . . . 79 126 B.2. Changes to draft-ietf-lisp-15.txt . . . . . . . . . . . . 79 127 B.3. Changes to draft-ietf-lisp-14.txt . . . . . . . . . . . . 79 128 B.4. Changes to draft-ietf-lisp-13.txt . . . . . . . . . . . . 80 129 B.5. Changes to draft-ietf-lisp-12.txt . . . . . . . . . . . . 80 130 B.6. Changes to draft-ietf-lisp-11.txt . . . . . . . . . . . . 82 131 B.7. Changes to draft-ietf-lisp-10.txt . . . . . . . . . . . . 82 132 B.8. Changes to draft-ietf-lisp-09.txt . . . . . . . . . . . . 83 133 B.9. Changes to draft-ietf-lisp-08.txt . . . . . . . . . . . . 83 134 B.10. Changes to draft-ietf-lisp-07.txt . . . . . . . . . . . . 85 135 B.11. Changes to draft-ietf-lisp-06.txt . . . . . . . . . . . . 87 136 B.12. Changes to draft-ietf-lisp-05.txt . . . . . . . . . . . . 88 137 B.13. Changes to draft-ietf-lisp-04.txt . . . . . . . . . . . . 88 138 B.14. Changes to draft-ietf-lisp-03.txt . . . . . . . . . . . . 90 139 B.15. Changes to draft-ietf-lisp-02.txt . . . . . . . . . . . . 90 140 B.16. Changes to draft-ietf-lisp-01.txt . . . . . . . . . . . . 91 141 B.17. Changes to draft-ietf-lisp-00.txt . . . . . . . . . . . . 91 142 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 92 144 1. Requirements Notation 146 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 147 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 148 document are to be interpreted as described in [RFC2119]. 150 2. Introduction 152 This document describes the Locator/Identifier Separation Protocol 153 (LISP), which provides a set of functions for routers to exchange 154 information used to map from non-routeable Endpoint Identifiers 155 (EIDs) to routeable Routing Locators (RLOCs). It also defines a 156 mechanism for these LISP routers to encapsulate IP packets addressed 157 with EIDs for transmission across an Internet that uses RLOCs for 158 routing and forwarding. 160 Creation of LISP was initially motivated by discussions during the 161 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 162 October, 2006 (see [RFC4984]). A key conclusion of the workshop was 163 that the Internet routing and addressing system was not scaling well 164 in the face of the explosive growth of new sites; one reason for this 165 poor scaling is the increasing number of multi-homed and other sites 166 that cannot be addressed as part of topologically- or provider-based 167 aggregated prefixes. Additional work that more completely described 168 the problem statement may be found in [RADIR]. 170 A basic observation, made many years ago in early networking research 171 such as that documented in [CHIAPPA] and [RFC4984], is that using a 172 single address field for both identifying a device and for 173 determining where it is topologically located in the network requires 174 optimization along two conflicting axes: for routing to be efficient, 175 the address must be assigned topologically; for collections of 176 devices to be easily and effectively managed, without the need for 177 renumbering in response to topological change (such as that caused by 178 adding or removing attachment points to the network or by mobility 179 events), the address must explicitly not be tied to the topology. 181 The approach that LISP takes to solving the routing scalability 182 problem is to replace IP addresses with two new types of numbers: 183 Routing Locators (RLOCs), which are topologically assigned to network 184 attachment points (and are therefore amenable to aggregation) and 185 used for routing and forwarding of packets through the network; and 186 Endpoint Identifiers (EIDs), which are assigned independently from 187 the network topology, are used for numbering devices, and are 188 aggregated along administrative boundaries. LISP then defines 189 functions for mapping between the two numbering spaces and for 190 encapsulating traffic originated by devices using non-routeable EIDs 191 for transport across a network infrastructure that routes and 192 forwards using RLOCs. Both RLOCs and EIDs are syntactically- 193 identical to IP addresses; it is the semantics of how they are used 194 that differs. 196 This document describes the protocol that implements these functions. 197 The database which stores the mappings between EIDs and RLOCs is 198 explicitly a separate "module" to facilitate experimentation with a 199 variety of approaches. One database design that is being developed 200 and prototyped as part of the LISP working group work is [ALT]. 201 Others that have been described but not implemented include [CONS], 202 [EMACS], [RPMD], [NERD]. Finally, [LISP-MS], documents a general- 203 purpose service interface for accessing a mapping database; this 204 interface is intended to make the mapping database modular so that 205 different approaches can be tried without the need to modify 206 installed xTRs. 208 This experimental specification has areas that require additional 209 experience and measurement. Results of such work may lead to 210 modifications and enhancements of protocol mechanisms defined in this 211 document. See Section 15 for specific, known issues that are in need 212 of further work during development, implementation, and prototype 213 deployment. 215 3. Definition of Terms 217 Provider Independent (PI) Addresses: PI addresses are an address 218 block assigned from a pool where blocks are not associated with 219 any particular location in the network (e.g. from a particular 220 service provider), and is therefore not topologically aggregatable 221 in the routing system. 223 Provider Assigned (PA) Addresses: PA addresses are an a address 224 block assigned to a site by each service provider to which a site 225 connects. Typically, each block is sub-block of a service 226 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 227 is aggregated into the larger block before being advertised into 228 the global Internet. Traditionally, IP multihoming has been 229 implemented by each multi-homed site acquiring its own, globally- 230 visible prefix. LISP uses only topologically-assigned and 231 aggregatable address blocks for RLOCs, eliminating this 232 demonstrably non-scalable practice. 234 Routing Locator (RLOC): A RLOC is an IPv4 or IPv6 address of an 235 egress tunnel router (ETR). A RLOC is the output of a EID-to-RLOC 236 mapping lookup. An EID maps to one or more RLOCs. Typically, 237 RLOCs are numbered from topologically-aggregatable blocks that are 238 assigned to a site at each point to which it attaches to the 239 global Internet; where the topology is defined by the connectivity 240 of provider networks, RLOCs can be thought of as PA addresses. 241 Multiple RLOCs can be assigned to the same ETR device or to 242 multiple ETR devices at a site. 244 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 245 IPv6) value used in the source and destination address fields of 246 the first (most inner) LISP header of a packet. The host obtains 247 a destination EID the same way it obtains an destination address 248 today, for example through a Domain Name System (DNS) [RFC1034] 249 lookup or Session Invitation Protocol (SIP) [RFC3261] exchange. 250 The source EID is obtained via existing mechanisms used to set a 251 host's "local" IP address. An EID is allocated to a host from an 252 EID-prefix block associated with the site where the host is 253 located. An EID can be used by a host to refer to other hosts. 254 EIDs MUST NOT be used as LISP RLOCs. Note that EID blocks may be 255 assigned in a hierarchical manner, independent of the network 256 topology, to facilitate scaling of the mapping database. In 257 addition, an EID block assigned to a site may have site-local 258 structure (subnetting) for routing within the site; this structure 259 is not visible to the global routing system. In theory, the bit 260 string that represents an EID for one device can represent an RLOC 261 for a different device. As the architecture is realized, if a 262 given bit string is both an RLOC and an EID, it must refer to the 263 same entity in both cases. When used in discussions with other 264 Locator/ID separation proposals, a LISP EID will be called a 265 "LEID". Throughout this document, any references to "EID" refers 266 to an LEID. 268 EID-prefix: An EID-prefix is a power-of-two block of EIDs which are 269 allocated to a site by an address allocation authority. EID- 270 prefixes are associated with a set of RLOC addresses which make up 271 a "database mapping". EID-prefix allocations can be broken up 272 into smaller blocks when an RLOC set is to be associated with the 273 smaller EID-prefix. A globally routed address block (whether PI 274 or PA) is not inherently an EID-prefix. A globally routed address 275 block may be used by its assignee as an EID block. The converse 276 is not supported. That is, a site which receives an explicitly 277 allocated EID-prefix may not use that EID-prefix as a globally 278 prefix. This would require coordination and cooperation with the 279 entities managing the mapping infrastructure. Once this has been 280 done, that block could be removed from the globally routed IP 281 system, if other suitable transition and access mechanisms are in 282 place. Discussion of such transition and access mechanisms can be 283 found in [INTERWORK] and [LISP-DEPLOY]. 285 End-system: An end-system is an IPv4 or IPv6 device that originates 286 packets with a single IPv4 or IPv6 header. The end-system 287 supplies an EID value for the destination address field of the IP 288 header when communicating globally (i.e. outside of its routing 289 domain). An end-system can be a host computer, a switch or router 290 device, or any network appliance. 292 Ingress Tunnel Router (ITR): An ITR is a router which accepts an IP 293 packet with a single IP header (more precisely, an IP packet that 294 does not contain a LISP header). The router treats this "inner" 295 IP destination address as an EID and performs an EID-to-RLOC 296 mapping lookup. The router then prepends an "outer" IP header 297 with one of its globally-routable RLOCs in the source address 298 field and the result of the mapping lookup in the destination 299 address field. Note that this destination RLOC may be an 300 intermediate, proxy device that has better knowledge of the EID- 301 to-RLOC mapping closer to the destination EID. In general, an ITR 302 receives IP packets from site end-systems on one side and sends 303 LISP-encapsulated IP packets toward the Internet on the other 304 side. 306 Specifically, when a service provider prepends a LISP header for 307 Traffic Engineering purposes, the router that does this is also 308 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 309 on the outer destination address (the originating ITR's supplied 310 RLOC) or the inner destination address (the originating hosts 311 supplied EID). 313 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 314 network that prepends an additional LISP header for Traffic 315 Engineering purposes. 317 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 318 packet where the destination address in the "outer" IP header is 319 one of its own RLOCs. The router strips the "outer" header and 320 forwards the packet based on the next IP header found. In 321 general, an ETR receives LISP-encapsulated IP packets from the 322 Internet on one side and sends decapsulated IP packets to site 323 end-systems on the other side. ETR functionality does not have to 324 be limited to a router device. A server host can be the endpoint 325 of a LISP tunnel as well. 327 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 328 network that strips an outer LISP header for Traffic Engineering 329 purposes. 331 xTR: A xTR is a reference to an ITR or ETR when direction of data 332 flow is not part of the context description. xTR refers to the 333 router that is the tunnel endpoint. Used synonymously with the 334 term "Tunnel Router". For example, "An xTR can be located at the 335 Customer Edge (CE) router", meaning both ITR and ETR functionality 336 is at the CE router. 338 EID-to-RLOC Cache: The EID-to-RLOC cache is a short-lived, on- 339 demand table in an ITR that stores, tracks, and is responsible for 340 timing-out and otherwise validating EID-to-RLOC mappings. This 341 cache is distinct from the full "database" of EID-to-RLOC 342 mappings, it is dynamic, local to the ITR(s), and relatively small 343 while the database is distributed, relatively static, and much 344 more global in scope. 346 EID-to-RLOC Database: The EID-to-RLOC database is a global 347 distributed database that contains all known EID-prefix to RLOC 348 mappings. Each potential ETR typically contains a small piece of 349 the database: the EID-to-RLOC mappings for the EID prefixes 350 "behind" the router. These map to one of the router's own, 351 globally-visible, IP addresses. The same database mapping entries 352 MUST be configured on all ETRs for a given site. In a steady 353 state the EID-prefixes for the site and the locator-set for each 354 EID-prefix MUST be the same on all ETRs. Procedures to enforce 355 and/or verify this are outside the scope of this document. Note 356 that there may be transient conditions when the EID-prefix for the 357 site and locator-set for each EID-prefix may not be the same on 358 all ETRs. This has no negative implications. 360 Recursive Tunneling: Recursive tunneling occurs when a packet has 361 more than one LISP IP header. Additional layers of tunneling may 362 be employed to implement traffic engineering or other re-routing 363 as needed. When this is done, an additional "outer" LISP header 364 is added and the original RLOCs are preserved in the "inner" 365 header. Any references to tunnels in this specification refers to 366 dynamic encapsulating tunnels and never are they statically 367 configured. 369 Reencapsulating Tunnels: Reencapsulating tunneling occurs when an 370 ETR removes a LISP header, then acts as an ITR to prepend another 371 LISP header. Doing this allows a packet to be re-routed by the 372 re-encapsulating router without adding the overhead of additional 373 tunnel headers. Any references to tunnels in this specification 374 refers to dynamic encapsulating tunnels and never are they 375 statically configured. When using multiple mapping database 376 systems, care must be taken to not create reencapsulation loops. 378 LISP Header: a term used in this document to refer to the outer 379 IPv4 or IPv6 header, a UDP header, and a LISP-specific 8-byte 380 header that follows the UDP header, an ITR prepends or an ETR 381 strips. 383 Address Family Identifier (AFI): a term used to describe an address 384 encoding in a packet. An address family currently pertains to an 385 IPv4 or IPv6 address. See [AFI]/[AFI-REGISTRY] and [RFC3232] for 386 details. An AFI value of 0 used in this specification indicates 387 an unspecified encoded address where the length of the address is 388 0 bytes following the 16-bit AFI value of 0. 390 Negative Mapping Entry: A negative mapping entry, also known as a 391 negative cache entry, is an EID-to-RLOC entry where an EID-prefix 392 is advertised or stored with no RLOCs. That is, the locator-set 393 for the EID-to-RLOC entry is empty or has an encoded locator count 394 of 0. This type of entry could be used to describe a prefix from 395 a non-LISP site, which is explicitly not in the mapping database. 396 There are a set of well defined actions that are encoded in a 397 Negative Map-Reply. 399 Data Probe: A data-probe is a LISP-encapsulated data packet where 400 the inner header destination address equals the outer header 401 destination address used to trigger a Map-Reply by a decapsulating 402 ETR. In addition, the original packet is decapsulated and 403 delivered to the destination host if the destination EID is in the 404 EID-prefix range configured on the ETR. Otherwise, the packet is 405 discarded. A Data Probe is used in some of the mapping database 406 designs to "probe" or request a Map-Reply from an ETR; in other 407 cases, Map-Requests are used. See each mapping database design 408 for details. 410 Proxy ITR (PITR): A PITR is also known as a PTR is defined and 411 described in [INTERWORK], a PITR acts like an ITR but does so on 412 behalf of non-LISP sites which send packets to destinations at 413 LISP sites. 415 Proxy ETR (PETR): A PETR is defined and described in [INTERWORK], a 416 PETR acts like an ETR but does so on behalf of LISP sites which 417 send packets to destinations at non-LISP sites. 419 Route-returnability: is an assumption that the underlying routing 420 system will deliver packets to the destination. When combined 421 with a nonce that is provided by a sender and returned by a 422 receiver limits off-path data insertion. 424 LISP site: is a set of routers in an edge network that are under a 425 single technical administration. LISP routers which reside in the 426 edge network are the demarcation points to separate the edge 427 network from the core network. 429 Client-side: a term used in this document to indicate a connection 430 initiation attempt by an EID. The ITR(s) at the LISP site are the 431 first to get involved in obtaining database map cache entries by 432 sending Map-Request messages. 434 Server-side: a term used in this document to indicate a connection 435 initiation attempt is being accepted for a destination EID. The 436 ETR(s) at the destination LISP site are the first to send Map- 437 Replies to the source site initiating the connection. The ETR(s) 438 at this destination site can obtain mappings by gleaning 439 information from Map-Requests, Data-Probes, or encapsulated 440 packets. 442 Locator Status Bits (LSBs): Locator status bits are present in the 443 LISP header. They are used by ITRs to inform ETRs about the up/ 444 down status of all ETRs at the local site. These bits are used as 445 a hint to convey up/down router status and not path reachability 446 status. The LSBs can be verified by use of one of the Locator 447 Reachability Algorithms described in Section 6.3. 449 4. Basic Overview 451 One key concept of LISP is that end-systems (hosts) operate the same 452 way they do today. The IP addresses that hosts use for tracking 453 sockets, connections, and for sending and receiving packets do not 454 change. In LISP terminology, these IP addresses are called Endpoint 455 Identifiers (EIDs). 457 Routers continue to forward packets based on IP destination 458 addresses. When a packet is LISP encapsulated, these addresses are 459 referred to as Routing Locators (RLOCs). Most routers along a path 460 between two hosts will not change; they continue to perform routing/ 461 forwarding lookups on the destination addresses. For routers between 462 the source host and the ITR as well as routers from the ETR to the 463 destination host, the destination address is an EID. For the routers 464 between the ITR and the ETR, the destination address is an RLOC. 466 Another key LISP concept is the "Tunnel Router". A tunnel router 467 prepends LISP headers on host-originated packets and strip them prior 468 to final delivery to their destination. The IP addresses in this 469 "outer header" are RLOCs. During end-to-end packet exchange between 470 two Internet hosts, an ITR prepends a new LISP header to each packet 471 and an egress tunnel router strips the new header. The ITR performs 472 EID-to-RLOC lookups to determine the routing path to the ETR, which 473 has the RLOC as one of its IP addresses. 475 Some basic rules governing LISP are: 477 o End-systems (hosts) only send to addresses which are EIDs. They 478 don't know addresses are EIDs versus RLOCs but assume packets get 479 to LISP routers, which in turn, deliver packets to the destination 480 the end-system has specified. The procedure a host uses to send 481 IP packets does not change. 483 o EIDs are always IP addresses assigned to hosts. 485 o LISP routers mostly deal with Routing Locator addresses. See 486 details later in Section 4.1 to clarify what is meant by "mostly". 488 o RLOCs are always IP addresses assigned to routers; preferably, 489 topologically-oriented addresses from provider CIDR blocks. 491 o When a router originates packets it may use as a source address 492 either an EID or RLOC. When acting as a host (e.g. when 493 terminating a transport session such as SSH, TELNET, or SNMP), it 494 may use an EID that is explicitly assigned for that purpose. An 495 EID that identifies the router as a host MUST NOT be used as an 496 RLOC; an EID is only routable within the scope of a site. A 497 typical BGP configuration might demonstrate this "hybrid" EID/RLOC 498 usage where a router could use its "host-like" EID to terminate 499 iBGP sessions to other routers in a site while at the same time 500 using RLOCs to terminate eBGP sessions to routers outside the 501 site. 503 o EIDs are not expected to be usable for global end-to-end 504 communication in the absence of an EID-to-RLOC mapping operation. 505 They are expected to be used locally for intra-site communication. 507 o EID prefixes are likely to be hierarchically assigned in a manner 508 which is optimized for administrative convenience and to 509 facilitate scaling of the EID-to-RLOC mapping database. The 510 hierarchy is based on a address allocation hierarchy which is 511 independent of the network topology. 513 o EIDs may also be structured (subnetted) in a manner suitable for 514 local routing within an autonomous system. 516 An additional LISP header may be prepended to packets by a TE-ITR 517 when re-routing of the path for a packet is desired. A potential 518 use-case for this would be an ISP router that needs to perform 519 traffic engineering for packets flowing through its network. In such 520 a situation, termed Recursive Tunneling, an ISP transit acts as an 521 additional ingress tunnel router and the RLOC it uses for the new 522 prepended header would be either a TE-ETR within the ISP (along 523 intra-ISP traffic engineered path) or a TE-ETR within another ISP (an 524 inter-ISP traffic engineered path, where an agreement to build such a 525 path exists). 