idnits 2.17.1 draft-ietf-lisp-rfc6830bis-00.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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (December 6, 2016) is 2691 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC5496' is defined on line 2154, but no explicit reference was found in the text == Unused Reference: 'RFC6115' is defined on line 2163, but no explicit reference was found in the text == Unused Reference: 'RFC6833' is defined on line 2171, but no explicit reference was found in the text == Unused Reference: 'RFC6836' is defined on line 2179, but no explicit reference was found in the text == Unused Reference: 'BGP-SEC' is defined on line 2193, but no explicit reference was found in the text == Unused Reference: 'CONS' is defined on line 2200, but no explicit reference was found in the text == Unused Reference: 'EMACS' is defined on line 2205, but no explicit reference was found in the text == Unused Reference: 'LISP-DDT' is defined on line 2225, but no explicit reference was found in the text == Unused Reference: 'RADIR' is defined on line 2259, but no explicit reference was found in the text == Unused Reference: 'RFC4107' is defined on line 2281, but no explicit reference was found in the text == Unused Reference: 'RFC6480' is defined on line 2300, but no explicit reference was found in the text == Unused Reference: 'RFC6518' is defined on line 2304, but no explicit reference was found in the text == Unused Reference: 'RFC6837' is defined on line 2320, but no explicit reference was found in the text ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Downref: Normative reference to an Informational RFC: RFC 3232 ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) ** Downref: Normative reference to an Informational RFC: RFC 6115 ** Obsolete normative reference: RFC 6833 (Obsoleted by RFC 9301) ** Obsolete normative reference: RFC 6834 (Obsoleted by RFC 9302) ** Downref: Normative reference to an Experimental RFC: RFC 6836 -- Duplicate reference: RFC6834, mentioned in 'RFC7215', was also mentioned in 'RFC6834'. ** Obsolete normative reference: RFC 6834 (ref. 'RFC7215') (Obsoleted by RFC 9302) Summary: 8 errors (**), 0 flaws (~~), 14 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Farinacci 3 Internet-Draft V. Fuller 4 Intended status: Standards Track D. Meyer 5 Expires: June 9, 2017 D. Lewis 6 Cisco Systems 7 A. Cabellos (Ed.) 8 UPC/BarcelonaTech 9 December 6, 2016 11 The Locator/ID Separation Protocol (LISP) 12 draft-ietf-lisp-rfc6830bis-00 14 Abstract 16 This document describes the Locator/ID Separation Protocol (LISP) 17 data-plane encapsulation protocol. LISP defines two namespaces, End- 18 point Identifiers (EIDs) that identify end-hosts and Routing Locators 19 (RLOCs) that identify network attachment points. With this, LISP 20 effectively separates control from data, and allows routers to create 21 overlay networks. LISP-capable routers exchange encapsulated packets 22 according to EID-to-RLOC mappings stored in a local map-cache. The 23 map-cache is populated by the LISP Control-Plane protocol 24 [REF_TO_RFC6833bis]. 26 LISP requires no change to either host protocol stacks or to underlay 27 routers and offers Traffic Engineering, multihoming and mobility, 28 among other features. 30 Status of This Memo 32 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. 35 Internet-Drafts are working documents of the Internet Engineering 36 Task Force (IETF). Note that other groups may also distribute 37 working documents as Internet-Drafts. The list of current Internet- 38 Drafts is at http://datatracker.ietf.org/drafts/current/. 40 Internet-Drafts are draft documents valid for a maximum of six months 41 and may be updated, replaced, or obsoleted by other documents at any 42 time. It is inappropriate to use Internet-Drafts as reference 43 material or to cite them other than as "work in progress." 45 This Internet-Draft will expire on June 9, 2017. 47 Copyright Notice 49 Copyright (c) 2016 IETF Trust and the persons identified as the 50 document authors. All rights reserved. 52 This document is subject to BCP 78 and the IETF Trust's Legal 53 Provisions Relating to IETF Documents 54 (http://trustee.ietf.org/license-info) in effect on the date of 55 publication of this document. Please review these documents 56 carefully, as they describe your rights and restrictions with respect 57 to this document. Code Components extracted from this document must 58 include Simplified BSD License text as described in Section 4.e of 59 the Trust Legal Provisions and are provided without warranty as 60 described in the Simplified BSD License. 62 Table of Contents 64 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 65 2. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 66 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 4 67 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . 9 68 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . 11 69 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . 13 70 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 14 71 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 15 72 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . 16 73 6. LISP EID-to-RLOC Map-Cache . . . . . . . . . . . . . . . . . 20 74 7. Dealing with Large Encapsulated Packets . . . . . . . . . . . 20 75 7.1. A Stateless Solution to MTU Handling . . . . . . . . . . 21 76 7.2. A Stateful Solution to MTU Handling . . . . . . . . . . . 22 77 8. Using Virtualization and Segmentation with LISP . . . . . . . 22 78 9. Routing Locator Selection . . . . . . . . . . . . . . . . . . 23 79 10. Routing Locator Reachability . . . . . . . . . . . . . . . . 24 80 10.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . . 27 81 10.2. RLOC-Probing Algorithm . . . . . . . . . . . . . . . . . 28 82 11. EID Reachability within a LISP Site . . . . . . . . . . . . . 29 83 12. Routing Locator Hashing . . . . . . . . . . . . . . . . . . . 29 84 13. Changing the Contents of EID-to-RLOC Mappings . . . . . . . . 30 85 13.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . . 31 86 13.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . . 32 87 13.3. Database Map-Versioning . . . . . . . . . . . . . . . . 33 88 14. Multicast Considerations . . . . . . . . . . . . . . . . . . 34 89 15. Router Performance Considerations . . . . . . . . . . . . . . 35 90 16. Mobility Considerations . . . . . . . . . . . . . . . . . . . 36 91 16.1. Slow Mobility . . . . . . . . . . . . . . . . . . . . . 36 92 16.2. Fast Mobility . . . . . . . . . . . . . . . . . . . . . 36 93 16.3. LISP Mobile Node Mobility . . . . . . . . . . . . . . . 37 94 17. LISP xTR Placement and Encapsulation Methods . . . . . . . . 37 95 17.1. First-Hop/Last-Hop xTRs . . . . . . . . . . . . . . . . 39 96 17.2. Border/Edge xTRs . . . . . . . . . . . . . . . . . . . . 39 97 17.3. ISP Provider Edge (PE) xTRs . . . . . . . . . . . . . . 40 98 17.4. LISP Functionality with Conventional NATs . . . . . . . 40 99 17.5. Packets Egressing a LISP Site . . . . . . . . . . . . . 40 100 18. Traceroute Considerations . . . . . . . . . . . . . . . . . . 41 101 18.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . 42 102 18.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . 42 103 18.3. Traceroute Using Mixed Locators . . . . . . . . . . . . 43 104 19. Security Considerations . . . . . . . . . . . . . . . . . . . 43 105 20. Network Management Considerations . . . . . . . . . . . . . . 43 106 21. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 43 107 21.1. LISP ACT and Flag Fields . . . . . . . . . . . . . . . . 44 108 21.2. LISP Address Type Codes . . . . . . . . . . . . . . . . 44 109 21.3. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . 44 110 21.4. LISP Key ID Numbers . . . . . . . . . . . . . . . . . . 44 111 22. References . . . . . . . . . . . . . . . . . . . . . . . . . 45 112 22.1. Normative References . . . . . . . . . . . . . . . . . . 45 113 22.2. Informative References . . . . . . . . . . . . . . . . . 47 114 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 51 115 Appendix B. Document Change Log . . . . . . . . . . . . . . . . 51 116 B.1. Changes to draft-ietf-lisp-rfc6830bis-00 . . . . . . . . 52 117 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 52 119 1. Introduction 121 This document describes the Locator/Identifier Separation Protocol 122 (LISP). LISP is an encapsulation protocol built around the 123 fundamental idea of separating the topological location of a network 124 attachment point from the node's identity [CHIAPPA]. As a result 125 LISP creates two namespaces: Endpoint Identifiers (EIDs), that are 126 used to identify end-hosts (e.g., nodes or Virtual Machines) and 127 routable Routing Locators (RLOCs), used to identify network 128 attachment points. LISP then defines functions for mapping between 129 the two numbering spaces and for encapsulating traffic originated by 130 devices using non-routable EIDs for transport across a network 131 infrastructure that routes and forwards using RLOCs. 133 LISP is an overlay protocol that separates control from data-plane, 134 this document specifies the data-plane, how LISP-capable routers 135 (Tunnel Routers) exchange packets by encapsulating them to the 136 appropriate location. Tunnel routers are equipped with a cache, 137 called map-cache, that contains EID-to-RLOC mappings. The map-cache 138 is populated using the LISP Control-Plane protocol [REF_to_6833bis]. 