527 In order to avoid excessive packet overhead as well as possible 528 encapsulation loops, this document mandates that a maximum of two 529 LISP headers can be prepended to a packet. For initial LISP 530 deployments, it is assumed two headers is sufficient, where the first 531 prepended header is used at a site for Location/Identity separation 532 and second prepended header is used inside a service provider for 533 Traffic Engineering purposes. 535 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 536 For example, the ITR for a particular end-to-end packet exchange 537 might be the first-hop or default router within a site for the source 538 host. Similarly, the egress tunnel router might be the last-hop 539 router directly-connected to the destination host. Another example, 540 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 541 could be the site's border router at the service provider attachment 542 point. Mixing and matching of site-operated, ISP-operated, and other 543 tunnel routers is allowed for maximum flexibility. See Section 8 for 544 more details. 546 4.1. Packet Flow Sequence 548 This section provides an example of the unicast packet flow with the 549 following conditions: 551 o Source host "host1.abc.example.com" is sending a packet to 552 "host2.xyz.example.com", exactly what host1 would do if the site 553 was not using LISP. 555 o Each site is multi-homed, so each tunnel router has an address 556 (RLOC) assigned from the service provider address block for each 557 provider to which that particular tunnel router is attached. 559 o The ITR(s) and ETR(s) are directly connected to the source and 560 destination, respectively, but the source and destination can be 561 located anywhere in LISP site. 563 o Map-Requests can be sent on the underlying routing system topology 564 or over an alternative topology [ALT]. 566 o Map-Replies are sent on the underlying routing system topology. 568 Client host1.abc.example.com wants to communicate with server 569 host2.xyz.example.com: 571 1. host1.abc.example.com wants to open a TCP connection to 572 host2.xyz.example.com. It does a DNS lookup on 573 host2.xyz.example.com. An A/AAAA record is returned. This 574 address is the destination EID. The locally-assigned address of 575 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 576 packet is built and forwarded through the LISP site as a normal 577 IP packet until it reaches a LISP ITR. 579 2. The LISP ITR must be able to map the EID destination to an RLOC 580 of one of the ETRs at the destination site. The specific method 581 used to do this is not described in this example. See [ALT] or 582 [CONS] for possible solutions. 584 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be 585 rate-limited. 587 4. When an alternate mapping system is not in use, the Map-Request 588 packet is routed through the underlying routing system. 589 Otherwise, the Map-Request packet is routed on an alternate 590 logical topology, for example the [ALT] database mapping system. 591 In either case, when the Map-Request arrives at one of the ETRs 592 at the destination site, it will process the packet as a control 593 message. 595 5. The ETR looks at the destination EID of the Map-Request and 596 matches it against the prefixes in the ETR's configured EID-to- 597 RLOC mapping database. This is the list of EID-prefixes the ETR 598 is supporting for the site it resides in. If there is no match, 599 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 600 returned to the ITR. 602 6. The ITR receives the Map-Reply message, parses the message (to 603 check for format validity) and stores the mapping information 604 from the packet. This information is stored in the ITR's EID-to- 605 RLOC mapping cache. Note that the map cache is an on-demand 606 cache. An ITR will manage its map cache in such a way that 607 optimizes for its resource constraints. 609 7. Subsequent packets from host1.abc.example.com to 610 host2.xyz.example.com will have a LISP header prepended by the 611 ITR using the appropriate RLOC as the LISP header destination 612 address learned from the ETR. Note the packet may be sent to a 613 different ETR than the one which returned the Map-Reply due to 614 the source site's hashing policy or the destination site's 615 locator-set policy. 617 8. The ETR receives these packets directly (since the destination 618 address is one of its assigned IP addresses), checks the validity 619 of the addresses, strips the LISP header, and forwards packets to 620 the attached destination host. 622 In order to defer the need for a mapping lookup in the reverse 623 direction, an ETR MAY create a cache entry that maps the source EID 624 (inner header source IP address) to the source RLOC (outer header 625 source IP address) in a received LISP packet. Such a cache entry is 626 termed a "gleaned" mapping and only contains a single RLOC for the 627 EID in question. More complete information about additional RLOCs 628 SHOULD be verified by sending a LISP Map-Request for that EID. Both 629 ITR and the ETR may also influence the decision the other makes in 630 selecting an RLOC. See Section 6 for more details. 632 5. LISP Encapsulation Details 634 Since additional tunnel headers are prepended, the packet becomes 635 larger and can exceed the MTU of any link traversed from the ITR to 636 the ETR. It is recommended in IPv4 that packets do not get 637 fragmented as they are encapsulated by the ITR. Instead, the packet 638 is dropped and an ICMP Too Big message is returned to the source. 640 This specification recommends that implementations support for one of 641 the proposed fragmentation and reassembly schemes. These two 642 existing schemes are detailed in Section 5.4. 644 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 645 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 646 header is in IPv4 packet format and the other header is in IPv6 647 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 648 is in IPv6 packet format and the other header is in IPv4 packet 649 format). The next sub-sections illustrate packet formats for the 650 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6) but all 4 651 combinations MUST be supported. 653 5.1. LISP IPv4-in-IPv4 Header Format 655 0 1 2 3 656 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 1 657 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 658 / |Version| IHL |Type of Service| Total Length | 659 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 660 | | Identification |Flags| Fragment Offset | 661 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 662 OH | Time to Live | Protocol = 17 | Header Checksum | 663 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 664 | | Source Routing Locator | 665 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 666 \ | Destination Routing Locator | 667 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 668 / | Source Port = xxxx | Dest Port = 4341 | 669 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 670 \ | UDP Length | UDP Checksum | 671 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 672 L |N|L|E|V|I|flags| Nonce/Map-Version | 673 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 674 S / | Instance ID/Locator Status Bits | 675 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 676 / |Version| IHL |Type of Service| Total Length | 677 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 678 | | Identification |Flags| Fragment Offset | 679 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 680 IH | Time to Live | Protocol | Header Checksum | 681 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 682 | | Source EID | 683 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 684 \ | Destination EID | 685 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 687 5.2. LISP IPv6-in-IPv6 Header Format 689 0 1 2 3 690 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 1 691 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 692 / |Version| Traffic Class | Flow Label | 693 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 695 | | Payload Length | Next Header=17| Hop Limit | 696 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 697 | | 698 O + + 699 u | | 700 t + Source Routing Locator + 701 e | | 702 r + + 703 | | 704 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 705 d | | 706 r + + 707 | | 708 ^ + Destination Routing Locator + 709 | | | 710 \ + + 711 \ | | 712 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 713 / | Source Port = xxxx | Dest Port = 4341 | 714 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 715 \ | UDP Length | UDP Checksum | 716 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 717 L |N|L|E|V|I|flags| Nonce/Map-Version | 718 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 719 S / | Instance ID/Locator Status Bits | 720 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 721 / |Version| Traffic Class | Flow Label | 722 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 723 / | Payload Length | Next Header | Hop Limit | 724 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 725 | | 726 I + + 727 n | | 728 n + Source EID + 729 e | | 730 r + + 731 | | 732 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 733 d | | 734 r + + 735 | | 736 ^ + Destination EID + 737 \ | | 738 \ + + 739 \ | | 740 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 742 5.3. Tunnel Header Field Descriptions 744 Inner Header: The inner header is the header on the datagram 745 received from the originating host. The source and destination IP 746 addresses are EIDs. 748 Outer Header: The outer header is a new header prepended by an ITR. 749 The address fields contain RLOCs obtained from the ingress 750 router's EID-to-RLOC cache. The IP protocol number is "UDP (17)" 751 from [RFC0768]. The DF bit of the Flags field is set to 0 when 752 the method in Section 5.4.1 is used and set to 1 when the method 753 in Section 5.4.2 is used. 755 UDP Header: The UDP header contains a ITR selected source port when 756 encapsulating a packet. See Section 6.5 for details on the hash 757 algorithm used to select a source port based on the 5-tuple of the 758 inner header. The destination port MUST be set to the well-known 759 IANA assigned port value 4341. 761 UDP Checksum: The UDP checksum field SHOULD be transmitted as zero 762 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 763 [UDP-TUNNELS]. When a packet with a zero UDP checksum is received 764 by an ETR, the ETR MUST accept the packet for decapsulation. When 765 an ITR transmits a non-zero value for the UDP checksum, it MUST 766 send a correctly computed value in this field. When an ETR 767 receives a packet with a non-zero UDP checksum, it MAY choose to 768 verify the checksum value. If it chooses to perform such 769 verification, and the verification fails, the packet MUST be 770 silently dropped. If the ETR chooses not to perform the 771 verification, or performs the verification successfully, the 772 packet MUST be accepted for decapsulation. The handling of UDP 773 checksums for all tunneling protocols, including LISP, is under 774 active discussion within the IETF. When that discussion 775 concludes, any necessary changes will be made to align LISP with 776 the outcome of the broader discussion. 778 UDP Length: The UDP length field is set for an IPv4 encapsulated 779 packet to be the sum of the inner header IPv4 Total Length plus 780 the UDP and LISP header lengths. For an IPv6 encapsulated packet, 781 the UDP length field is the sum of the inner header IPv6 Payload 782 Length, the size of the IPv6 header (40 bytes), and the size of 783 the UDP and LISP headers. 785 N: The N bit is the nonce-present bit. When this bit is set to 1, 786 the low-order 24-bits of the first 32-bits of the LISP header 787 contains a Nonce. See Section 6.3.1 for details. Both N and V 788 bits MUST NOT be set in the same packet. If they are, a 789 decapsulating ETR MUST treat the "Nonce/Map-Version" field as 790 having a Nonce value present. 792 L: The L bit is the Locator Status Bits field enabled bit. When this 793 bit is set to 1, the Locator Status Bits in the second 32-bits of 794 the LISP header are in use. 796 x 1 x x 0 x x x 797 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 798 |N|L|E|V|I|flags| Nonce/Map-Version | 799 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 800 | Locator Status Bits | 801 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 803 E: The E bit is the echo-nonce-request bit. When this bit is set to 804 1, the N bit MUST be 1. This bit SHOULD be ignored and has no 805 meaning when the N bit is set to 0. See Section 6.3.1 for 806 details. 808 V: The V bit is the Map-Version present bit. When this bit is set to 809 1, the N bit MUST be 0. Refer to Section 6.6.3 for more details. 810 This bit indicates that the LISP header is encoded as: 812 0 x 0 1 x x x x 813 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 814 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version | 815 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 816 | Instance ID/Locator Status Bits | 817 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 819 I: The I bit is the Instance ID bit. See Section 5.5 for more 820 details. When this bit is set to 1, the Locator Status Bits field 821 is reduced to 8-bits and the high-order 24-bits are used as an 822 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 823 are transmitted as zero and ignored on receipt. The format of the 824 LISP header would look like: 826 x x x x 1 x x x 827 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 828 |N|L|E|V|I|flags| Nonce/Map-Version | 829 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 830 | Instance ID | LSBs | 831 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 833 flags: The flags field is a 3-bit field is reserved for future flag 834 use. It MUST be set to 0 on transmit and MUST be ignored on 835 receipt. 837 LISP Nonce: The LISP nonce field is a 24-bit value that is randomly 838 generated by an ITR when the N-bit is set to 1. Nonce generation 839 algorithms are an implementation matter but are required to 840 generate different nonces when sending to different destinations. 841 However, the same nonce can be used for a period of time to the 842 same destination. The nonce is also used when the E-bit is set to 843 request the nonce value to be echoed by the other side when 844 packets are returned. When the E-bit is clear but the N-bit is 845 set, a remote ITR is either echoing a previously requested echo- 846 nonce or providing a random nonce. See Section 6.3.1 for more 847 details. 849 LISP Locator Status Bits: The locator status bits field in the LISP 850 header is set by an ITR to indicate to an ETR the up/down status 851 of the Locators in the source site. Each RLOC in a Map-Reply is 852 assigned an ordinal value from 0 to n-1 (when there are n RLOCs in 853 a mapping entry). The Locator Status Bits are numbered from 0 to 854 n-1 from the least significant bit of field. The field is 32-bits 855 when the I-bit is set to 0 and is 8 bits when the I-bit is set to 856 1. When a Locator Status Bit is set to 1, the ITR is indicating 857 to the ETR the RLOC associated with the bit ordinal has up status. 858 See Section 6.3 for details on how an ITR can determine the status 859 of the ETRs at the same site. When a site has multiple EID- 860 prefixes which result in multiple mappings (where each could have 861 a different locator-set), the Locator Status Bits setting in an 862 encapsulated packet MUST reflect the mapping for the EID-prefix 863 that the inner-header source EID address matches. If the LSB for 864 an anycast locator is set to 1, then there is at least one RLOC 865 with that address the ETR is considered 'up'. 867 When doing ITR/PITR encapsulation: 869 o The outer header Time to Live field (or Hop Limit field, in case 870 of IPv6) SHOULD be copied from the inner header Time to Live 871 field. 873 o The outer header Type of Service field (or the Traffic Class 874 field, in the case of IPv6) SHOULD be copied from the inner header 875 Type of Service field (with one caveat, see below). 877 When doing ETR/PETR decapsulation: 879 o The inner header Time to Live field (or Hop Limit field, in case 880 of IPv6) SHOULD be copied from the outer header Time to Live 881 field, when the Time to Live field of the outer header is less 882 than the Time to Live of the inner header. Failing to perform 883 this check can cause the Time to Live of the inner header to 884 increment across encapsulation/decapsulation cycle. This check is 885 also performed when doing initial encapsulation when a packet 886 comes to an ITR or PITR destined for a LISP site. 888 o The inner header Type of Service field (or the Traffic Class 889 field, in the case of IPv6) SHOULD be copied from the outer header 890 Type of Service field (with one caveat, see below). 892 Note if an ETR/PETR is also an ITR/PITR and choose to reencapsulate 893 after decapsulating, the net effect of this is that the new outer 894 header will carry the same Time to Live as the old outer header minus 895 1. 897 Copying the TTL serves two purposes: first, it preserves the distance 898 the host intended the packet to travel; second, and more importantly, 899 it provides for suppression of looping packets in the event there is 900 a loop of concatenated tunnels due to misconfiguration. See 901 Section 9.3 for TTL exception handling for traceroute packets. 903 The ECN field occupies bits 6 and 7 of both the IPv4 Type of Service 904 field and the IPv6 Traffic Class field [RFC3168]. The ECN field 905 requires special treatment in order to avoid discarding indications 906 of congestion [RFC3168]. ITR encapsulation MUST copy the 2-bit ECN 907 field from the inner header to the outer header. Re-encapsulation 908 MUST copy the 2-bit ECN field from the stripped outer header to the 909 new outer header. If the ECN field contains a congestion indication 910 codepoint (the value is '11', the Congestion Experienced (CE) 911 codepoint), then ETR decapsulation MUST copy the 2-bit ECN field from 912 the stripped outer header to the surviving inner header that is used 913 to forward the packet beyond the ETR. These requirements preserve 914 Congestion Experienced (CE) indications when a packet that uses ECN 915 traverses a LISP tunnel and becomes marked with a CE indication due 916 to congestion between the tunnel endpoints. 918 5.4. Dealing with Large Encapsulated Packets 920 This section proposes two mechanisms to deal with packets that exceed 921 the path MTU between the ITR and ETR. 923 It is left to the implementor to decide if the stateless or stateful 924 mechanism should be implemented. Both or neither can be used since 925 it is a local decision in the ITR regarding how to deal with MTU 926 issues, and sites can interoperate with differing mechanisms. 928 Both stateless and stateful mechanisms also apply to Reencapsulating 929 and Recursive Tunneling. So any actions below referring to an ITR 930 also apply to an TE-ITR. 932 5.4.1. A Stateless Solution to MTU Handling 934 An ITR stateless solution to handle MTU issues is described as 935 follows: 937 1. Define an architectural constant S for the maximum size of a 938 packet, in bytes, an ITR would like to receive from a source 939 inside of its site. 941 2. Define L to be the maximum size, in bytes, a packet of size S 942 would be after the ITR prepends the LISP header, UDP header, and 943 outer network layer header of size H. 945 3. Calculate: S + H = L. 947 When an ITR receives a packet from a site-facing interface and adds H 948 bytes worth of encapsulation to yield a packet size greater than L 949 bytes, it resolves the MTU issue by first splitting the original 950 packet into 2 equal-sized fragments. A LISP header is then prepended 951 to each fragment. The size of the encapsulated fragments is then 952 (S/2 + H), which is less than the ITR's estimate of the path MTU 953 between the ITR and its correspondent ETR. 955 When an ETR receives encapsulated fragments, it treats them as two 956 individually encapsulated packets. It strips the LISP headers then 957 forwards each fragment to the destination host of the destination 958 site. The two fragments are reassembled at the destination host into 959 the single IP datagram that was originated by the source host. 961 This behavior is performed by the ITR when the source host originates 962 a packet with the DF field of the IP header is set to 0. When the DF 963 field of the IP header is set to 1, or the packet is an IPv6 packet 964 originated by the source host, the ITR will drop the packet when the 965 size is greater than L, and sends an ICMP Too Big message to the 966 source with a value of S, where S is (L - H). 968 When the outer header encapsulation uses an IPv4 header, an 969 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 970 can be avoided. An implementation MAY set the DF bit in such headers 971 to 0 if it has good reason to believe there are unresolvable path MTU 972 issues between the sending ITR and the receiving ETR. 974 This specification recommends that L be defined as 1500. 976 5.4.2. A Stateful Solution to MTU Handling 978 An ITR stateful solution to handle MTU issues is described as follows 979 and was first introduced in [OPENLISP]: 981 1. The ITR will keep state of the effective MTU for each locator per 982 mapping cache entry. The effective MTU is what the core network 983 can deliver along the path between ITR and ETR. 985 2. When an IPv6 encapsulated packet or an IPv4 encapsulated packet 986 with DF bit set to 1, exceeds what the core network can deliver, 987 one of the intermediate routers on the path will send an ICMP Too 988 Big message to the ITR. The ITR will parse the ICMP message to 989 determine which locator is affected by the effective MTU change 990 and then record the new effective MTU value in the mapping cache 991 entry. 993 3. When a packet is received by the ITR from a source inside of the 994 site and the size of the packet is greater than the effective MTU 995 stored with the mapping cache entry associated with the 996 destination EID the packet is for, the ITR will send an ICMP Too 997 Big message back to the source. The packet size advertised by 998 the ITR in the ICMP Too Big message is the effective MTU minus 999 the LISP encapsulation length. 1001 Even though this mechanism is stateful, it has advantages over the 1002 stateless IP fragmentation mechanism, by not involving the 1003 destination host with reassembly of ITR fragmented packets. 1005 5.5. Using Virtualization and Segmentation with LISP 1007 When multiple organizations inside of a LISP site are using private 1008 addresses [RFC1918] as EID-prefixes, their address spaces MUST remain 1009 segregated due to possible address duplication. An Instance ID in 1010 the address encoding can aid in making the entire AFI based address 1011 unique. See IANA Considerations Section 14.2 for details for 1012 possible address encodings. 1014 An Instance ID can be carried in a LISP encapsulated packet. An ITR 1015 that prepends a LISP header, will copy a 24-bit value, used by the 1016 LISP router to uniquely identify the address space. The value is 1017 copied to the Instance ID field of the LISP header and the I-bit is 1018 set to 1. 1020 When an ETR decapsulates a packet, the Instance ID from the LISP 1021 header is used as a table identifier to locate the forwarding table 1022 to use for the inner destination EID lookup. 1024 For example, a 802.1Q VLAN tag or VPN identifier could be used as a 1025 24-bit Instance ID. 1027 6. EID-to-RLOC Mapping 1029 6.1. LISP IPv4 and IPv6 Control Plane Packet Formats 1031 The following UDP packet formats are used by the LISP control-plane. 1033 0 1 2 3 1034 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 1035 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1036 |Version| IHL |Type of Service| Total Length | 1037 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1038 | Identification |Flags| Fragment Offset | 1039 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1040 | Time to Live | Protocol = 17 | Header Checksum | 1041 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1042 | Source Routing Locator | 1043 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1044 | Destination Routing Locator | 1045 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1046 / | Source Port | Dest Port | 1047 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1048 \ | UDP Length | UDP Checksum | 1049 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1050 | | 1051 | LISP Message | 1052 | | 1053 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1055 0 1 2 3 1056 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 1057 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1058 |Version| Traffic Class | Flow Label | 1059 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1060 | Payload Length | Next Header=17| Hop Limit | 1061 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1062 | | 1063 + + 1064 | | 1065 + Source Routing Locator + 1066 | | 1067 + + 1068 | | 1069 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1070 | | 1071 + + 1072 | | 1073 + Destination Routing Locator + 1074 | | 1075 + + 1076 | | 1077 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1078 / | Source Port | Dest Port | 1079 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1080 \ | UDP Length | UDP Checksum | 1081 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1082 | | 1083 | LISP Message | 1084 | | 1085 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1087 The LISP UDP-based messages are the Map-Request and Map-Reply 1088 messages. When a UDP Map-Request is sent, the UDP source port is 1089 chosen by the sender and the destination UDP port number is set to 1090 4342. When a UDP Map-Reply is sent, the source UDP port number is 1091 set to 4342 and the destination UDP port number is copied from the 1092 source port of either the Map-Request or the invoking data packet. 1093 Implementations MUST be prepared to accept packets when either the 1094 source port or destination UDP port is set to 4342 due to NATs 1095 changing port number values. 1097 The UDP Length field will reflect the length of the UDP header and 1098 the LISP Message payload. 1100 The UDP Checksum is computed and set to non-zero for Map-Request, 1101 Map-Reply, Map-Register and ECM control messages. It MUST be checked 1102 on receipt and if the checksum fails, the packet MUST be dropped. 1104 LISP-CONS [CONS] uses TCP to send LISP control messages. The format 1105 of control messages includes the UDP header so the checksum and 1106 length fields can be used to protect and delimit message boundaries. 1108 This main LISP specification is the authoritative source for message 1109 format definitions for the Map-Request and Map-Reply messages. 1111 6.1.1. LISP Packet Type Allocations 1113 This section will be the authoritative source for allocating LISP 1114 Type values. Current allocations are: 1116 Reserved: 0 b'0000' 1117 LISP Map-Request: 1 b'0001' 1118 LISP Map-Reply: 2 b'0010' 1119 LISP Map-Register: 3 b'0011' 1120 LISP Map-Notify: 4 b'0100' 1121 LISP Encapsulated Control Message: 8 b'1000' 1123 6.1.2. Map-Request Message Format 1125 0 1 2 3 1126 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 1127 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1128 |Type=1 |A|M|P|S|p|s| Reserved | IRC | Record Count | 1129 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1130 | Nonce . . . | 1131 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1132 | . . . Nonce | 1133 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1134 | Source-EID-AFI | Source EID Address ... | 1135 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1136 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... | 1137 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1138 | ... | 1139 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1140 | ITR-RLOC-AFI n | ITR-RLOC Address n ... | 1141 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1142 / | Reserved | EID mask-len | EID-prefix-AFI | 1143 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1144 \ | EID-prefix ... | 1145 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1146 | Map-Reply Record ... | 1147 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1148 | Mapping Protocol Data | 1149 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1151 Packet field descriptions: 1153 Type: 1 (Map-Request) 1155 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 1156 Requests sent by an ITR. 1158 M: This is the map-data-present bit, when set, it indicates a Map- 1159 Reply Record segment is included in the Map-Request. 1161 P: This is the probe-bit which indicates that a Map-Request SHOULD be 1162 treated as a locator reachability probe. The receiver SHOULD 1163 respond with a Map-Reply with the probe-bit set, indicating the 1164 Map-Reply is a locator reachability probe reply, with the nonce 1165 copied from the Map-Request. See Section 6.3.2 for more details. 1167 S: This is the Solicit-Map-Request (SMR) bit. See Section 6.6.2 for 1168 details. 1170 p: This is the PITR bit. This bit is set to 1 when a PITR sends a 1171 Map-Request. 1173 s: This is the SMR-invoked bit. This bit is set to 1 when an xTR is 1174 sending a Map-Request in response to a received SMR-based Map- 1175 Request. 1177 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1178 receipt. 1180 IRC: This 5-bit field is the ITR-RLOC Count which encodes the 1181 additional number of (ITR-RLOC-AFI, ITR-RLOC Address) fields 1182 present in this message. At least one (ITR-RLOC-AFI, ITR-RLOC- 1183 Address) pair must always be encoded. Multiple ITR-RLOC Address 1184 fields are used so a Map-Replier can select which destination 1185 address to use for a Map-Reply. The IRC value ranges from 0 to 1186 31. For a value of 0, there is 1 ITR-RLOC address encoded, and 1187 for a value of 1, there are 2 ITR-RLOC addresses encoded and so on 1188 up to 31 which encodes a total of 32 ITR-RLOC addresses. 1190 Record Count: The number of records in this Map-Request message. A 1191 record is comprised of the portion of the packet that is labeled 1192 'Rec' above and occurs the number of times equal to Record Count. 1193 For this version of the protocol, a receiver MUST accept and 1194 process Map-Requests that contain one or more records, but a 1195 sender MUST only send Map-Requests containing one record. Support 1196 for requesting multiple EIDs in a single Map-Request message will 1197 be specified in a future version of the protocol. 1199 Nonce: An 8-byte random value created by the sender of the Map- 1200 Request. This nonce will be returned in the Map-Reply. The 1201 security of the LISP mapping protocol depends critically on the 1202 strength of the nonce in the Map-Request message. The nonce 1203 SHOULD be generated by a properly seeded pseudo-random (or strong 1204 random) source. See [RFC4086] for advice on generating security- 1205 sensitive random data. 1207 Source-EID-AFI: Address family of the "Source EID Address" field. 1209 Source EID Address: This is the EID of the source host which 1210 originated the packet which is invoking this Map-Request. When 1211 Map-Requests are used for refreshing a map-cache entry or for 1212 RLOC-probing, an AFI value 0 is used and this field is of zero 1213 length. 1215 ITR-RLOC-AFI: Address family of the "ITR-RLOC Address" field that 1216 follows this field. 1218 ITR-RLOC Address: Used to give the ETR the option of selecting the 1219 destination address from any address family for the Map-Reply 1220 message. This address MUST be a routable RLOC address of the 1221 sender of the Map-Request message. 1223 EID mask-len: Mask length for EID prefix. 1225 EID-prefix-AFI: Address family of EID-prefix according to [AFI] 1227 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1228 address-family. When a Map-Request is sent by an ITR because a 1229 data packet is received for a destination where there is no 1230 mapping entry, the EID-prefix is set to the destination IP address 1231 of the data packet. And the 'EID mask-len' is set to 32 or 128 1232 for IPv4 or IPv6, respectively. When an xTR wants to query a site 1233 about the status of a mapping it already has cached, the EID- 1234 prefix used in the Map-Request has the same mask-length as the 1235 EID-prefix returned from the site when it sent a Map-Reply 1236 message. 1238 Map-Reply Record: When the M bit is set, this field is the size of a 1239 single "Record" in the Map-Reply format. This Map-Reply record 1240 contains the EID-to-RLOC mapping entry associated with the Source 1241 EID. This allows the ETR which will receive this Map-Request to 1242 cache the data if it chooses to do so. 1244 Mapping Protocol Data: See [CONS] for details. This field is 1245 optional and present when the UDP length indicates there is enough 1246 space in the packet to include it. 1248 6.1.3. EID-to-RLOC UDP Map-Request Message 1250 A Map-Request is sent from an ITR when it needs a mapping for an EID, 1251 wants to test an RLOC for reachability, or wants to refresh a mapping 1252 before TTL expiration. For the initial case, the destination IP 1253 address used for the Map-Request is the destination-EID from the 1254 packet which had a mapping cache lookup failure. For the latter two 1255 cases, the destination IP address used for the Map-Request is one of 1256 the RLOC addresses from the locator-set of the map cache entry. The 1257 source address is either an IPv4 or IPv6 RLOC address depending if 1258 the Map-Request is using an IPv4 versus IPv6 header, respectively. 1259 In all cases, the UDP source port number for the Map-Request message 1260 is an ITR/PITR selected 16-bit value and the UDP destination port 1261 number is set to the well-known destination port number 4342. A 1262 successful Map-Reply, which is one that has a nonce that matches an 1263 outstanding Map-Request nonce, will update the cached set of RLOCs 1264 associated with the EID prefix range. 1266 One or more Map-Request (ITR-RLOC-AFI, ITR-RLOC-Address) fields MUST 1267 be filled in by the ITR. The number of fields (minus 1) encoded MUST 1268 be placed in the IRC field. The ITR MAY include all locally 1269 configured locators in this list or just provide one locator address 1270 from each address family it supports. If the ITR erroneously 1271 provides no ITR-RLOC addresses, the Map-Replier MUST drop the Map- 1272 Request. 1274 Map-Requests can also be LISP encapsulated using UDP destination port 1275 4342 with a LISP type value set to "Encapsulated Control Message", 1276 when sent from an ITR to a Map-Resolver. Likewise, Map-Requests are 1277 LISP encapsulated the same way from a Map-Server to an ETR. Details 1278 on encapsulated Map-Requests and Map-Resolvers can be found in 1279 [LISP-MS]. 1281 Map-Requests MUST be rate-limited. It is recommended that a Map- 1282 Request for the same EID-prefix be sent no more than once per second. 1284 An ITR that is configured with mapping database information (i.e. it 1285 is also an ETR) may optionally include those mappings in a Map- 1286 Request. When an ETR configured to accept and verify such 1287 "piggybacked" mapping data receives such a Map-Request and it does 1288 not have this mapping in the map-cache, it may originate a "verifying 1289 Map-Request", addressed to the map-requesting ITR. If the ETR has a 1290 map-cache entry that matches the "piggybacked" EID and the RLOC is in 1291 the locator-set for the entry, then it may send the "verifying Map- 1292 Request" directly to the originating Map-Request source. If the RLOC 1293 is not in the locator-set, then the ETR MUST send the "verifying Map- 1294 Request" to the "piggybacked" EID. Doing this forces the "verifying 1295 Map-Request" to go through the mapping database system to reach the 1296 authoritative source of information about that EID, guarding against 1297 RLOC-spoofing in in the "piggybacked" mapping data. 1299 6.1.4. Map-Reply Message Format 1301 0 1 2 3 1302 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 1303 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1304 |Type=2 |P|E|S| Reserved | Record Count | 1305 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1306 | Nonce . . . | 1307 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1308 | . . . Nonce | 1309 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1310 | | Record TTL | 1311 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1312 R | Locator Count | EID mask-len | ACT |A| Reserved | 1313 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1314 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1315 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1316 r | EID-prefix | 1317 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1318 | /| Priority | Weight | M Priority | M Weight | 1319 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1320 | o | Unused Flags |L|p|R| Loc-AFI | 1321 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1322 | \| Locator | 1323 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1324 | Mapping Protocol Data | 1325 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1327 Packet field descriptions: 1329 Type: 2 (Map-Reply) 1331 P: This is the probe-bit which indicates that the Map-Reply is in 1332 response to a locator reachability probe Map-Request. The nonce 1333 field MUST contain a copy of the nonce value from the original 1334 Map-Request. See Section 6.3.2 for more details. 1336 E: Indicates that the ETR which sends this Map-Reply message is 1337 advertising that the site is enabled for the Echo-Nonce locator 1338 reachability algorithm. See Section 6.3.1 for more details. 1340 S: This is the Security bit. When set to 1 the field following the 1341 Mapping Protocol Data field will have the following format. The 1342 detailed format of the Authentication Data Content is for further 1343 study. 1345 0 1 2 3 1346 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 1347 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1348 | AD Type | Authentication Data Content . . . | 1349 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1351 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1352 receipt. 1354 Record Count: The number of records in this reply message. A record 1355 is comprised of that portion of the packet labeled 'Record' above 1356 and occurs the number of times equal to Record count. 1358 Nonce: A 24-bit value set in a Data-Probe packet or a 64-bit value 1359 from the Map-Request is echoed in this Nonce field of the Map- 1360 Reply. 1362 Record TTL: The time in minutes the recipient of the Map-Reply will 1363 store the mapping. If the TTL is 0, the entry SHOULD be removed 1364 from the cache immediately. If the value is 0xffffffff, the 1365 recipient can decide locally how long to store the mapping. 1367 Locator Count: The number of Locator entries. A locator entry 1368 comprises what is labeled above as 'Loc'. The locator count can 1369 be 0 indicating there are no locators for the EID-prefix. 1371 EID mask-len: Mask length for EID prefix. 1373 ACT: This 3-bit field describes negative Map-Reply actions. These 1374 bits are used only when the 'Locator Count' field is set to 0. 1375 The action bits are encoded only in Map-Reply messages. The 1376 actions defined are used by an ITR or PTR when a destination EID 1377 matches a negative mapping cache entry. Unassigned values should 1378 cause a map-cache entry to be created and, when packets match this 1379 negative cache entry, they will be dropped. The current assigned 1380 values are: 1382 (0) No-Action: The map-cache is kept alive and no packet 1383 encapsulation occurs. 1385 (1) Natively-Forward: The packet is not encapsulated or dropped 1386 but natively forwarded. 1388 (2) Send-Map-Request: The packet invokes sending a Map-Request. 1390 (3) Drop: A packet that matches this map-cache entry is dropped. 1391 An ICMP Unreachable message SHOULD be sent. 1393 A: The Authoritative bit, when sent by a UDP-based message is always 1394 set to 1 by an ETR. See [CONS] for TCP-based Map-Replies. When a 1395 Map-Server is proxy Map-Replying [LISP-MS] for a LISP site, the 1396 Authoritative bit is set to 0. This indicates to requesting ITRs 1397 that the Map-Reply was not originated by a LISP node managed at 1398 the site that owns the EID-prefix. 1400 Map-Version Number: When this 12-bit value is non-zero the Map-Reply 1401 sender is informing the ITR what the version number is for the 1402 EID-record contained in the Map-Reply. The ETR can allocate this 1403 number internally but MUST coordinate this value with other ETRs 1404 for the site. When this value is 0, there is no versioning 1405 information conveyed. The Map-Version Number can be included in 1406 Map-Request and Map-Register messages. See Section 6.6.3 for more 1407 details. 1409 EID-prefix-AFI: Address family of EID-prefix according to [AFI]. 1411 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1412 address-family. 1414 Priority: each RLOC is assigned a unicast priority. Lower values 1415 are more preferable. When multiple RLOCs have the same priority, 1416 they may be used in a load-split fashion. A value of 255 means 1417 the RLOC MUST NOT be used for unicast forwarding. 1419 Weight: when priorities are the same for multiple RLOCs, the weight 1420 indicates how to balance unicast traffic between them. Weight is 1421 encoded as a relative weight of total unicast packets that match 1422 the mapping entry. For example if there are 4 locators in a 1423 locator set, where the weights assigned are 30, 20, 20, and 10, 1424 the first locator will get 37.5% of the traffic, the 2nd and 3rd 1425 locators will get 25% of traffic and the 4th locator will get 1426 12.5% of the traffic. If all weights for a locator-set are equal, 1427 receiver of the Map-Reply will decide how to load-split traffic. 1429 See Section 6.5 for a suggested hash algorithm to distribute load 1430 across locators with same priority and equal weight values. 1432 M Priority: each RLOC is assigned a multicast priority used by an 1433 ETR in a receiver multicast site to select an ITR in a source 1434 multicast site for building multicast distribution trees. A value 1435 of 255 means the RLOC MUST NOT be used for joining a multicast 1436 distribution tree. 1438 M Weight: when priorities are the same for multiple RLOCs, the 1439 weight indicates how to balance building multicast distribution 1440 trees across multiple ITRs. The weight is encoded as a relative 1441 weight (similar to the unicast Weights) of total number of trees 1442 built to the source site identified by the EID-prefix. If all 1443 weights for a locator-set are equal, the receiver of the Map-Reply 1444 will decide how to distribute multicast state across ITRs. 1446 Unused Flags: set to 0 when sending and ignored on receipt. 1448 L: when this bit is set, the locator is flagged as a local locator to 1449 the ETR that is sending the Map-Reply. When a Map-Server is doing 1450 proxy Map-Replying [LISP-MS] for a LISP site, the L bit is set to 1451 0 for all locators in this locator-set. 1453 p: when this bit is set, an ETR informs the RLOC-probing ITR that the 1454 locator address, for which this bit is set, is the one being RLOC- 1455 probed and may be different from the source address of the Map- 1456 Reply. An ITR that RLOC-probes a particular locator, MUST use 1457 this locator for retrieving the data structure used to store the 1458 fact that the locator is reachable. The "p" bit is set for a 1459 single locator in the same locator set. If an implementation sets 1460 more than one "p" bit erroneously, the receiver of the Map-Reply 1461 MUST select the first locator. The "p" bit MUST NOT be set for 1462 locator-set records sent in Map-Request and Map-Register messages. 