140 LISP does not require changes to either host protocol stack or to 141 underlay routers. By separating the EID from the RLOC space, LISP 142 offers native Traffic Engineering, multihoming and mobility, among 143 other features. 145 Creation of LISP was initially motivated by discussions during the 146 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 147 October 2006 (see [RFC4984]) 149 This document specifies the LISP data-plane encapsulation and other 150 xTR functionality while [REF_to_6833bis] specifies the LISP control 151 plane. LISP deployment guidelines can be found in [RFC7215] and 152 [RFC6835] describes considerations for network operational 153 management. Finally, [LISP-INTRO] describes the LISP architecture. 155 2. Requirements Notation 157 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 158 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 159 document are to be interpreted as described in [RFC2119]. 161 3. Definition of Terms 163 Provider-Independent (PI) Addresses: PI addresses are an address 164 block assigned from a pool where blocks are not associated with 165 any particular location in the network (e.g., from a particular 166 service provider) and are therefore not topologically aggregatable 167 in the routing system. 169 Provider-Assigned (PA) Addresses: PA addresses are an address block 170 assigned to a site by each service provider to which a site 171 connects. Typically, each block is a sub-block of a service 172 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 173 is aggregated into the larger block before being advertised into 174 the global Internet. Traditionally, IP multihoming has been 175 implemented by each multihomed site acquiring its own globally 176 visible prefix. LISP uses only topologically assigned and 177 aggregatable address blocks for RLOCs, eliminating this 178 demonstrably non-scalable practice. 180 Routing Locator (RLOC): An RLOC is an IPv4 [RFC0791] or IPv6 181 [RFC2460] address of an Egress Tunnel Router (ETR). An RLOC is 182 the output of an EID-to-RLOC mapping lookup. An EID maps to one 183 or more RLOCs. Typically, RLOCs are numbered from topologically 184 aggregatable blocks that are assigned to a site at each point to 185 which it attaches to the global Internet; where the topology is 186 defined by the connectivity of provider networks, RLOCs can be 187 thought of as PA addresses. Multiple RLOCs can be assigned to the 188 same ETR device or to multiple ETR devices at a site. 190 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 191 IPv6) value used in the source and destination address fields of 192 the first (most inner) LISP header of a packet. The host obtains 193 a destination EID the same way it obtains a destination address 194 today, for example, through a Domain Name System (DNS) [RFC1034] 195 lookup or Session Initiation Protocol (SIP) [RFC3261] exchange. 196 The source EID is obtained via existing mechanisms used to set a 197 host's "local" IP address. An EID used on the public Internet 198 must have the same properties as any other IP address used in that 199 manner; this means, among other things, that it must be globally 200 unique. An EID is allocated to a host from an EID-Prefix block 201 associated with the site where the host is located. An EID can be 202 used by a host to refer to other hosts. EIDs MUST NOT be used as 203 LISP RLOCs. Note that EID blocks MAY be assigned in a 204 hierarchical manner, independent of the network topology, to 205 facilitate scaling of the mapping database. In addition, an EID 206 block assigned to a site may have site-local structure 207 (subnetting) for routing within the site; this structure is not 208 visible to the global routing system. In theory, the bit string 209 that represents an EID for one device can represent an RLOC for a 210 different device. As the architecture is realized, if a given bit 211 string is both an RLOC and an EID, it must refer to the same 212 entity in both cases. When used in discussions with other 213 Locator/ID separation proposals, a LISP EID will be called an 214 "LEID". Throughout this document, any references to "EID" refer 215 to an LEID. 217 EID-Prefix: An EID-Prefix is a power-of-two block of EIDs that are 218 allocated to a site by an address allocation authority. EID- 219 Prefixes are associated with a set of RLOC addresses that make up 220 a "database mapping". EID-Prefix allocations can be broken up 221 into smaller blocks when an RLOC set is to be associated with the 222 larger EID-Prefix block. A globally routed address block (whether 223 PI or PA) is not inherently an EID-Prefix. A globally routed 224 address block MAY be used by its assignee as an EID block. The 225 converse is not supported. That is, a site that receives an 226 explicitly allocated EID-Prefix may not use that EID-Prefix as a 227 globally routed prefix. This would require coordination and 228 cooperation with the entities managing the mapping infrastructure. 229 Once this has been done, that block could be removed from the 230 globally routed IP system, if other suitable transition and access 231 mechanisms are in place. Discussion of such transition and access 232 mechanisms can be found in [RFC6832] and [RFC7215]. 234 End-system: An end-system is an IPv4 or IPv6 device that originates 235 packets with a single IPv4 or IPv6 header. The end-system 236 supplies an EID value for the destination address field of the IP 237 header when communicating globally (i.e., outside of its routing 238 domain). An end-system can be a host computer, a switch or router 239 device, or any network appliance. 241 Ingress Tunnel Router (ITR): An ITR is a router that resides in a 242 LISP site. Packets sent by sources inside of the LISP site to 243 destinations outside of the site are candidates for encapsulation 244 by the ITR. The ITR treats the IP destination address as an EID 245 and performs an EID-to-RLOC mapping lookup. The router then 246 prepends an "outer" IP header with one of its globally routable 247 RLOCs in the source address field and the result of the mapping 248 lookup in the destination address field. Note that this 249 destination RLOC MAY be an intermediate, proxy device that has 250 better knowledge of the EID-to-RLOC mapping closer to the 251 destination EID. In general, an ITR receives IP packets from site 252 end-systems on one side and sends LISP-encapsulated IP packets 253 toward the Internet on the other side. 255 Specifically, when a service provider prepends a LISP header for 256 Traffic Engineering purposes, the router that does this is also 257 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 258 on the outer destination address (the originating ITR's supplied 259 RLOC) or the inner destination address (the originating host's 260 supplied EID). 262 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 263 network that prepends an additional LISP header for Traffic 264 Engineering purposes. 266 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 267 packet where the destination address in the "outer" IP header is 268 one of its own RLOCs. The router strips the "outer" header and 269 forwards the packet based on the next IP header found. In 270 general, an ETR receives LISP-encapsulated IP packets from the 271 Internet on one side and sends decapsulated IP packets to site 272 end-systems on the other side. ETR functionality does not have to 273 be limited to a router device. A server host can be the endpoint 274 of a LISP tunnel as well. 276 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 277 network that strips an outer LISP header for Traffic Engineering 278 purposes. 280 xTR: An xTR is a reference to an ITR or ETR when direction of data 281 flow is not part of the context description. "xTR" refers to the 282 router that is the tunnel endpoint and is used synonymously with 283 the term "Tunnel Router". For example, "An xTR can be located at 284 the Customer Edge (CE) router" indicates both ITR and ETR 285 functionality at the CE router. 287 LISP Router: A LISP router is a router that performs the functions 288 of any or all of the following: ITR, ETR, Proxy-ITR (PITR), or 289 Proxy-ETR (PETR). 291 EID-to-RLOC Map-Cache: The EID-to-RLOC map-cache is a short-lived, 292 on-demand table in an ITR that stores, tracks, and is responsible 293 for timing out and otherwise validating EID-to-RLOC mappings. 294 This cache is distinct from the full "database" of EID-to-RLOC 295 mappings; it is dynamic, local to the ITR(s), and relatively 296 small, while the database is distributed, relatively static, and 297 much more global in scope. 