1464 R: set when the sender of a Map-Reply has a route to the locator in 1465 the locator data record. This receiver may find this useful to 1466 know if the locator is up but not necessarily reachable from the 1467 receiver's point of view. See also Section 6.4 for another way 1468 the R-bit may be used. 1470 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1471 assigned to an ETR. Note that the destination RLOC address MAY be 1472 an anycast address. A source RLOC can be an anycast address as 1473 well. The source or destination RLOC MUST NOT be the broadcast 1474 address (255.255.255.255 or any subnet broadcast address known to 1475 the router), and MUST NOT be a link-local multicast address. The 1476 source RLOC MUST NOT be a multicast address. The destination RLOC 1477 SHOULD be a multicast address if it is being mapped from a 1478 multicast destination EID. 1480 Mapping Protocol Data: See [CONS] for details. This field is 1481 optional and present when the UDP length indicates there is enough 1482 space in the packet to include it. The Mapping Protocol Data is 1483 used when needed by the particular mapping system. 1485 6.1.5. EID-to-RLOC UDP Map-Reply Message 1487 A Map-Reply returns an EID-prefix with a prefix length that is less 1488 than or equal to the EID being requested. The EID being requested is 1489 either from the destination field of an IP header of a Data-Probe or 1490 the EID record of a Map-Request. The RLOCs in the Map-Reply are 1491 globally-routable IP addresses of all ETRs for the LISP site. Each 1492 RLOC conveys status reachability but does not convey path 1493 reachability from a requesters perspective. Separate testing of path 1494 reachability is required, See Section 6.3 for details. 1496 Note that a Map-Reply may contain different EID-prefix granularity 1497 (prefix + length) than the Map-Request which triggers it. This might 1498 occur if a Map-Request were for a prefix that had been returned by an 1499 earlier Map-Reply. In such a case, the requester updates its cache 1500 with the new prefix information and granularity. For example, a 1501 requester with two cached EID-prefixes that are covered by a Map- 1502 Reply containing one, less-specific prefix, replaces the entry with 1503 the less-specific EID-prefix. Note that the reverse, replacement of 1504 one less-specific prefix with multiple more-specific prefixes, can 1505 also occur but not by removing the less-specific prefix rather by 1506 adding the more-specific prefixes which during a lookup will override 1507 the less-specific prefix. 1509 When an ETR is configured with overlapping EID-prefixes, a Map- 1510 Request with an EID that longest matches any EID-prefix MUST be 1511 returned in a single Map-Reply message. For instance, if an ETR had 1512 database mapping entries for EID-prefixes: 1514 10.0.0.0/8 1515 10.1.0.0/16 1516 10.1.1.0/24 1517 10.1.2.0/24 1519 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record 1520 count of 1 to be returned with a mapping record EID-prefix of 1521 10.1.1.0/24. 1523 A Map-Request for EID 10.1.5.5, would cause a Map-Reply with a record 1524 count of 3 to be returned with mapping records for EID-prefixes 1525 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24. 1527 Note that not all overlapping EID-prefixes need to be returned, only 1528 the more specifics (note in the second example above 10.0.0.0/8 was 1529 not returned for requesting EID 10.1.5.5) entries for the matching 1530 EID-prefix of the requesting EID. When more than one EID-prefix is 1531 returned, all SHOULD use the same Time-to-Live value so they can all 1532 time out at the same time. When a more specific EID-prefix is 1533 received later, its Time-to-Live value in the Map-Reply record can be 1534 stored even when other less specifics exist. When a less specific 1535 EID-prefix is received later, its map-cache expiration time SHOULD be 1536 set to the minimum expiration time of any more specific EID-prefix in 1537 the map-cache. This is done so the integrity of the EID-prefix set 1538 is wholly maintained so no more-specific entries are removed from the 1539 map-cache while keeping less-specific entries. 1541 Map-Replies SHOULD be sent for an EID-prefix no more often than once 1542 per second to the same requesting router. For scalability, it is 1543 expected that aggregation of EID addresses into EID-prefixes will 1544 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1545 prefix range thereby reducing the number of Map-Request messages. 1547 Map-Reply records can have an empty locator-set. A negative Map- 1548 Reply is a Map-Reply with an empty locator-set. Negative Map-Replies 1549 convey special actions by the sender to the ITR or PTR which have 1550 solicited the Map-Reply. There are two primary applications for 1551 Negative Map-Replies. The first is for a Map-Resolver to instruct an 1552 ITR or PTR when a destination is for a LISP site versus a non-LISP 1553 site. And the other is to source quench Map-Requests which are sent 1554 for non-allocated EIDs. 1556 For each Map-Reply record, the list of locators in a locator-set MUST 1557 appear in the same order for each ETR that originates a Map-Reply 1558 message. The locator-set MUST be sorted in order of ascending IP 1559 address where an IPv4 locator address is considered numerically 'less 1560 than' an IPv6 locator address. 1562 When sending a Map-Reply message, the destination address is copied 1563 from the one of the ITR-RLOC fields from the Map-Request. The ETR 1564 can choose a locator address from one of the address families it 1565 supports. For Data-Probes, the destination address of the Map-Reply 1566 is copied from the source address of the Data-Probe message which is 1567 invoking the reply. The source address of the Map-Reply is one of 1568 the local IP addresses chosen to allow uRPF checks to succeed in the 1569 upstream service provider. The destination port of a Map-Reply 1570 message is copied from the source port of the Map-Request or Data- 1571 Probe and the source port of the Map-Reply message is set to the 1572 well-known UDP port 4342. 1574 6.1.5.1. Traffic Redirection with Coarse EID-Prefixes 1576 When an ETR is misconfigured or compromised, it could return coarse 1577 EID-prefixes in Map-Reply messages it sends. The EID-prefix could 1578 cover EID-prefixes which are allocated to other sites redirecting 1579 their traffic to the locators of the compromised site. 1581 To solve this problem, there are two basic solutions that could be 1582 used. The first is to have Map-Servers proxy-map-reply on behalf of 1583 ETRs so their registered EID-prefixes are the ones returned in Map- 1584 Replies. Since the interaction between an ETR and Map-Server is 1585 secured with shared-keys, it is more difficult for an ETR to 1586 misbehave. The second solution is to have ITRs and PTRs cache EID- 1587 prefixes with mask-lengths that are greater than or equal to a 1588 configured prefix length. This limits the damage to a specific width 1589 of any EID-prefix advertised, but needs to be coordinated with the 1590 allocation of site prefixes. These solutions can be used 1591 independently or at the same time. 1593 At the time of this writing, other approaches are being considered 1594 and researched. 1596 6.1.6. Map-Register Message Format 1598 The usage details of the Map-Register message can be found in 1599 specification [LISP-MS]. This section solely defines the message 1600 format. 1602 The message is sent in UDP with a destination UDP port of 4342 and a 1603 randomly selected UDP source port number. 1605 The Map-Register message format is: 1607 0 1 2 3 1608 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 1609 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1610 |Type=3 |P| Reserved |M| Record Count | 1611 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1612 | Nonce . . . | 1613 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1614 | . . . Nonce | 1615 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1616 | Key ID | Authentication Data Length | 1617 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1618 ~ Authentication Data ~ 1619 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1620 | | Record TTL | 1621 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1622 R | Locator Count | EID mask-len | ACT |A| Reserved | 1623 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1624 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1625 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1626 r | EID-prefix | 1627 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1628 | /| Priority | Weight | M Priority | M Weight | 1629 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1630 | o | Unused Flags |L|p|R| Loc-AFI | 1631 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1632 | \| Locator | 1633 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1635 Packet field descriptions: 1637 Type: 3 (Map-Register) 1639 P: This is the proxy-map-reply bit, when set to 1 an ETR sends a Map- 1640 Register message requesting for the Map-Server to proxy Map-Reply. 1641 The Map-Server will send non-authoritative Map-Replies on behalf 1642 of the ETR. Details on this usage will be provided in a future 1643 version of this draft. 1645 Reserved: It MUST be set to 0 on transmit and MUST be ignored on 1646 receipt. 1648 M: This is the want-map-notify bit, when set to 1 an ETR is 1649 requesting for a Map-Notify message to be returned in response to 1650 sending a Map-Register message. The Map-Notify message sent by a 1651 Map-Server is used to an acknowledge receipt of a Map-Register 1652 message. 1654 Record Count: The number of records in this Map-Register message. A 1655 record is comprised of that portion of the packet labeled 'Record' 1656 above and occurs the number of times equal to Record count. 1658 Nonce: This 8-byte Nonce field is set to 0 in Map-Register messages. 1660 Key ID: A configured ID to find the configured Message 1661 Authentication Code (MAC) algorithm and key value used for the 1662 authentication function. See Section 14.4 for codepoint 1663 assignments. 1665 Authentication Data Length: The length in bytes of the 1666 Authentication Data field that follows this field. The length of 1667 the Authentication Data field is dependent on the Message 1668 Authentication Code (MAC) algorithm used. The length field allows 1669 a device that doesn't know the MAC algorithm to correctly parse 1670 the packet. 1672 Authentication Data: The message digest used from the output of the 1673 Message Authentication Code (MAC) algorithm. The entire Map- 1674 Register payload is authenticated with this field preset to 0. 1675 After the MAC is computed, it is placed in this field. 1676 Implementations of this specification MUST include support for 1677 HMAC-SHA-1-96 [RFC2404] and support for HMAC-SHA-256-128 [RFC6234] 1678 is recommended. 1680 The definition of the rest of the Map-Register can be found in the 1681 Map-Reply section. 1683 6.1.7. Map-Notify Message Format 1685 The usage details of the Map-Notify message can be found in 1686 specification [LISP-MS]. This section solely defines the message 1687 format. 1689 The message is sent inside a UDP packet with a source UDP port equal 1690 to 4342 and a destination port equal to the source port from the Map- 1691 Register message this Map-Notify message is responding to. 1693 The Map-Notify message format is: 1695 0 1 2 3 1696 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 1697 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1698 |Type=4 | Reserved | Record Count | 1699 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1700 | Nonce . . . | 1701 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1702 | . . . Nonce | 1703 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1704 | Key ID | Authentication Data Length | 1705 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1706 ~ Authentication Data ~ 1707 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1708 | | Record TTL | 1709 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1710 R | Locator Count | EID mask-len | ACT |A| Reserved | 1711 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1712 c | Rsvd | Map-Version Number | EID-prefix-AFI | 1713 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1714 r | EID-prefix | 1715 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1716 | /| Priority | Weight | M Priority | M Weight | 1717 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1718 | o | Unused Flags |L|p|R| Loc-AFI | 1719 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1720 | \| Locator | 1721 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1723 Packet field descriptions: 1725 Type: 4 (Map-Notify) 1727 The Map-Notify message has the same contents as a Map-Register 1728 message. See Map-Register section for field descriptions. 1730 6.1.8. Encapsulated Control Message Format 1732 An Encapsulated Control Message (ECM) is used to encapsulate control 1733 packets sent between xTRs and the mapping database system described 1734 in [LISP-MS]. 1736 0 1 2 3 1737 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 1738 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1739 / | IPv4 or IPv6 Header | 1740 OH | (uses RLOC addresses) | 1741 \ | | 1742 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1743 / | Source Port = xxxx | Dest Port = 4342 | 1744 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1745 \ | UDP Length | UDP Checksum | 1746 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1747 LH |Type=8 |S| Reserved | 1748 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1749 / | IPv4 or IPv6 Header | 1750 IH | (uses RLOC or EID addresses) | 1751 \ | | 1752 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1753 / | Source Port = xxxx | Dest Port = yyyy | 1754 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1755 \ | UDP Length | UDP Checksum | 1756 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1757 LCM | LISP Control Message | 1758 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1760 Packet header descriptions: 1762 OH: The outer IPv4 or IPv6 header which uses RLOC addresses in the 1763 source and destination header address fields. 1765 UDP: The outer UDP header with destination port 4342. The source 1766 port is randomly allocated. The checksum field MUST be non-zero. 1768 LH: Type 8 is defined to be a "LISP Encapsulated Control Message" 1769 and what follows is either an IPv4 or IPv6 header as encoded by 1770 the first 4 bits after the reserved field. 1772 S: This is the Security bit. When set to 1 the field following the 1773 Reserved field will have the following format. The detailed 1774 format of the Authentication Data Content is for further study. 1776 0 1 2 3 1777 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 1778 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1779 | AD Type | Authentication Data Content . . . | 1780 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1782 IH: The inner IPv4 or IPv6 header which can use either RLOC or EID 1783 addresses in the header address fields. When a Map-Request is 1784 encapsulated in this packet format the destination address in this 1785 header is an EID. 1787 UDP: The inner UDP header where the port assignments depends on the 1788 control packet being encapsulated. When the control packet is a 1789 Map-Request or Map-Register, the source port is ITR/PITR selected 1790 and the destination port is 4342. When the control packet is a 1791 Map-Reply, the source port is 4342 and the destination port is 1792 assigned from the source port of the invoking Map-Request. Port 1793 number 4341 MUST NOT be assigned to either port. The checksum 1794 field MUST be non-zero. 1796 LCM: The format is one of the control message formats described in 1797 this section. At this time, only Map-Request messages and PIM 1798 Join-Prune messages [MLISP] are allowed to be encapsulated. 1799 Encapsulating other types of LISP control messages are for further 1800 study. When Map-Requests are sent for RLOC-probing purposes (i.e 1801 the probe-bit is set), they MUST NOT be sent inside Encapsulated 1802 Control Messages. 1804 6.2. Routing Locator Selection 1806 Both client-side and server-side may need control over the selection 1807 of RLOCs for conversations between them. This control is achieved by 1808 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1809 messages. Alternatively, RLOC information may be gleaned from 1810 received tunneled packets or EID-to-RLOC Map-Request messages. 1812 The following enumerates different scenarios for choosing RLOCs and 1813 the controls that are available: 1815 o Server-side returns one RLOC. Client-side can only use one RLOC. 1816 Server-side has complete control of the selection. 1818 o Server-side returns a list of RLOC where a subset of the list has 1819 the same best priority. Client can only use the subset list 1820 according to the weighting assigned by the server-side. In this 1821 case, the server-side controls both the subset list and load- 1822 splitting across its members. The client-side can use RLOCs 1823 outside of the subset list if it determines that the subset list 1824 is unreachable (unless RLOCs are set to a Priority of 255). Some 1825 sharing of control exists: the server-side determines the 1826 destination RLOC list and load distribution while the client-side 1827 has the option of using alternatives to this list if RLOCs in the 1828 list are unreachable. 1830 o Server-side sets weight of 0 for the RLOC subset list. In this 1831 case, the client-side can choose how the traffic load is spread 1832 across the subset list. Control is shared by the server-side 1833 determining the list and the client determining load distribution. 1834 Again, the client can use alternative RLOCs if the server-provided 1835 list of RLOCs are unreachable. 1837 o Either side (more likely on the server-side ETR) decides not to 1838 send a Map-Request. For example, if the server-side ETR does not 1839 send Map-Requests, it gleans RLOCs from the client-side ITR, 1840 giving the client-side ITR responsibility for bidirectional RLOC 1841 reachability and preferability. Server-side ETR gleaning of the 1842 client-side ITR RLOC is done by caching the inner header source 1843 EID and the outer header source RLOC of received packets. The 1844 client-side ITR controls how traffic is returned and can alternate 1845 using an outer header source RLOC, which then can be added to the 1846 list the server-side ETR uses to return traffic. Since no 1847 Priority or Weights are provided using this method, the server- 1848 side ETR MUST assume each client-side ITR RLOC uses the same best 1849 Priority with a Weight of zero. In addition, since EID-prefix 1850 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1851 on tunnel routers can grow to be very large. 1853 o A "gleaned" map-cache entry, one learned from the source RLOC of a 1854 received encapsulated packet, is only stored and used for a few 1855 seconds, pending verification. Verification is performed by 1856 sending a Map-Request to the source EID (the inner header IP 1857 source address) of the received encapsulated packet. A reply to 1858 this "verifying Map-Request" is used to fully populate the map- 1859 cache entry for the "gleaned" EID and is stored and used for the 1860 time indicated from the TTL field of a received Map-Reply. When a 1861 verified map-cache entry is stored, data gleaning no longer occurs 1862 for subsequent packets which have a source EID that matches the 1863 EID-prefix of the verified entry. 1865 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1866 reachable when the R-bit for the locator record is set to 1. When 1867 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1868 RLOC. Neither the information contained in a Map-Reply or that 1869 stored in the mapping database system provides reachability 1870 information for RLOCs. Note that reachability is not part of the 1871 mapping system and is determined using one or more of the Routing 1872 Locator Reachability Algorithms described in the next section. 1874 6.3. Routing Locator Reachability 1876 Several mechanisms for determining RLOC reachability are currently 1877 defined: 1879 1. An ETR may examine the Locator Status Bits in the LISP header of 1880 an encapsulated data packet received from an ITR. If the ETR is 1881 also acting as an ITR and has traffic to return to the original 1882 ITR site, it can use this status information to help select an 1883 RLOC. 1885 2. An ITR may receive an ICMP Network or ICMP Host Unreachable 1886 message for an RLOC it is using. This indicates that the RLOC is 1887 likely down. Note, trusting ICMP messages may not be desirable 1888 but neither is ignoring them completely. Implementations are 1889 encouraged to follow current best practices in treating these 1890 conditions. 1892 3. An ITR which participates in the global routing system can 1893 determine that an RLOC is down if no BGP RIB route exists that 1894 matches the RLOC IP address. 1896 4. An ITR may receive an ICMP Port Unreachable message from a 1897 destination host. This occurs if an ITR attempts to use 1898 interworking [INTERWORK] and LISP-encapsulated data is sent to a 1899 non-LISP-capable site. 1901 5. An ITR may receive a Map-Reply from a ETR in response to a 1902 previously sent Map-Request. The RLOC source of the Map-Reply is 1903 likely up since the ETR was able to send the Map-Reply to the 1904 ITR. 1906 6. When an ETR receives an encapsulated packet from an ITR, the 1907 source RLOC from the outer header of the packet is likely up. 1909 7. An ITR/ETR pair can use the Locator Reachability Algorithms 1910 described in this section, namely Echo-Noncing or RLOC-Probing. 1912 When determining Locator up/down reachability by examining the 1913 Locator Status Bits from the LISP encapsulated data packet, an ETR 1914 will receive up to date status from an encapsulating ITR about 1915 reachability for all ETRs at the site. CE-based ITRs at the source 1916 site can determine reachability relative to each other using the site 1917 IGP as follows: 1919 o Under normal circumstances, each ITR will advertise a default 1920 route into the site IGP. 1922 o If an ITR fails or if the upstream link to its PE fails, its 1923 default route will either time-out or be withdrawn. 1925 Each ITR can thus observe the presence or lack of a default route 1926 originated by the others to determine the Locator Status Bits it sets 1927 for them. 1929 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1930 Locator Status Bits in a LISP encapsulated packet are numbered from 0 1931 to n-1 starting with the least significant bit. For example, if an 1932 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1933 value 2), then all ITRs at the site will clear the 3rd least 1934 significant bit (xxxx x0xx) of the Locator Status Bits field for the 1935 packets they encapsulate. 1937 When an ETR decapsulates a packet, it will check for any change in 1938 the Locator Status Bits field. When a bit goes from 1 to 0, the ETR 1939 if acting also as an ITR, will refrain from encapsulating packets to 1940 an RLOC that is indicated as down. It will only resume using that 1941 RLOC if the corresponding Locator Status Bit returns to a value of 1. 1942 Locator Status Bits are associated with a locator-set per EID-prefix. 1943 Therefore, when a locator becomes unreachable, the Locator Status Bit 1944 that corresponds to that locator's position in the list returned by 1945 the last Map-Reply will be set to zero for that particular EID- 1946 prefix. 1948 When ITRs at the site are not deployed in CE routers, the IGP can 1949 still be used to determine the reachability of Locators provided they 1950 are injected into the IGP. This is typically done when a /32 address 1951 is configured on a loopback interface. 1953 When ITRs receive ICMP Network or Host Unreachable messages as a 1954 method to determine unreachability, they will refrain from using 1955 Locators which are described in Locator lists of Map-Replies. 1956 However, using this approach is unreliable because many network 1957 operators turn off generation of ICMP Unreachable messages. 1959 If an ITR does receive an ICMP Network or Host Unreachable message, 1960 it MAY originate its own ICMP Unreachable message destined for the 1961 host that originated the data packet the ITR encapsulated. 1963 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1964 locator address from a locator-set in a mapping entry matches a 1965 prefix. If it does not find one and BGP is running in the Default 1966 Free Zone (DFZ), it can decide to not use the locator even though the 1967 Locator Status Bits indicate the locator is up. In this case, the 1968 path from the ITR to the ETR that is assigned the locator is not 1969 available. More details are in [LOC-ID-ARCH]. 1971 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 1972 Reply is returned, reachability of the Locator has been determined. 1973 Obviously, sending such probes increases the number of control 1974 messages originated by tunnel routers for active flows, so Locators 1975 are assumed to be reachable when they are advertised. 1977 This assumption does create a dependency: Locator unreachability is 1978 detected by the receipt of ICMP Host Unreachable messages. When an 1979 Locator has been determined to be unreachable, it is not used for 1980 active traffic; this is the same as if it were listed in a Map-Reply 1981 with priority 255. 1983 The ITR can test the reachability of the unreachable Locator by 1984 sending periodic Requests. Both Requests and Replies MUST be rate- 1985 limited. Locator reachability testing is never done with data 1986 packets since that increases the risk of packet loss for end-to-end 1987 sessions. 1989 When an ETR decapsulates a packet, it knows that it is reachable from 1990 the encapsulating ITR because that is how the packet arrived. In 1991 most cases, the ETR can also reach the ITR but cannot assume this to 1992 be true due to the possibility of path asymmetry. In the presence of 1993 unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD NOT 1994 use the lack of return traffic as an indication that the ETR is 1995 unreachable. Instead, it MUST use an alternate mechanisms to 1996 determine reachability. 1998 6.3.1. Echo Nonce Algorithm 2000 When data flows bidirectionally between locators from different 2001 sites, a data-plane mechanism called "nonce echoing" can be used to 2002 determine reachability between an ITR and ETR. When an ITR wants to 2003 solicit a nonce echo, it sets the N and E bits and places a 24-bit 2004 nonce in the LISP header of the next encapsulated data packet. 2006 When this packet is received by the ETR, the encapsulated packet is 2007 forwarded as normal. When the ETR next sends a data packet to the 2008 ITR, it includes the nonce received earlier with the N bit set and E 2009 bit cleared. The ITR sees this "echoed nonce" and knows the path to 2010 and from the ETR is up. 2012 The ITR will set the E-bit and N-bit for every packet it sends while 2013 in echo-nonce-request state. The time the ITR waits to process the 2014 echoed nonce before it determines the path is unreachable is variable 2015 and a choice left for the implementation. 2017 If the ITR is receiving packets from the ETR but does not see the 2018 nonce echoed while being in echo-nonce-request state, then the path 2019 to the ETR is unreachable. This decision may be overridden by other 2020 locator reachability algorithms. Once the ITR determines the path to 2021 the ETR is down it can switch to another locator for that EID-prefix. 2023 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 2024 be implementing both ITR and ETR functionality for the echo nonce 2025 mechanism to operate. 2027 The ITR and ETR may both go into echo-nonce-request state at the same 2028 time. The number of packets sent or the time during which echo nonce 2029 requests are sent is an implementation specific setting. However, 2030 when an ITR is in echo-nonce-request state, it can echo the ETR's 2031 nonce in the next set of packets that it encapsulates and then 2032 subsequently, continue sending echo-nonce-request packets. 2034 This mechanism does not completely solve the forward path 2035 reachability problem as traffic may be unidirectional. That is, the 2036 ETR receiving traffic at a site may not be the same device as an ITR 2037 which transmits traffic from that site or the site to site traffic is 2038 unidirectional so there is no ITR returning traffic. 2040 The echo-nonce algorithm is bilateral. That is, if one side sets the 2041 E-bit and the other side is not enabled for echo-noncing, then the 2042 echoing of the nonce does not occur and the requesting side may 2043 regard the locator unreachable erroneously. An ITR SHOULD only set 2044 the E-bit in a encapsulated data packet when it knows the ETR is 2045 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 2046 Reply message. 2048 Note that other locator reachability mechanisms are being researched 2049 and can be used to compliment or even override the Echo Nonce 2050 Algorithm. See next section for an example of control-plane probing. 2052 6.3.2. RLOC Probing Algorithm 2054 RLOC Probing is a method that an ITR or PTR can use to determine the 2055 reachability status of one or more locators that it has cached in a 2056 map-cache entry. The probe-bit of the Map-Request and Map-Reply 2057 messages are used for RLOC Probing. 2059 RLOC probing is done in the control-plane on a timer basis where an 2060 ITR or PTR will originate a Map-Request destined to a locator address 2061 from one of its own locator addresses. A Map-Request used as an 2062 RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or on the 2063 ALT like one would when soliciting mapping data. The EID record 2064 encoded in the Map-Request is the EID-prefix of the map-cache entry 2065 cached by the ITR or PTR. The ITR may include a mapping data record 2066 for its own database mapping information which contains the local 2067 EID-prefixes and RLOCs for its site. 2069 When an ETR receives a Map-Request message with the probe-bit set, it 2070 returns a Map-Reply with the probe-bit set. The source address of 2071 the Map-Reply is set according to the procedure described in 2072 Section 6.1.5. The Map-Reply SHOULD contain mapping data for the 2073 EID-prefix contained in the Map-Request. This provides the 2074 opportunity for the ITR or PTR, which sent the RLOC-probe to get 2075 mapping updates if there were changes to the ETR's database mapping 2076 entries. 2078 There are advantages and disadvantages of RLOC Probing. The greatest 2079 benefit of RLOC Probing is that it can handle many failure scenarios 2080 allowing the ITR to determine when the path to a specific locator is 2081 reachable or has become unreachable, thus providing a robust 2082 mechanism for switching to using another locator from the cached 2083 locator. RLOC Probing can also provide rough RTT estimates between a 2084 pair of locators which can be useful for network management purposes 2085 as well as for selecting low delay paths. The major disadvantage of 2086 RLOC Probing is in the number of control messages required and the 2087 amount of bandwidth used to obtain those benefits, especially if the 2088 requirement for failure detection times are very small. 2090 Continued research and testing will attempt to characterize the 2091 tradeoffs of failure detection times versus message overhead. 2093 6.4. EID Reachability within a LISP Site 2095 A site may be multihomed using two or more ETRs. The hosts and 2096 infrastructure within a site will be addressed using one or more EID 2097 prefixes that are mapped to the RLOCs of the relevant ETRs in the 2098 mapping system. One possible failure mode is for an ETR to lose 2099 reachability to one or more of the EID prefixes within its own site. 2100 When this occurs when the ETR sends Map-Replies, it can clear the 2101 R-bit associated with its own locator. And when the ETR is also an 2102 ITR, it can clear its locator-status-bit in the encapsulation data 2103 header. 2105 It is recognized there are no simple solutions to the site 2106 partitioning problem because it is hard to know which part of the 2107 EID-prefix range is partitioned. And which locators can reach any 2108 sub-ranges of the EID-prefixes. This problem is under investigation 2109 with the expectation that experiments will tell us more. Note, this 2110 is not a new problem introduced by the LISP architecture. The 2111 problem exists today when a multi-homed site uses BGP to advertise 2112 its reachability upstream. 2114 6.5. Routing Locator Hashing 2116 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 2117 a requesting ITR, the locator-set for the EID-prefix may contain 2118 different priority values for each locator address. When more than 2119 one best priority locator exists, the ITR can decide how to load 2120 share traffic against the corresponding locators. 2122 The following hash algorithm may be used by an ITR to select a 2123 locator for a packet destined to an EID for the EID-to-RLOC mapping: 2125 1. Either a source and destination address hash can be used or the 2126 traditional 5-tuple hash which includes the source and 2127 destination addresses, source and destination TCP, UDP, or SCTP 2128 port numbers and the IP protocol number field or IPv6 next- 2129 protocol fields of a packet a host originates from within a LISP 2130 site. When a packet is not a TCP, UDP, or SCTP packet, the 2131 source and destination addresses only from the header are used to 2132 compute the hash. 2134 2. Take the hash value and divide it by the number of locators 2135 stored in the locator-set for the EID-to-RLOC mapping. 2137 3. The remainder will be yield a value of 0 to "number of locators 2138 minus 1". Use the remainder to select the locator in the 2139 locator-set. 2141 Note that when a packet is LISP encapsulated, the source port number 2142 in the outer UDP header needs to be set. Selecting a hashed value 2143 allows core routers which are attached to Link Aggregation Groups 2144 (LAGs) to load-split the encapsulated packets across member links of 2145 such LAGs. Otherwise, core routers would see a single flow, since 2146 packets have a source address of the ITR, for packets which are 2147 originated by different EIDs at the source site. A suggested setting 2148 for the source port number computed by an ITR is a 5-tuple hash 2149 function on the inner header, as described above. 2151 Many core router implementations use a 5-tuple hash to decide how to 2152 balance packet load across members of a LAG. The 5-tuple hash 2153 includes the source and destination addresses of the packet and the 2154 source and destination ports when the protocol number in the packet 2155 is TCP or UDP. For this reason, UDP encoding is used for LISP 2156 encapsulation. 2158 6.6. Changing the Contents of EID-to-RLOC Mappings 2160 Since the LISP architecture uses a caching scheme to retrieve and 2161 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 2162 date mapping is to re-request the mapping. However, the ITRs do not 2163 know when the mappings change and the ETRs do not keep track of which 2164 ITRs requested its mappings. For scalability reasons, we want to 2165 maintain this approach but need to provide a way for ETRs change 2166 their mappings and inform the sites that are currently communicating 2167 with the ETR site using such mappings. 2169 When adding a new locator record in lexicographic order to the end of 2170 a locator-set, it is easy to update mappings. We assume new mappings 2171 will maintain the same locator ordering as the old mapping but just 2172 have new locators appended to the end of the list. So some ITRs can 2173 have a new mapping while other ITRs have only an old mapping that is 2174 used until they time out. When an ITR has only an old mapping but 2175 detects bits set in the loc-status-bits that correspond to locators 2176 beyond the list it has cached, it simply ignores them. However, this 2177 can only happen for locator addresses that are lexicographically 2178 greater than the locator addresses in the existing locator-set. 2180 When a locator record is inserted in the middle of a locator-set, to 2181 maintain lexicographic order, the SMR procedure in Section 6.6.2 is 2182 used to inform ITRs and PTRs of the new locator-status-bit mappings. 2184 When a locator record is removed from a locator-set, ITRs that have 2185 the mapping cached will not use the removed locator because the xTRs 2186 will set the loc-status-bit to 0. So even if the locator is in the 2187 list, it will not be used. For new mapping requests, the xTRs can 2188 set the locator AFI to 0 (indicating an unspecified address), as well 2189 as setting the corresponding loc-status-bit to 0. This forces ITRs 2190 with old or new mappings to avoid using the removed locator. 2192 If many changes occur to a mapping over a long period of time, one 2193 will find empty record slots in the middle of the locator-set and new 2194 records appended to the locator-set. At some point, it would be 2195 useful to compact the locator-set so the loc-status-bit settings can 2196 be efficiently packed. 2198 We propose here three approaches for locator-set compaction, one 2199 operational and two protocol mechanisms. The operational approach 2200 uses a clock sweep method. The protocol approaches use the concept 2201 of Solicit-Map-Requests and Map-Versioning. 2203 6.6.1. Clock Sweep 2205 The clock sweep approach uses planning in advance and the use of 2206 count-down TTLs to time out mappings that have already been cached. 2207 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 2208 there is a 24 hour window to time out old mappings. The following 2209 clock sweep procedure is used: 2211 1. 24 hours before a mapping change is to take effect, a network 2212 administrator configures the ETRs at a site to start the clock 2213 sweep window. 2215 2. During the clock sweep window, ETRs continue to send Map-Reply 2216 messages with the current (unchanged) mapping records. The TTL 2217 for these mappings is set to 1 hour. 2219 3. 24 hours later, all previous cache entries will have timed out, 2220 and any active cache entries will time out within 1 hour. During 2221 this 1 hour window the ETRs continue to send Map-Reply messages 2222 with the current (unchanged) mapping records with the TTL set to 2223 1 minute. 2225 4. At the end of the 1 hour window, the ETRs will send Map-Reply 2226 messages with the new (changed) mapping records. So any active 2227 caches can get the new mapping contents right away if not cached, 2228 or in 1 minute if they had the mapping cached. The new mappings 2229 are cached with a time to live equal to the TTL in the Map-Reply. 2231 6.6.2. Solicit-Map-Request (SMR) 2233 Soliciting a Map-Request is a selective way for ETRs, at the site 2234 where mappings change, to control the rate they receive requests for 2235 Map-Reply messages. SMRs are also used to tell remote ITRs to update 2236 the mappings they have cached. 2238 Since the ETRs don't keep track of remote ITRs that have cached their 2239 mappings, they do not know which ITRs need to have their mappings 2240 updated. As a result, an ETR will solicit Map-Requests (called an 2241 SMR message) from those sites to which it has been sending 2242 encapsulated data to for the last minute. In particular, an ETR will 2243 send an SMR an ITR to which it has recently sent encapsulated data. 2245 An SMR message is simply a bit set in a Map-Request message. An ITR 2246 or PTR will send a Map-Request when they receive an SMR message. 2247 Both the SMR sender and the Map-Request responder MUST rate-limited 2248 these messages. Rate-limiting can be implemented as a global rate- 2249 limiter or one rate-limiter per SMR destination. 2251 The following procedure shows how a SMR exchange occurs when a site 2252 is doing locator-set compaction for an EID-to-RLOC mapping: 2254 1. When the database mappings in an ETR change, the ETRs at the site 2255 begin to send Map-Requests with the SMR bit set for each locator 2256 in each map-cache entry the ETR caches. 2258 2. A remote ITR which receives the SMR message will schedule sending 2259 a Map-Request message to the source locator address of the SMR 2260 message or to the mapping database system. A newly allocated 2261 random nonce is selected and the EID-prefix used is the one 2262 copied from the SMR message. If the source locator is the only 2263 locator in the cached locator-set, the remote ITR SHOULD send a 2264 Map-Request to the database mapping system just in case the 2265 single locator has changed and may no longer be reachable to 2266 accept the Map-Request. 2268 3. The remote ITR MUST rate-limit the Map-Request until it gets a 2269 Map-Reply while continuing to use the cached mapping. When Map 2270 Versioning is used, described in Section 6.6.3, an SMR sender can 2271 detect if an ITR is using the most up to date database mapping. 2273 4. The ETRs at the site with the changed mapping will reply to the 2274 Map-Request with a Map-Reply message that has a nonce from the 2275 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate 2276 limited. This is important to avoid Map-Reply implosion. 2278 5. The ETRs, at the site with the changed mapping, record the fact 2279 that the site that sent the Map-Request has received the new 2280 mapping data in the mapping cache entry for the remote site so 2281 the loc-status-bits are reflective of the new mapping for packets 2282 going to the remote site. The ETR then stops sending SMR 2283 messages. 2285 Experimentation is in progress to determine the appropriate rate- 2286 limit parameters. 2288 For security reasons an ITR MUST NOT process unsolicited Map-Replies. 2289 To avoid map-cache entry corruption by a third-party, a sender of an 2290 SMR-based Map-Request MUST be verified. If an ITR receives an SMR- 2291 based Map-Request and the source is not in the locator-set for the 2292 stored map-cache entry, then the responding Map-Request MUST be sent 2293 with an EID destination to the mapping database system. Since the 2294 mapping database system is more secure to reach an authoritative ETR, 2295 it will deliver the Map-Request to the authoritative source of the 2296 mapping data. 2298 When an ITR receives an SMR-based Map-Request for which it does not 2299 have a cached mapping for the EID in the SMR message, it MAY not send 2300 a SMR-invoked Map-Request. This scenario can occur when an ETR sends 2301 SMR messages to all locators in the locator-set it has stored in its 2302 map-cache but the remote ITRs that receive the SMR may not be sending 2303 packets to the site. There is no point in updating the ITRs until 2304 they need to send, in which case, they will send Map-Requests to 2305 obtain a map-cache entry. 2307 6.6.3. Database Map Versioning 2309 When there is unidirectional packet flow between an ITR and ETR, and 2310 the EID-to-RLOC mappings change on the ETR, it needs to inform the 2311 ITR so encapsulation can stop to a removed locator and start to a new 2312 locator in the locator-set. 2314 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 2315 number. This is known as the Destination Map-Version Number. ITRs 2316 include the Destination Map-Version Number in packets they 2317 encapsulate to the site. When an ETR decapsulates a packet and 2318 detects the Destination Map-Version Number is less than the current 2319 version for its mapping, the SMR procedure described in Section 6.6.2 2320 occurs. 2322 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 2323 Version number. This is known as the Source Map-Version Number. 