299 EID-to-RLOC Database: The EID-to-RLOC Database is a global 300 distributed database that contains all known EID-Prefix-to-RLOC 301 mappings. Each potential ETR typically contains a small piece of 302 the database: the EID-to-RLOC mappings for the EID-Prefixes 303 "behind" the router. These map to one of the router's own 304 globally visible IP addresses. The same database mapping entries 305 MUST be configured on all ETRs for a given site. In a steady 306 state, the EID-Prefixes for the site and the Locator-Set for each 307 EID-Prefix MUST be the same on all ETRs. Procedures to enforce 308 and/or verify this are outside the scope of this document. Note 309 that there MAY be transient conditions when the EID-Prefix for the 310 site and Locator-Set for each EID-Prefix may not be the same on 311 all ETRs. This has no negative implications, since a partial set 312 of Locators can be used. 314 Recursive Tunneling: Recursive Tunneling occurs when a packet has 315 more than one LISP IP header. Additional layers of tunneling MAY 316 be employed to implement Traffic Engineering or other re-routing 317 as needed. When this is done, an additional "outer" LISP header 318 is added, and the original RLOCs are preserved in the "inner" 319 header. Any references to tunnels in this specification refer to 320 dynamic encapsulating tunnels; they are never statically 321 configured. 323 Re-encapsulating Tunnels: Re-encapsulating Tunneling occurs when an 324 ETR removes a LISP header, then acts as an ITR to prepend another 325 LISP header. Doing this allows a packet to be re-routed by the 326 re-encapsulating router without adding the overhead of additional 327 tunnel headers. Any references to tunnels in this specification 328 refer to dynamic encapsulating tunnels; they are never statically 329 configured. When using multiple mapping database systems, care 330 must be taken to not create re-encapsulation loops through 331 misconfiguration. 333 LISP Header: LISP header is a term used in this document to refer 334 to the outer IPv4 or IPv6 header, a UDP header, and a LISP- 335 specific 8-octet header that follow the UDP header and that an ITR 336 prepends or an ETR strips. 338 Address Family Identifier (AFI): AFI is a term used to describe an 339 address encoding in a packet. An address family currently 340 pertains to an IPv4 or IPv6 address. See [AFI] and [RFC3232] for 341 details. An AFI value of 0 used in this specification indicates 342 an unspecified encoded address where the length of the address is 343 0 octets following the 16-bit AFI value of 0. 345 Negative Mapping Entry: A negative mapping entry, also known as a 346 negative cache entry, is an EID-to-RLOC entry where an EID-Prefix 347 is advertised or stored with no RLOCs. That is, the Locator-Set 348 for the EID-to-RLOC entry is empty or has an encoded Locator count 349 of 0. This type of entry could be used to describe a prefix from 350 a non-LISP site, which is explicitly not in the mapping database. 351 There are a set of well-defined actions that are encoded in a 352 Negative Map-Reply. 354 Data-Probe: A Data-Probe is a LISP-encapsulated data packet where 355 the inner-header destination address equals the outer-header 356 destination address used to trigger a Map-Reply by a decapsulating 357 ETR. In addition, the original packet is decapsulated and 358 delivered to the destination host if the destination EID is in the 359 EID-Prefix range configured on the ETR. Otherwise, the packet is 360 discarded. A Data-Probe is used in some of the mapping database 361 designs to "probe" or request a Map-Reply from an ETR; in other 362 cases, Map-Requests are used. See each mapping database design 363 for details. When using Data-Probes, by sending Map-Requests on 364 the underlying routing system, EID-Prefixes must be advertised. 365 However, this is discouraged if the core is to scale by having 366 less EID-Prefixes stored in the core router's routing tables. 368 Proxy-ITR (PITR): A PITR is defined and described in [RFC6832]. A 369 PITR acts like an ITR but does so on behalf of non-LISP sites that 370 send packets to destinations at LISP sites. 372 Proxy-ETR (PETR): A PETR is defined and described in [RFC6832]. A 373 PETR acts like an ETR but does so on behalf of LISP sites that 374 send packets to destinations at non-LISP sites. 376 Route-returnability: Route-returnability is an assumption that the 377 underlying routing system will deliver packets to the destination. 378 When combined with a nonce that is provided by a sender and 379 returned by a receiver, this limits off-path data insertion. A 380 route-returnability check is verified when a message is sent with 381 a nonce, another message is returned with the same nonce, and the 382 destination of the original message appears as the source of the 383 returned message. 385 LISP site: LISP site is a set of routers in an edge network that are 386 under a single technical administration. LISP routers that reside 387 in the edge network are the demarcation points to separate the 388 edge network from the core network. 390 Client-side: Client-side is a term used in this document to indicate 391 a connection initiation attempt by an EID. The ITR(s) at the LISP 392 site are the first to get involved in obtaining database Map-Cache 393 entries by sending Map-Request messages. 