2324 When an ETR decapsulates a packet and detects the Source Map-Version 2325 Number is greater than the last Map-Version Number sent in a Map- 2326 Reply from the ITR's site, the ETR will send a Map-Request to one of 2327 the ETRs for the source site. 2329 A Map-Version Number is used as a sequence number per EID-prefix. So 2330 values that are greater, are considered to be more recent. A value 2331 of 0 for the Source Map-Version Number or the Destination Map-Version 2332 Number conveys no versioning information and an ITR does no 2333 comparison with previously received Map-Version Numbers. 2335 A Map-Version Number can be included in Map-Register messages as 2336 well. This is a good way for the Map-Server can assure that all ETRs 2337 for a site registering to it will be Map-Version number synchronized. 2339 See [VERSIONING] for a more detailed analysis and description of 2340 Database Map Versioning. 2342 7. Router Performance Considerations 2344 LISP is designed to be very hardware-based forwarding friendly. A 2345 few implementation techniques can be used to incrementally implement 2346 LISP: 2348 o When a tunnel encapsulated packet is received by an ETR, the outer 2349 destination address may not be the address of the router. This 2350 makes it challenging for the control plane to get packets from the 2351 hardware. This may be mitigated by creating special FIB entries 2352 for the EID-prefixes of EIDs served by the ETR (those for which 2353 the router provides an RLOC translation). These FIB entries are 2354 marked with a flag indicating that control plane processing should 2355 be performed. The forwarding logic of testing for particular IP 2356 protocol number value is not necessary. There are a few proven 2357 cases where no changes to existing deployed hardware were needed 2358 to support the LISP data-plane. 2360 o On an ITR, prepending a new IP header consists of adding more 2361 bytes to a MAC rewrite string and prepending the string as part of 2362 the outgoing encapsulation procedure. Routers that support GRE 2363 tunneling [RFC2784] or 6to4 tunneling [RFC3056] may already 2364 support this action. 2366 o A packet's source address or interface the packet was received on 2367 can be used to select a VRF (Virtual Routing/Forwarding). The 2368 VRF's routing table can be used to find EID-to-RLOC mappings. 2370 For performance issues related to map-cache management, see section 2371 Section 12. 2373 8. Deployment Scenarios 2375 This section will explore how and where ITRs and ETRs can be deployed 2376 and will discuss the pros and cons of each deployment scenario. For 2377 a more detailed deployment recommendation, refer to [LISP-DEPLOY]. 2379 There are two basic deployment trade-offs to consider: centralized 2380 versus distributed caches and flat, recursive, or re-encapsulating 2381 tunneling. When deciding on centralized versus distributed caching, 2382 the following issues should be considered: 2384 o Are the tunnel routers spread out so that the caches are spread 2385 across all the memories of each router? A centralized cache is 2386 when an ITR keeps a cache for all the EIDs it is encapsulating to. 2387 The packet takes a direct path to the destination locator. A 2388 distributed cache is when an ITR needs help from other re- 2389 encapsulating routers because it does not store all the cache 2390 entries for the EIDs is it encapsulating to. So the packet takes 2391 a path through re-encapsulating routers that have a different set 2392 of cache entries. 2394 o Should management "touch points" be minimized by choosing few 2395 tunnel routers, just enough for redundancy? 2397 o In general, using more ITRs doesn't increase management load, 2398 since caches are built and stored dynamically. On the other hand, 2399 more ETRs does require more management since EID-prefix-to-RLOC 2400 mappings need to be explicitly configured. 2402 When deciding on flat, recursive, or re-encapsulation tunneling, the 2403 following issues should be considered: 2405 o Flat tunneling implements a single tunnel between source site and 2406 destination site. This generally offers better paths between 2407 sources and destinations with a single tunnel path. 2409 o Recursive tunneling is when tunneled traffic is again further 2410 encapsulated in another tunnel, either to implement VPNs or to 2411 perform Traffic Engineering. When doing VPN-based tunneling, the 2412 site has some control since the site is prepending a new tunnel 2413 header. In the case of TE-based tunneling, the site may have 2414 control if it is prepending a new tunnel header, but if the site's 2415 ISP is doing the TE, then the site has no control. Recursive 2416 tunneling generally will result in suboptimal paths but at the 2417 benefit of steering traffic to resource available parts of the 2418 network. 2420 o The technique of re-encapsulation ensures that packets only 2421 require one tunnel header. So if a packet needs to be rerouted, 2422 it is first decapsulated by the ETR and then re-encapsulated with 2423 a new tunnel header using a new RLOC. 2425 The next sub-sections will survey where tunnel routers can reside in 2426 the network. 2428 8.1. First-hop/Last-hop Tunnel Routers 2430 By locating tunnel routers close to hosts, the EID-prefix set is at 2431 the granularity of an IP subnet. So at the expense of more EID- 2432 prefix-to-RLOC sets for the site, the caches in each tunnel router 2433 can remain relatively small. But caches always depend on the number 2434 of non-aggregated EID destination flows active through these tunnel 2435 routers. 2437 With more tunnel routers doing encapsulation, the increase in control 2438 traffic grows as well: since the EID-granularity is greater, more 2439 Map-Requests and Map-Replies are traveling between more routers. 2441 The advantage of placing the caches and databases at these stub 2442 routers is that the products deployed in this part of the network 2443 have better price-memory ratios then their core router counterparts. 2444 Memory is typically less expensive in these devices and fewer routes 2445 are stored (only IGP routes). These devices tend to have excess 2446 capacity, both for forwarding and routing state. 2448 LISP functionality can also be deployed in edge switches. These 2449 devices generally have layer-2 ports facing hosts and layer-3 ports 2450 facing the Internet. Spare capacity is also often available in these 2451 devices as well. 2453 8.2. Border/Edge Tunnel Routers 2455 Using customer-edge (CE) routers for tunnel endpoints allows the EID 2456 space associated with a site to be reachable via a small set of RLOCs 2457 assigned to the CE routers for that site. This is the default 2458 behavior envisioned in the rest of this specification. 2460 This offers the opposite benefit of the first-hop/last-hop tunnel 2461 router scenario: the number of mapping entries and network management 2462 touch points are reduced, allowing better scaling. 2464 One disadvantage is that less of the network's resources are used to 2465 reach host endpoints thereby centralizing the point-of-failure domain 2466 and creating network choke points at the CE router. 2468 Note that more than one CE router at a site can be configured with 2469 the same IP address. In this case an RLOC is an anycast address. 2470 This allows resilience between the CE routers. That is, if a CE 2471 router fails, traffic is automatically routed to the other routers 2472 using the same anycast address. However, this comes with the 2473 disadvantage where the site cannot control the entrance point when 2474 the anycast route is advertised out from all border routers. Another 2475 disadvantage of using anycast locators is the limited advertisement 2476 scope of /32 (or /128 for IPv6) routes. 2478 8.3. ISP Provider-Edge (PE) Tunnel Routers 2480 Use of ISP PE routers as tunnel endpoint routers is not the typical 2481 deployment scenario envisioned in the specification. This section 2482 attempts to capture some of reasoning behind this preference of 2483 implementing LISP on CE routers. 2485 Use of ISP PE routers as tunnel endpoint routers gives an ISP, rather 2486 than a site, control over the location of the egress tunnel 2487 endpoints. That is, the ISP can decide if the tunnel endpoints are 2488 in the destination site (in either CE routers or last-hop routers 2489 within a site) or at other PE edges. The advantage of this case is 2490 that two tunnel headers can be avoided. By having the PE be the 2491 first router on the path to encapsulate, it can choose a TE path 2492 first, and the ETR can decapsulate and re-encapsulate for a tunnel to 2493 the destination end site. 2495 An obvious disadvantage is that the end site has no control over 2496 where its packets flow or the RLOCs used. Other disadvantages 2497 include the difficulty in synchronizing path liveness updates between 2498 CE and PE routers. 2500 As mentioned in earlier sections a combination of these scenarios is 2501 possible at the expense of extra packet header overhead, if both site 2502 and provider want control, then recursive or re-encapsulating tunnels 2503 are used. 2505 8.4. LISP Functionality with Conventional NATs 2507 LISP routers can be deployed behind Network Address Translator (NAT) 2508 devices to provide the same set of packet services hosts have today 2509 when they are addressed out of private address space. 2511 It is important to note that a locator address in any LISP control 2512 message MUST be a globally routable address and therefore SHOULD NOT 2513 contain [RFC1918] addresses. If a LISP router is configured with 2514 private addresses, they MUST be used only in the outer IP header so 2515 the NAT device can translate properly. Otherwise, EID addresses MUST 2516 be translated before encapsulation is performed. Both NAT 2517 translation and LISP encapsulation functions could be co-located in 2518 the same device. 2520 More details on LISP address translation can be found in [INTERWORK]. 2522 8.5. Packets Egressing a LISP Site 2524 When a LISP site is using two ITRs for redundancy, the failure of one 2525 ITR will likely shift outbound traffic to the second. This second 2526 ITR's cache may not not be populated with the same EID-to-RLOC 2527 mapping entries as the first. If this second ITR does not have these 2528 mappings, traffic will be dropped while the mappings are retrieved 2529 from the mapping system. The retrieval of these messages may 2530 increase the load of requests being sent into the mapping system. 2531 Deployment and experimentation will determine whether this issue 2532 requires more attention. 2534 9. Traceroute Considerations 2536 When a source host in a LISP site initiates a traceroute to a 2537 destination host in another LISP site, it is highly desirable for it 2538 to see the entire path. Since packets are encapsulated from ITR to 2539 ETR, the hop across the tunnel could be viewed as a single hop. 2540 However, LISP traceroute will provide the entire path so the user can 2541 see 3 distinct segments of the path from a source LISP host to a 2542 destination LISP host: 2544 Segment 1 (in source LISP site based on EIDs): 2546 source-host ---> first-hop ... next-hop ---> ITR 2548 Segment 2 (in the core network based on RLOCs): 2550 ITR ---> next-hop ... next-hop ---> ETR 2552 Segment 3 (in the destination LISP site based on EIDs): 2554 ETR ---> next-hop ... last-hop ---> destination-host 2556 For segment 1 of the path, ICMP Time Exceeded messages are returned 2557 in the normal matter as they are today. The ITR performs a TTL 2558 decrement and test for 0 before encapsulating. So the ITR hop is 2559 seen by the traceroute source has an EID address (the address of 2560 site-facing interface). 2562 For segment 2 of the path, ICMP Time Exceeded messages are returned 2563 to the ITR because the TTL decrement to 0 is done on the outer 2564 header, so the destination of the ICMP messages are to the ITR RLOC 2565 address, the source RLOC address of the encapsulated traceroute 2566 packet. The ITR looks inside of the ICMP payload to inspect the 2567 traceroute source so it can return the ICMP message to the address of 2568 the traceroute client as well as retaining the core router IP address 2569 in the ICMP message. This is so the traceroute client can display 2570 the core router address (the RLOC address) in the traceroute output. 2571 The ETR returns its RLOC address and responds to the TTL decrement to 2572 0 like the previous core routers did. 2574 For segment 3, the next-hop router downstream from the ETR will be 2575 decrementing the TTL for the packet that was encapsulated, sent into 2576 the core, decapsulated by the ETR, and forwarded because it isn't the 2577 final destination. If the TTL is decremented to 0, any router on the 2578 path to the destination of the traceroute, including the next-hop 2579 router or destination, will send an ICMP Time Exceeded message to the 2580 source EID of the traceroute client. The ICMP message will be 2581 encapsulated by the local ITR and sent back to the ETR in the 2582 originated traceroute source site, where the packet will be delivered 2583 to the host. 2585 9.1. IPv6 Traceroute 2587 IPv6 traceroute follows the procedure described above since the 2588 entire traceroute data packet is included in ICMP Time Exceeded 2589 message payload. Therefore, only the ITR needs to pay special 2590 attention for forwarding ICMP messages back to the traceroute source. 2592 9.2. IPv4 Traceroute 2594 For IPv4 traceroute, we cannot follow the above procedure since IPv4 2595 ICMP Time Exceeded messages only include the invoking IP header and 8 2596 bytes that follow the IP header. Therefore, when a core router sends 2597 an IPv4 Time Exceeded message to an ITR, all the ITR has in the ICMP 2598 payload is the encapsulated header it prepended followed by a UDP 2599 header. The original invoking IP header, and therefore the identity 2600 of the traceroute source is lost. 2602 The solution we propose to solve this problem is to cache traceroute 2603 IPv4 headers in the ITR and to match them up with corresponding IPv4 2604 Time Exceeded messages received from core routers and the ETR. The 2605 ITR will use a circular buffer for caching the IPv4 and UDP headers 2606 of traceroute packets. It will select a 16-bit number as a key to 2607 find them later when the IPv4 Time Exceeded messages are received. 2608 When an ITR encapsulates an IPv4 traceroute packet, it will use the 2609 16-bit number as the UDP source port in the encapsulating header. 2610 When the ICMP Time Exceeded message is returned to the ITR, the UDP 2611 header of the encapsulating header is present in the ICMP payload 2612 thereby allowing the ITR to find the cached headers for the 2613 traceroute source. The ITR puts the cached headers in the payload 2614 and sends the ICMP Time Exceeded message to the traceroute source 2615 retaining the source address of the original ICMP Time Exceeded 2616 message (a core router or the ETR of the site of the traceroute 2617 destination). 2619 The signature of a traceroute packet comes in two forms. The first 2620 form is encoded as a UDP message where the destination port is 2621 inspected for a range of values. The second form is encoded as an 2622 ICMP message where the IP identification field is inspected for a 2623 well-known value. 2625 9.3. Traceroute using Mixed Locators 2627 When either an IPv4 traceroute or IPv6 traceroute is originated and 2628 the ITR encapsulates it in the other address family header, you 2629 cannot get all 3 segments of the traceroute. Segment 2 of the 2630 traceroute can not be conveyed to the traceroute source since it is 2631 expecting addresses from intermediate hops in the same address format 2632 for the type of traceroute it originated. Therefore, in this case, 2633 segment 2 will make the tunnel look like one hop. All the ITR has to 2634 do to make this work is to not copy the inner TTL to the outer, 2635 encapsulating header's TTL when a traceroute packet is encapsulated 2636 using an RLOC from a different address family. This will cause no 2637 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2639 10. Mobility Considerations 2641 There are several kinds of mobility of which only some might be of 2642 concern to LISP. Essentially they are as follows. 2644 10.1. Site Mobility 2646 A site wishes to change its attachment points to the Internet, and 2647 its LISP Tunnel Routers will have new RLOCs when it changes upstream 2648 providers. Changes in EID-RLOC mappings for sites are expected to be 2649 handled by configuration, outside of the LISP protocol. 2651 10.2. Slow Endpoint Mobility 2653 An individual endpoint wishes to move, but is not concerned about 2654 maintaining session continuity. Renumbering is involved. LISP can 2655 help with the issues surrounding renumbering [RFC4192] [LISA96] by 2656 decoupling the address space used by a site from the address spaces 2657 used by its ISPs. [RFC4984] 2659 10.3. Fast Endpoint Mobility 2661 Fast endpoint mobility occurs when an endpoint moves relatively 2662 rapidly, changing its IP layer network attachment point. Maintenance 2663 of session continuity is a goal. This is where the Mobile IPv4 2664 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used, 2665 and primarily where interactions with LISP need to be explored. 2667 The problem is that as an endpoint moves, it may require changes to 2668 the mapping between its EID and a set of RLOCs for its new network 2669 location. When this is added to the overhead of mobile IP binding 2670 updates, some packets might be delayed or dropped. 2672 In IPv4 mobility, when an endpoint is away from home, packets to it 2673 are encapsulated and forwarded via a home agent which resides in the 2674 home area the endpoint's address belongs to. The home agent will 2675 encapsulate and forward packets either directly to the endpoint or to 2676 a foreign agent which resides where the endpoint has moved to. 2677 Packets from the endpoint may be sent directly to the correspondent 2678 node, may be sent via the foreign agent, or may be reverse-tunneled 2679 back to the home agent for delivery to the mobile node. As the 2680 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 2681 mappings are required for communication between the mobile node and 2682 the home agent, whether via foreign agent or not. As a mobile 2683 endpoint changes networks, up to three LISP mapping changes may be 2684 required: 2686 o The mobile node moves from an old location to a new visited 2687 network location and notifies its home agent that it has done so. 2688 The Mobile IPv4 control packets the mobile node sends pass through 2689 one of the new visited network's ITRs, which needs a EID-RLOC 2690 mapping for the home agent. 2692 o The home agent might not have the EID-RLOC mappings for the mobile 2693 node's "care-of" address or its foreign agent in the new visited 2694 network, in which case it will need to acquire them. 2696 o When packets are sent directly to the correspondent node, it may 2697 be that no traffic has been sent from the new visited network to 2698 the correspondent node's network, and the new visited network's 2699 ITR will need to obtain an EID-RLOC mapping for the correspondent 2700 node's site. 2702 In addition, if the IPv4 endpoint is sending packets from the new 2703 visited network using its original EID, then LISP will need to 2704 perform a route-returnability check on the new EID-RLOC mapping for 2705 that EID. 2707 In IPv6 mobility, packets can flow directly between the mobile node 2708 and the correspondent node in either direction. The mobile node uses 2709 its "care-of" address (EID). In this case, the route-returnability 2710 check would not be needed but one more LISP mapping lookup may be 2711 required instead: 2713 o As above, three mapping changes may be needed for the mobile node 2714 to communicate with its home agent and to send packets to the 2715 correspondent node. 2717 o In addition, another mapping will be needed in the correspondent 2718 node's ITR, in order for the correspondent node to send packets to 2719 the mobile node's "care-of" address (EID) at the new network 2720 location. 2722 When both endpoints are mobile the number of potential mapping 2723 lookups increases accordingly. 2725 As a mobile node moves there are not only mobility state changes in 2726 the mobile node, correspondent node, and home agent, but also state 2727 changes in the ITRs and ETRs for at least some EID-prefixes. 2729 The goal is to support rapid adaptation, with little delay or packet 2730 loss for the entire system. Also IP mobility can be modified to 2731 require fewer mapping changes. In order to increase overall system 2732 performance, there may be a need to reduce the optimization of one 2733 area in order to place fewer demands on another. 2735 In LISP, one possibility is to "glean" information. When a packet 2736 arrives, the ETR could examine the EID-RLOC mapping and use that 2737 mapping for all outgoing traffic to that EID. It can do this after 2738 performing a route-returnability check, to ensure that the new 2739 network location does have a internal route to that endpoint. 2740 However, this does not cover the case where an ITR (the node assigned 2741 the RLOC) at the mobile-node location has been compromised. 2743 Mobile IP packet exchange is designed for an environment in which all 2744 routing information is disseminated before packets can be forwarded. 2745 In order to allow the Internet to grow to support expected future 2746 use, we are moving to an environment where some information may have 2747 to be obtained after packets are in flight. Modifications to IP 2748 mobility should be considered in order to optimize the behavior of 2749 the overall system. Anything which decreases the number of new EID- 2750 RLOC mappings needed when a node moves, or maintains the validity of 2751 an EID-RLOC mapping for a longer time, is useful. 2753 10.4. Fast Network Mobility 2755 In addition to endpoints, a network can be mobile, possibly changing 2756 xTRs. A "network" can be as small as a single router and as large as 2757 a whole site. This is different from site mobility in that it is 2758 fast and possibly short-lived, but different from endpoint mobility 2759 in that a whole prefix is changing RLOCs. However, the mechanisms 2760 are the same and there is no new overhead in LISP. A map request for 2761 any endpoint will return a binding for the entire mobile prefix. 2763 If mobile networks become a more common occurrence, it may be useful 2764 to revisit the design of the mapping service and allow for dynamic 2765 updates of the database. 2767 The issue of interactions between mobility and LISP needs to be 2768 explored further. Specific improvements to the entire system will 2769 depend on the details of mapping mechanisms. Mapping mechanisms 2770 should be evaluated on how well they support session continuity for 2771 mobile nodes. 2773 10.5. LISP Mobile Node Mobility 2775 A mobile device can use the LISP infrastructure to achieve mobility 2776 by implementing the LISP encapsulation and decapsulation functions 2777 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 2778 node" can use topologically-independent EID IP addresses that are not 2779 advertised into and do not impose a cost on the global routing 2780 system. These EIDs are maintained at the edges of the mapping system 2781 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 2782 only the correspondents of the LISP mobile node. 2784 Refer to the LISP Mobility Architecture specification [LISP-MN] for 2785 more details. 2787 11. Multicast Considerations 2789 A multicast group address, as defined in the original Internet 2790 architecture is an identifier of a grouping of topologically 2791 independent receiver host locations. The address encoding itself 2792 does not determine the location of the receiver(s). The multicast 2793 routing protocol, and the network-based state the protocol creates, 2794 determines where the receivers are located. 2796 In the context of LISP, a multicast group address is both an EID and 2797 a Routing Locator. Therefore, no specific semantic or action needs 2798 to be taken for a destination address, as it would appear in an IP 2799 header. Therefore, a group address that appears in an inner IP 2800 header built by a source host will be used as the destination EID. 2801 The outer IP header (the destination Routing Locator address), 2802 prepended by a LISP router, will use the same group address as the 2803 destination Routing Locator. 2805 Having said that, only the source EID and source Routing Locator 2806 needs to be dealt with. Therefore, an ITR merely needs to put its 2807 own IP address in the source Routing Locator field when prepending 2808 the outer IP header. This source Routing Locator address, like any 2809 other Routing Locator address MUST be globally routable. 2811 Therefore, an EID-to-RLOC mapping does not need to be performed by an 2812 ITR when a received data packet is a multicast data packet or when 2813 processing a source-specific Join (either by IGMPv3 or PIM). But the 2814 source Routing Locator is decided by the multicast routing protocol 2815 in a receiver site. That is, an EID to Routing Locator translation 2816 is done at control-time. 2818 Another approach is to have the ITR not encapsulate a multicast 2819 packet and allow the host built packet to flow into the core even if 2820 the source address is allocated out of the EID namespace. If the 2821 RPF-Vector TLV [RFC5496] is used by PIM in the core, then core 2822 routers can RPF to the ITR (the Locator address which is injected 2823 into core routing) rather than the host source address (the EID 2824 address which is not injected into core routing). 2826 To avoid any EID-based multicast state in the network core, the first 2827 approach is chosen for LISP-Multicast. Details for LISP-Multicast 2828 and Interworking with non-LISP sites is described in specification 2829 [MLISP]. 2831 12. Security Considerations 2833 It is believed that most of the security mechanisms will be part of 2834 the mapping database service when using control plane procedures for 2835 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 2836 as described in this specification, protection is provided against 2837 ETR spoofing by using Return-Routability mechanisms evidenced by the 2838 use of a 24-bit Nonce field in the LISP encapsulation header and a 2839 64-bit Nonce field in the LISP control message. 2841 The nonce, coupled with the ITR accepting only solicited Map-Replies 2842 provides a basic level of security, in many ways similar to the 2843 security experienced in the current Internet routing system. It is 2844 hard for off-path attackers to launch attacks against these LISP 2845 mechanisms, as they do not have the nonce values. Sending a large 2846 number of packets to accidentally find the right nonce value is 2847 possible, but would already by itself be a denial-of-service attack. 2848 On-path attackers can perform far more serious attacks, but on-path 2849 attackers can launch serious attacks in the current Internet as well, 2850 including eavesdropping, blocking or redirecting traffic. 2852 LISP does not rely on a PKI infrastructure or a more heavy weight 2853 authentication system. These systems challenge the scalability of 2854 LISP which was a primary design goal. 2856 DoS attack prevention will depend on implementations rate-limiting 2857 Map-Requests and Map-Replies to the control plane as well as rate- 2858 limiting the number of data-triggered Map-Replies. 2860 An incorrectly implemented or malicious ITR might choose to ignore 2861 the priority and weights provided by the ETR in its Map-Reply. This 2862 traffic steering would be limited to the traffic that is sent by this 2863 ITR's site, and no more severe than if the site initiated a bandwidth 2864 DoS attack on (one of) the ETR's ingress links. The ITR's site would 2865 typically gain no benefit from not respecting the weights, and would 2866 likely to receive better service by abiding by them. 2868 To deal with map-cache exhaustion attempts in an ITR/PTR, the 2869 implementation should consider putting a maximum cap on the number of 2870 entries stored with a reserve list for special or frequently accessed 2871 sites. This should be a configuration policy control set by the 2872 network administrator who manages ITRs and PTRs. When overlapping 2873 EID-prefixes occur across multiple map-cache entries, the integrity 2874 of the set must be wholly maintained. So if a more-specific entry 2875 cannot be added due to reaching the maximum cap, then none of the 2876 less specifics should be stored in the map-cache. 2878 Given that the ITR/PTR maintains a cache of EID-to-RLOC mappings, 2879 cache sizing and maintenance is an issue to be kept in mind during 2880 implementation. It is a good idea to have instrumentation in place 2881 to detect thrashing of the cache. Implementation experimentation 2882 will be used to determine which cache management strategies work 2883 best. In general, it is difficult to defend against cache trashing 2884 attacks. It should be noted that an undersized cache in an ITR/PTR 2885 not only causes adverse affect on the site or region they support, 2886 but may also cause increased Map-Request load on the mapping system. 2888 There is a security risk implicit in the fact that ETRs generate the 2889 EID prefix to which they are responding. An ETR can claim a shorter 2890 prefix than it is actually responsible for. Various mechanisms to 2891 ameliorate or resolve this issue will be examined in the future, 2892 [LISP-SEC]. 2894 Spoofing of inner header addresses of LISP encapsulated packets is 2895 possible like with any tunneling mechanism. ITRs MUST verify the 2896 source address of a packet to be an EID that belongs to the site's 2897 EID-prefix range prior to encapsulation. An ETR must only 2898 decapsulate and forward datagrams with an inner header destination 2899 that matches one of its EID-prefix ranges. If, upon receipt and 2900 decapsulation, the destination EID of a datagram does not match one 2901 of the ETR's configured EID-prefixes, the ETR MUST drop the datagram. 2902 If a LISP encapsulated packet arrives at an ETR, it SHOULD compare 2903 the inner header source EID address and the outer header source RLOC 2904 address with the mapping that exists in the mapping database. Then 2905 when spoofing attacks occur, the outer header source RLOC address can 2906 be used to trace back the attack to the source site, using existing 2907 operational tools. 2909 This experimental specification does not address automated key 2910 management (AKM). BCP 107 provides guidance in this area. In 2911 addition, at the time of this writing, substantial work is being 2912 undertaken to improve security of the routing system [KARP], [RPKI], 2913 [BGP-SEC], [LISP-SEC]. Future work on LISP should address BCP-107 as 2914 well as other open security considerations, which may require changes 2915 to this specification. 2917 13. Network Management Considerations 2919 Considerations for Network Management tools exist so the LISP 2920 protocol suite can be operationally managed. The mechanisms can be 2921 found in [LISP-MIB] and [LISP-LIG]. 2923 14. IANA Considerations 2925 This section provides guidance to the Internet Assigned Numbers 2926 Authority (IANA) regarding registration of values related to the LISP 2927 specification, in accordance with BCP 26 and RFC 5226 [RFC5226]. 2929 There are two name spaces in LISP that require registration: 2931 o LISP IANA registry allocations should not be made for purposes 2932 unrelated to LISP routing or transport protocols. 2934 o The following policies are used here with the meanings defined in 2935 BCP 26: "Specification Required", "IETF Review", "Experimental 2936 Use", "First Come First Served". 2938 14.1. LISP ACT and Flag Fields 2940 New ACT values (Section 6.1.5) can be allocated through IETF review 2941 or IESG approval. Four values have already been allocated by this 2942 specification (Section 6.1.5). 2944 In addition, the LISP protocol has a number of flag and reserved 2945 fields, such as the LISP header flags field (Section 5.3). New bits 2946 for flags can be taken into use from these fields through IETF review 2947 or IESG approval, but these need not be managed by IANA. 2949 14.2. LISP Address Type Codes 2951 LISP Address [LCAF] type codes have a range from 0 to 255. New type 2952 codes MUST be allocated consecutively starting at 0. Type Codes 0 - 2953 127 are to be assigned by IETF review or IESG approval. 2955 Type Codes 128 - 255 are available on a First Come First Served 2956 policy. 2958 This registry, initially empty, is constructed for future-use 2959 experimental work of LCAF values. See [LCAF] for details for other 2960 possible unapproved address encodings. The unapproved LCAF encodings 2961 are an area for further study and experimentation. 2963 14.3. LISP UDP Port Numbers 2965 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2966 LISP data-plane and control-plane operation, respectively. 2968 14.4. LISP Key ID Numbers 2970 The following Key ID values are defined by this specification as used 2971 in any packet type that references a Key ID field: 2973 Name Number Defined in 2974 ----------------------------------------------- 2975 None 0 n/a 2976 HMAC-SHA-1-96 1 [RFC2404] 2977 HMAC-SHA-256-128 2 [RFC6234] 2979 15. Known Open Issues and Areas of Future Work 2981 As an experimental specification, this work is, by definition, 2982 incomplete. Specific areas where additional experience and work are 2983 needed include: 2985 o At present, only [ALT] is defined for implementing a database of 2986 EID-to-RLOC mapping information. Additional research on other 2987 mapping database systems is strongly encouraged. 2989 o Failure and recovery of LISP site partitioning (see Section 6.4), 2990 in the presence of redundant configuration (see Section 8.5) needs 2991 further research and experimentation. 2993 o The characteristics of map-cache management under exceptional 2994 conditions, such as denial-of-service attacks are not fully 2995 understood. Further experience is needed to determine whether 2996 current caching methods are practical or in need of further 2997 development. See Section 12 for additional thoughts and 2998 considerations. 3000 o Preliminary work has been done to ensure that sites employing LISP 3001 can interconnect with the rest of the Internet. This work is 3002 documented in [INTERWORK] but further experimentation and 3003 experience is needed. 3005 o At present, no mechanism for automated key management for message 3006 authentication is defined. Addressing automated key management is 3007 necessary before this specification could be developed into a 3008 standards track RFC. See Section 12 for further details regarding 3009 security considerations. 3011 16. References 3013 16.1. Normative References 3015 [ALT] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, "LISP 3016 Alternative Topology (LISP-ALT)", 3017 draft-ietf-lisp-alt-09.txt (work in progress). 3019 [LISP-MS] Farinacci, D. and V. Fuller, "LISP Map Server", 3020 draft-ietf-lisp-ms-12.txt (work in progress). 3022 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 3023 August 1980. 3025 [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and 3026 E. Lear, "Address Allocation for Private Internets", 3027 BCP 5, RFC 1918, February 1996. 3029 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3030 Requirement Levels", BCP 14, RFC 2119, March 1997. 3032 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 3033 ESP and AH", RFC 2404, November 1998. 3035 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3036 of Explicit Congestion Notification (ECN) to IP", 3037 RFC 3168, September 2001. 3039 [RFC3232] Reynolds, J., "Assigned Numbers: RFC 1700 is Replaced by 3040 an On-line Database", RFC 3232, January 2002. 3042 [RFC4086] Eastlake, D., Schiller, J., and S. Crocker, "Randomness 3043 Requirements for Security", BCP 106, RFC 4086, June 2005. 3045 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 3046 (CIDR): The Internet Address Assignment and Aggregation 3047 Plan", BCP 122, RFC 4632, August 2006. 3049 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 3050 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 3051 May 2008. 3053 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 3054 Forwarding (RPF) Vector TLV", RFC 5496, March 2009. 3056 [RFC5944] Perkins, C., "IP Mobility Support for IPv4, Revised", 3057 RFC 5944, November 2010. 3059 [RFC6234] Eastlake, D. and T. Hansen, "US Secure Hash Algorithms 3060 (SHA and SHA-based HMAC and HKDF)", RFC 6234, May 2011. 3062 [RFC6275] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support 3063 in IPv6", RFC 6275, July 2011. 3065 [UDP-TUNNELS] 3066 Eubanks, M. and P. Chimento, "UDP Checksums for Tunneled 3067 Packets", draft-eubanks-chimento-6man-01.txt (work in 3068 progress), October 2010. 3070 [VERSIONING] 3071 Iannone, L., Saucez, D., and O. Bonaventure, "LISP Mapping 3072 Versioning", draft-ietf-lisp-map-versioning-05.txt (work 3073 in progress). 3075 16.2. Informative References 3077 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3078 NUMBERS 3079 http://www.iana.org/assignments/address-family-numbers. 3081 [AFI-REGISTRY] 3082 IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 3083 NUMBER registry http://www.iana.org/assignments/ 3084 address-family-numbers/ 3085 address-family-numbers.xml#address-family-numbers-1. 3087 [BGP-SEC] Lepinski, M., "An Overview of BGPSEC", 3088 draft-lepinski-bgpsec-overview-00.txt (work in progress), 3089 March 2011. 3091 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 3092 Enhancement to the Internet Architecture", Internet- 3093 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt. 3095 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 3096 Content distribution Overlay Network Service for LISP", 3097 draft-meyer-lisp-cons-04.txt (work in progress). 3099 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 3100 Mappings Multicast Across Cooperating Systems for LISP", 3101 draft-curran-lisp-emacs-00.txt (work in progress). 3103 [INTERWORK] 3104 Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 3105 "Interworking LISP with IPv4 and IPv6", 3106 draft-ietf-lisp-interworking-02.txt (work in progress). 3108 [KARP] Lebovitz, G. and M. Bhatia, "Keying and Authentication for 3109 Routing Protocols (KARP)Design Guidelines", 3110 draft-ietf-karp-design-guide-06.txt (work in progress), 3111 October 2011. 3113 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 3114 Address Format", draft-farinacci-lisp-lcaf-06.txt (work in 3115 progress). 3117 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 3118 "Renumbering: Threat or Menace?", Usenix . 3120 [LISP-DEPLOY] 3121 Jakab, L., Coras, F., Domingo-Pascual, J., and D. Lewis, 3122 "LISP Network Element Deployment Considerations", 3123 draft-ietf-lisp-deployment-02.txt (work in progress). 3125 [LISP-LIG] 3126 Farinacci, D. and D. Meyer, "LISP Internet Groper (LIG)", 3127 draft-ietf-lisp-lig-06.txt (work in progress). 3129 [LISP-MAIN] 3130 Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 3131 "Locator/ID Separation Protocol (LISP)", 3132 draft-farinacci-lisp-12.txt (work in progress). 3134 [LISP-MIB] 3135 Schudel, G., Jain, A., and V. Moreno, "LISP MIB", 3136 draft-ietf-lisp-mib-02.txt (work in progress). 3138 [LISP-MN] Farinacci, D., Fuller, V., Lewis, D., and D. Meyer, "LISP 3139 Mobility Architecture", draft-meyer-lisp-mn-06.txt (work 3140 in progress). 3142 [LISP-SEC] 3143 Maino, F., Ermagon, V., Cabellos, A., Sausez, D., and O. 3144 Bonaventure, "LISP-Security (LISP-SEC)", 3145 draft-ietf-lisp-sec-00.txt (work in progress). 3147 [LOC-ID-ARCH] 3148 Meyer, D. and D. Lewis, "Architectural Implications of 3149 Locator/ID Separation", 3150 draft-meyer-loc-id-implications-01.txt (work in progress). 3152 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 3153 "LISP for Multicast Environments", 3154 draft-ietf-lisp-multicast-10.txt (work in progress). 3156 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 3157 draft-lear-lisp-nerd-08.txt (work in progress). 3159 [OPENLISP] 3160 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 3161 Report", draft-iannone-openlisp-implementation-01.txt 3162 (work in progress). 3164 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 3165 draft-narten-radir-problem-statement-05.txt (work in 3166 progress). 3168 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 3169 STD 13, RFC 1034, November 1987. 3171 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 3172 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 3173 March 2000. 3175 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 3176 via IPv4 Clouds", RFC 3056, February 2001. 3178 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 3179 A., Peterson, J., Sparks, R., Handley, M., and E. 3180 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 3181 June 2002. 3183 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 3184 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 3185 September 2005. 3187 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 3188 Optimization for Mobile IPv6", RFC 4866, May 2007. 3190 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 3191 Workshop on Routing and Addressing", RFC 4984, 3192 September 2007. 3194 [RPKI] Lepinski, M., "An Infrastructure to Support Secure 3195 Internet Routing", draft-ietf-sidr-arch-13.txt (work in 3196 progress), February 2011. 3198 [RPMD] Handley, M., Huici, F., and A. Greenhalgh, "RPMD: Protocol 3199 for Routing Protocol Meta-data Dissemination", 3200 draft-handley-p2ppush-unpublished-2007726.txt (work in 3201 progress). 3203 Appendix A. Acknowledgments 3205 An initial thank you goes to Dave Oran for planting the seeds for the 3206 initial ideas for LISP. His consultation continues to provide value 3207 to the LISP authors. 3209 A special and appreciative thank you goes to Noel Chiappa for 3210 providing architectural impetus over the past decades on separation 3211 of location and identity, as well as detailed review of the LISP 3212 architecture and documents, coupled with enthusiasm for making LISP a 3213 practical and incremental transition for the Internet. 3215 The authors would like to gratefully acknowledge many people who have 3216 contributed discussion and ideas to the making of this proposal. 3217 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 3218 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 3219 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 3220 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 3221 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 3222 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 3223 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 3224 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 3225 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 3226 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 3227 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 3228 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 3229 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 3230 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 3231 Ermagan, Albert Cabellos, Fabio Maino, Victor Moreno, Chris White, 3232 Clarence Filsfils, and Alia Atlas. 3234 This work originated in the Routing Research Group (RRG) of the IRTF. 3235 The individual submission [LISP-MAIN] was converted into this IETF 3236 LISP working group draft. 3238 The LISP working group would like to give a special thanks to Jari 3239 Arkko, the Internet Area AD at the time the set of LISP documents 3240 were being prepared for IESG last call, for his meticulous review and 3241 detail commentary on the 7 working group last call drafts progressing 3242 toward experimental RFCs. 3244 Appendix B. Document Change Log 3246 B.1. Changes to draft-ietf-lisp-16.txt 3248 o Posted October 2011 after AD review by Jari. 3250 B.2. Changes to draft-ietf-lisp-15.txt 3252 o Posted July 2011. Fixing IDnits errors. 3254 o Change description on how to select a source address for RLOC- 3255 probe Map-Replies to refer to the "EID-to-RLOC Map-Reply Message" 3256 section. 3258 B.3. Changes to draft-ietf-lisp-14.txt 3260 o Post working group last call and pre-IESG last call review. 3262 o Indicate that an ICMP Unreachable message should be sent when a 3263 packet matches a drop-based negative map-cache entry. 3265 o Indicate how a map-cache set of overlapping EID-prefixes must 3266 maintain integrity when the map-cache maximum cap is reached. 3268 o Add Joel's description for the definition of an EID, that the bit 3269 string value can be an RLOC for another device in abstract but the 3270 architecture allows it to be an EID of one device and the same 3271 value as an RLOC for another device. 3273 o In the "Tunnel Encapsulation Details" section, indicate that 4 3274 combinations of encapsulation are supported. 3276 o Add what ETR should do for a Data-Probe when received for a 3277 destination EID outside of its EID-prefix range. This was added 3278 in the Data Probe definition section. 3280 o Added text indicating that more-specific EID-prefixes must not be 3281 removed when less-specific entries stay in the map-cache. This is 3282 to preserve the integrity of the EID-prefix set. 3284 o Add clarifying text in the Security Considerations section about 3285 how an ETR must not decapsulate and forward a packet that is not 3286 for its configured EID-prefix range. 3288 B.4. Changes to draft-ietf-lisp-13.txt 3290 o Posted June 2011 to complete working group last call. 3292 o Tracker item 87. Put Yakov suggested wording in the EID-prefix 3293 definition section to reference [INTERWORK] and [LISP-DEPLOY] 3294 about discussion on transition and access mechanisms. 3296 o Change "ITRs" to "ETRs" in the Locator Status Bit definition 3297 section and data packet description section per Damien's comment. 3299 o Remove the normative reference to [LISP-SEC] when describing the 3300 S-bit in the ECM and Map-Reply headers. 3302 o Tracker item 54. Added text from John Scudder in the "Packets 3303 Egressing a LISP Site" section. 3305 o Add sentence to the "Reencapsulating Tunnel" definition about how 3306 reencapsulation loops can occur when not coordinating among 3307 multiple mapping database systems. 3309 o Remove "In theory" from a sentence in the Security Considerations 3310 section. 3312 o Remove Security Area Statement title and reword section with 3313 Eliot's provided text. The text was agreed upon by LISP-WG chairs 3314 and Security ADs. 3316 o Remove word "potential" from the over-claiming paragraph of the 3317 Security Considerations section per Stephen's request. 3319 o Wordsmithing and other editorial comments from Alia. 3321 B.5. Changes to draft-ietf-lisp-12.txt 3323 o Posted April 2011. 3325 o Tracker item 87. Provided rewording how an EID-prefix can be 3326 reused in the definition section of "EID-prefix". 3328 o Tracker item 95. Change "eliminate" to "defer" in section 4.1. 3330 o Tracker item 110. Added that the Mapping Protocol Data field in 3331 the Map-Reply message is only used when needed by the particular 3332 Mapping Database System. 3334 o Tracker item 111. Indicate that if an LSB that is associated with 3335 an anycast address, that there is at least one RLOC that is up. 3337 o Tracker item 108. Make clear the R-bit does not define RLOC path 3338 reachability. 3340 o Tracker item 107. Indicate that weights are relative to each 3341 other versus requiring an addition of up to 100%. 3343 o Tracker item 46. Add a sentence how LISP products should be sized 3344 for the appropriate demand so cache thrashing is avoided. 3346 o Change some references of RFC 5226 to [AFI] per Luigi. 3348 o Per Luigi, make reference to "EID-AFI" consistent to "EID-prefix- 3349 AFI". 3351 o Tracker item 66. Indicate that appending locators to a locator- 3352 set is done when the added locators are lexicographically greater 3353 than the previous ones in the set. 3355 o Tracker item 87. Once again reword the definition of the EID- 3356 prefix to reflect recent comments. 3358 o Tracker item 70. Added text to security section on what the 3359 implications could be if an ITR does not obey priority and weights 3360 from a Map-Reply message. 3362 o Tracker item 54. Added text to the new section titled "Packets 3363 Egressing a LISP Site" to describe the implications when two or 3364 more ITRs exist at a site where only one ITR is used for egress 3365 traffic and when there is a shift of traffic to the others, how 3366 the map-cache will need to be populated in those new egress ITRs. 3368 o Tracker item 33. Make more clear in the Routing Locator Selection 3369 section what an ITR should do when it sees an R-bit of 0 in a 3370 locator-record of a Map-Reply. 3372 o Tracker item 33. Add paragraph to the EID Reachability section 3373 indicating that site partitioning is under investigation. 3375 o Tracker item 58. Added last paragraph of Security Considerations 3376 section about how to protect inner header EID address spoofing 3377 attacks. 3379 o Add suggested Sam text to indicate that all security concerns need 3380 not be addressed for moving document to Experimental RFC status. 3381 Put this in a subsection of the Security Considerations section. 3383 B.6. Changes to draft-ietf-lisp-11.txt 3385 o Posted March 30, 2011. 3387 o Change IANA URL. The URL we had pointed to a general protocol 3388 numbers page. 3390 o Added the "s" bit to the Map-Request to allow SMR-invoked Map- 3391 Requests to be sent to a MN ETR via the map-server. 3393 o Generalize text for the definition of Reencapsuatling tunnels. 3395 o Add paragraph suggested by Joel to explain how implementation 3396 experimentation will be used to determine the proper cache 3397 management techniques. 3399 o Add Yakov provided text for the definition of "EID-to-RLOC 3400 "Database". 3402 o Add reference in Section 8, Deployment Scenarios, to the 3403 draft-jakab-lisp-deploy-02.txt draft. 3405 o Clarify sentence about no hardware changes needed to support LISP 3406 encapsulation. 3408 o Add paragraph about what is the procedure when a locator is 3409 inserted in the middle of a locator-set. 3411 o Add a definition for Locator Status Bits so we can emphasize they 3412 are used as a hint for router up/down status and not path 3413 reachability. 3415 o Change "BGP RIB" to "RIB" per Clarence's comment. 3417 o Fixed complaints by IDnits. 3419 o Add subsection to Security Considerations section indicating how 3420 EID-prefix overclaiming in Map-Replies is for further study and 3421 add a reference to LISP-SEC. 3423 B.7. Changes to draft-ietf-lisp-10.txt 3425 o Posted March 2011. 3427 o Add p-bit to Map-Request so there is documentary reasons to know 3428 when a PITR has sent a Map-Request to an ETR. 3430 o Add Map-Notify message which is used to acknowledge a Map-Register 3431 message sent to a Map-Server. 3433 o Add M-bit to the Map-Register message so an ETR that wants an 3434 acknowledgment for the Map-Register can request one. 3436 o Add S-bit to the ECM and Map-Reply messages to describe security 3437 data that can be present in each message. Then refer to 3438 [LISP-SEC] for expansive details. 3440 o Add Network Management Considerations section and point to the MIB 3441 and LIG drafts. 3443 o Remove the word "simple" per Yakov's comments. 3445 B.8. Changes to draft-ietf-lisp-09.txt 3447 o Posted October 2010. 3449 o Add to IANA Consideration section about the use of LCAF Type 3450 values that accepted and maintained by the IANA registry and not 3451 the LCAF specification. 3453 o Indicate that implementations should be able to receive LISP 3454 control messages when either UDP port is 4342, so they can be 3455 robust in the face of intervening NAT boxes. 3457 o Add paragraph to SMR section to indicate that an ITR does not need 3458 to respond to an SMR-based Map-Request when it has no map-cache 3459 entry for the SMR source's EID-prefix. 3461 B.9. Changes to draft-ietf-lisp-08.txt 3463 o Posted August 2010. 3465 o In section 6.1.6, remove statement about setting TTL to 0 in Map- 3466 Register messages. 3468 o Clarify language in section 6.1.5 about Map-Replying to Data- 3469 Probes or Map-Requests. 3471 o Indicate that outer TTL should only be copied to inner TTL when it 3472 is less than inner TTL. 3474 o Indicate a source-EID for RLOC-probes are encoded with an AFI 3475 value of 0. 3477 o Indicate that SMRs can have a global or per SMR destination rate- 3478 limiter. 3480 o Add clarifications to the SMR procedures. 3482 o Add definitions for "client-side" and 'server-side" terms used in 3483 this specification. 3485 o Clear up language in section 6.4, last paragraph. 3487 o Change ACT of value 0 to "no-action". This is so we can RLOC- 3488 probe a PETR and have it return a Map-Reply with a locator-set of 3489 size 0. The way it is spec'ed the map-cache entry has action 3490 "dropped". Drop-action is set to 3. 3492 o Add statement about normalizing locator weights. 3494 o Clarify R-bit definition in the Map-Reply locator record. 3496 o Add section on EID Reachability within a LISP site. 3498 o Clarify another disadvantage of using anycast locators. 3500 o Reworded Abstract. 3502 o Change section 2.0 Introduction to remove obsolete information 3503 such as the LISP variant definitions. 3505 o Change section 5 title from "Tunneling Details" to "LISP 3506 Encapsulation Details". 3508 o Changes to section 5 to include results of network deployment 3509 experience with MTU. Recommend that implementations use either 3510 the stateful or stateless handling. 3512 o Make clarification wordsmithing to Section 7 and 8. 3514 o Identify that if there is one locator in the locator-set of a map- 3515 cache entry, that an SMR from that locator should be responded to 3516 by sending the the SMR-invoked Map-Request to the database mapping 3517 system rather than to the RLOC itself (which may be unreachable). 3519 o When describing Unicast and Multicast Weights indicate the the 3520 values are relative weights rather than percentages. So it 3521 doesn't imply the sum of all locator weights in the locator-set 3522 need to be 100. 3524 o Do some wordsmithing on copying TTL and TOS fields. 3526 o Numerous wordsmithing changes from Dave Meyer. He fine toothed 3527 combed the spec. 3529 o Removed Section 14 "Prototype Plans and Status". We felt this 3530 type of section is no longer appropriate for a protocol 3531 specification. 3533 o Add clarification text for the IRC description per Damien's 3534 commentary. 3536 o Remove text on copying nonce from SMR to SMR-invoked Map- Request 3537 per Vina's comment about a possible DoS vector. 3539 o Clarify (S/2 + H) in the stateless MTU section. 3541 o Add text to reflect Damien's comment about the description of the 3542 "ITR-RLOC Address" field in the Map-Request. that the list of RLOC 3543 addresses are local addresses of the Map-Requester. 3545 B.10. Changes to draft-ietf-lisp-07.txt 3547 o Posted April 2010. 3549 o Added I-bit to data header so LSB field can also be used as an 3550 Instance ID field. When this occurs, the LSB field is reduced to 3551 8-bits (from 32-bits). 3553 o Added V-bit to the data header so the 24-bit nonce field can also 3554 be used for source and destination version numbers. 3556 o Added Map-Version 12-bit value to the EID-record to be used in all 3557 of Map-Request, Map-Reply, and Map-Register messages. 3559 o Added multiple ITR-RLOC fields to the Map-Request packet so an ETR 3560 can decide what address to select for the destination of a Map- 3561 Reply. 3563 o Added L-bit (Local RLOC bit) and p-bit (Probe-Reply RLOC bit) to 3564 the Locator-Set record of an EID-record for a Map-Reply message. 3565 The L-bit indicates which RLOCs in the locator-set are local to 3566 the sender of the message. The P-bit indicates which RLOC is the 3567 source of a RLOC-probe Reply (Map-Reply) message. 3569 o Add reference to the LISP Canonical Address Format [LCAF] draft. 3571 o Made editorial and clarification changes based on comments from 3572 Dhirendra Trivedi. 3574 o Added wordsmithing comments from Joel Halpern on DF=1 setting. 3576 o Add John Zwiebel clarification to Echo Nonce Algorithm section 3577 6.3.1. 3579 o Add John Zwiebel comment about expanding on proxy-map-reply bit 3580 for Map-Register messages. 3582 o Add NAT section per Ron Bonica comments. 3584 o Fix IDnits issues per Ron Bonica. 3586 o Added section on Virtualization and Segmentation to explain the 3587 use if the Instance ID field in the data header. 3589 o There are too many P-bits, keep their scope to the packet format 3590 description and refer to them by name every where else in the 3591 spec. 3593 o Scanned all occurrences of "should", "should not", "must" and 3594 "must not" and uppercased them. 3596 o John Zwiebel offered text for section 4.1 to modernize the 3597 example. Thanks Z! 3599 o Make it more clear in the definition of "EID-to-RLOC Database" 3600 that all ETRs need to have the same database mapping. This 3601 reflects a comment from John Scudder. 3603 o Add a definition "Route-returnability" to the Definition of Terms 3604 section. 3606 o In section 9.2, add text to describe what the signature of 3607 traceroute packets can look like. 3609 o Removed references to Data Probe for introductory example. Data- 3610 probes are still part of the LISP design but not encouraged. 3612 o Added the definition for "LISP site" to the Definition of Terms" 3613 section. 3615 B.11. Changes to draft-ietf-lisp-06.txt 3617 Editorial based changes: 3619 o Posted December 2009. 3621 o Fix typo for flags in LISP data header. Changed from "4" to "5". 3623 o Add text to indicate that Map-Register messages must contain a 3624 computed UDP checksum. 3626 o Add definitions for PITR and PETR. 3628 o Indicate an AFI value of 0 is an unspecified address. 3630 o Indicate that the TTL field of a Map-Register is not used and set 3631 to 0 by the sender. This change makes this spec consistent with 3632 [LISP-MS]. 3634 o Change "... yield a packet size of L bytes" to "... yield a packet 3635 size greater than L bytes". 3637 o Clarify section 6.1.5 on what addresses and ports are used in Map- 3638 Reply messages. 3640 o Clarify that LSBs that go beyond the number of locators do not to 3641 be SMRed when the locator addresses are greater lexicographically 3642 than the locator in the existing locator-set. 3644 o Add Gregg, Srini, and Amit to acknowledgment section. 3646 o Clarify in the definition of a LISP header what is following the 3647 UDP header. 3649 o Clarify "verifying Map-Request" text in section 6.1.3. 3651 o Add Xu Xiaohu to the acknowledgment section for introducing the 3652 problem of overlapping EID-prefixes among multiple sites in an RRG 3653 email message. 3655 Design based changes: 3657 o Use stronger language to have the outer IPv4 header set DF=1 so we 3658 can avoid fragment reassembly in an ETR or PETR. This will also 3659 make IPv4 and IPv6 encapsulation have consistent behavior. 3661 o Map-Requests should not be sent in ECM with the Probe bit is set. 3662 These type of Map-Requests are used as RLOC-probes and are sent 3663 directly to locator addresses in the underlying network. 3665 o Add text in section 6.1.5 about returning all EID-prefixes in a 3666 Map-Reply sent by an ETR when there are overlapping EID-prefixes 3667 configure. 3669 o Add text in a new subsection of section 6.1.5 about dealing with 3670 Map-Replies with coarse EID-prefixes. 3672 B.12. Changes to draft-ietf-lisp-05.txt 3674 o Posted September 2009. 3676 o Added this Document Change Log appendix. 3678 o Added section indicating that encapsulated Map-Requests must use 3679 destination UDP port 4342. 3681 o Don't use AH in Map-Registers. Put key-id, auth-length, and auth- 3682 data in Map-Register payload. 3684 o Added Jari to acknowledgment section. 3686 o State the source-EID is set to 0 when using Map-Requests to 3687 refresh or RLOC-probe. 3689 o Make more clear what source-RLOC should be for a Map-Request. 3691 o The LISP-CONS authors thought that the Type definitions for CONS 3692 should be removed from this specification. 3694 o Removed nonce from Map-Register message, it wasn't used so no need 3695 for it. 3697 o Clarify what to do for unspecified Action bits for negative Map- 3698 Replies. Since No Action is a drop, make value 0 Drop. 3700 B.13. Changes to draft-ietf-lisp-04.txt 3702 o Posted September 2009. 3704 o How do deal with record count greater than 1 for a Map-Request. 3705 Damien and Joel comment. Joel suggests: 1) Specify that senders 3706 compliant with the current document will always set the count to 3707 1, and note that the count is included for future extensibility. 3708 2) Specify what a receiver compliant with the draft should do if 3709 it receives a request with a count greater than 1. Presumably, it 3710 should send some error back? 3712 o Add Fred Templin in acknowledgment section. 3714 o Add Margaret and Sam to the acknowledgment section for their great 3715 comments. 3717 o Say more about LAGs in the UDP section per Sam Hartman's comment. 3719 o Sam wants to use MAY instead of SHOULD for ignoring checksums on 3720 ETR. From the mailing list: "You'd need to word it as an ITR MAY 3721 send a zero checksum, an ETR MUST accept a 0 checksum and MAY 3722 ignore the checksum completely. And of course we'd need to 3723 confirm that can actually be implemented. In particular, hardware 3724 that verifies UDP checksums on receive needs to be checked to make 3725 sure it permits 0 checksums." 3727 o Margaret wants a reference to 3728 http://www.ietf.org/id/draft-eubanks-chimento-6man-00.txt. 3730 o Fix description in Map-Request section. Where we describe Map- 3731 Reply Record, change "R-bit" to "M-bit". 3733 o Add the mobility bit to Map-Replies. So PTRs don't probe so often 3734 for MNs but often enough to get mapping updates. 3736 o Indicate SHA1 can be used as well for Map-Registers. 3738 o More Fred comments on MTU handling. 3740 o Isidor comment about spec'ing better periodic Map-Registers. Will 3741 be fixed in draft-ietf-lisp-ms-02.txt. 3743 o Margaret's comment on gleaning: "The current specification does 3744 not make it clear how long gleaned map entries should be retained 3745 in the cache, nor does it make it clear how/ when they will be 3746 validated. The LISP spec should, at the very least, include a 3747 (short) default lifetime for gleaned entries, require that they be 3748 validated within a short period of time, and state that a new 3749 gleaned entry should never overwrite an entry that was obtained 3750 from the mapping system. The security implications of storing 3751 "gleaned" entries should also be explored in detail." 3753 o Add section on RLOC-probing per working group feedback. 3755 o Change "loc-reach-bits" to "loc-status-bits" per comment from 3756 Noel. 3758 o Remove SMR-bit from data-plane. Dino prefers to have it in the 3759 control plane only. 3761 o Change LISP header to allow a "Research Bit" so the Nonce and LSB 3762 fields can be turned off and used for another future purpose. For 3763 Luigi et al versioning convergence. 3765 o Add a N-bit to the data header suggested by Noel. Then the nonce 3766 field could be used when N is not 1. 3768 o Clarify that when E-bit is 0, the nonce field can be an echoed 3769 nonce or a random nonce. Comment from Jesper. 3771 o Indicate when doing data-gleaning that a verifying Map-Request is 3772 sent to the source-EID of the gleaned data packet so we can avoid 3773 map-cache corruption by a 3rd party. Comment from Pedro. 3775 o Indicate that a verifying Map-Request, for accepting mapping data, 3776 should be sent over the ALT (or to the EID). 3778 o Reference IPsec RFC 4302. Comment from Sam and Brian Weis. 3780 o Put E-bit in Map-Reply to tell ITRs that the ETR supports echo- 3781 noncing. Comment by Pedro and Dino. 3783 o Jesper made a comment to loosen the language about requiring the 3784 copy of inner TTL to outer TTL since the text to get mixed-AF 3785 traceroute to work would violate the "MUST" clause. Changed from 3786 MUST to SHOULD in section 5.3. 3788 B.14. Changes to draft-ietf-lisp-03.txt 3790 o Posted July 2009. 3792 o Removed loc-reach-bits longword from control packets per Damien 3793 comment. 3795 o Clarifications in MTU text from Roque. 3797 o Added text to indicate that the locator-set be sorted by locator 3798 address from Isidor. 3800 o Clarification text from John Zwiebel in Echo-Nonce section. 3802 B.15. Changes to draft-ietf-lisp-02.txt 3804 o Posted July 2009. 3806 o Encapsulation packet format change to add E-bit and make loc- 3807 reach-bits 32-bits in length. 3809 o Added Echo-Nonce Algorithm section. 3811 o Clarification how ECN bits are copied. 3813 o Moved S-bit in Map-Request. 3815 o Added P-bit in Map-Request and Map-Reply messages to anticipate 3816 RLOC-Probe Algorithm. 3818 o Added to Mobility section to reference [LISP-MN]. 3820 B.16. Changes to draft-ietf-lisp-01.txt 3822 o Posted 2 days after draft-ietf-lisp-00.txt in May 2009. 3824 o Defined LEID to be a "LISP EID". 3826 o Indicate encapsulation use IPv4 DF=0. 3828 o Added negative Map-Reply messages with drop, native-forward, and 3829 send-map-request actions. 3831 o Added Proxy-Map-Reply bit to Map-Register. 3833 B.17. Changes to draft-ietf-lisp-00.txt 3835 o Posted May 2009. 3837 o Rename of draft-farinacci-lisp-12.txt. 3839 o Acknowledgment to RRG. 3841 Authors' Addresses 3843 Dino Farinacci 3844 cisco Systems 3845 Tasman Drive 3846 San Jose, CA 95134 3847 USA 3849 Email: dino@cisco.com 3851 Vince Fuller 3852 cisco Systems 3853 Tasman Drive 3854 San Jose, CA 95134 3855 USA 3857 Email: vaf@cisco.com 3859 Dave Meyer 3860 cisco Systems 3861 170 Tasman Drive 3862 San Jose, CA 3863 USA 3865 Email: dmm@cisco.com 3867 Darrel Lewis 3868 cisco Systems 3869 170 Tasman Drive 3870 San Jose, CA 3871 USA 3873 Email: darlewis@cisco.com