395 Server-side: Server-side is a term used in this document to indicate 396 that a connection initiation attempt is being accepted for a 397 destination EID. The ETR(s) at the destination LISP site are the 398 first to send Map-Replies to the source site initiating the 399 connection. The ETR(s) at this destination site can obtain 400 mappings by gleaning information from Map-Requests, Data-Probes, 401 or encapsulated packets. 403 Locator-Status-Bits (LSBs): Locator-Status-Bits are present in the 404 LISP header. They are used by ITRs to inform ETRs about the up/ 405 down status of all ETRs at the local site. These bits are used as 406 a hint to convey up/down router status and not path reachability 407 status. The LSBs can be verified by use of one of the Locator 408 reachability algorithms described in Section 10. 410 Anycast Address: Anycast Address is a term used in this document to 411 refer to the same IPv4 or IPv6 address configured and used on 412 multiple systems at the same time. An EID or RLOC can be an 413 anycast address in each of their own address spaces. 415 4. Basic Overview 417 One key concept of LISP is that end-systems operate the same way they 418 do today. The IP addresses that hosts use for tracking sockets and 419 connections, and for sending and receiving packets, do not change. 420 In LISP terminology, these IP addresses are called Endpoint 421 Identifiers (EIDs). 423 Routers continue to forward packets based on IP destination 424 addresses. When a packet is LISP encapsulated, these addresses are 425 referred to as Routing Locators (RLOCs). Most routers along a path 426 between two hosts will not change; they continue to perform routing/ 427 forwarding lookups on the destination addresses. For routers between 428 the source host and the ITR as well as routers from the ETR to the 429 destination host, the destination address is an EID. For the routers 430 between the ITR and the ETR, the destination address is an RLOC. 432 Another key LISP concept is the "Tunnel Router". A Tunnel Router 433 prepends LISP headers on host-originated packets and strips them 434 prior to final delivery to their destination. The IP addresses in 435 this "outer header" are RLOCs. During end-to-end packet exchange 436 between two Internet hosts, an ITR prepends a new LISP header to each 437 packet, and an ETR strips the new header. The ITR performs EID-to- 438 RLOC lookups to determine the routing path to the ETR, which has the 439 RLOC as one of its IP addresses. 441 Some basic rules governing LISP are: 443 o End-systems only send to addresses that are EIDs. They don't know 444 that addresses are EIDs versus RLOCs but assume that packets get 445 to their intended destinations. In a system where LISP is 446 deployed, LISP routers intercept EID-addressed packets and assist 447 in delivering them across the network core where EIDs cannot be 448 routed. The procedure a host uses to send IP packets does not 449 change. 451 o EIDs are typically IP addresses assigned to hosts. 453 o Other types of EID are supported by LISP, see [LCAF] for further 454 information. 456 o LISP routers mostly deal with Routing Locator addresses. See 457 details in Section 4.1 to clarify what is meant by "mostly". 459 o RLOCs are always IP addresses assigned to routers, preferably 460 topologically oriented addresses from provider CIDR (Classless 461 Inter-Domain Routing) blocks. 463 o When a router originates packets, it may use as a source address 464 either an EID or RLOC. When acting as a host (e.g., when 465 terminating a transport session such as Secure SHell (SSH), 466 TELNET, or the Simple Network Management Protocol (SNMP)), it may 467 use an EID that is explicitly assigned for that purpose. An EID 468 that identifies the router as a host MUST NOT be used as an RLOC; 469 an EID is only routable within the scope of a site. A typical BGP 470 configuration might demonstrate this "hybrid" EID/RLOC usage where 471 a router could use its "host-like" EID to terminate iBGP sessions 472 to other routers in a site while at the same time using RLOCs to 473 terminate eBGP sessions to routers outside the site. 475 o Packets with EIDs in them are not expected to be delivered end-to- 476 end in the absence of an EID-to-RLOC mapping operation. They are 477 expected to be used locally for intra-site communication or to be 478 encapsulated for inter-site communication. 480 o EID-Prefixes are likely to be hierarchically assigned in a manner 481 that is optimized for administrative convenience and to facilitate 482 scaling of the EID-to-RLOC mapping database. The hierarchy is 483 based on an address allocation hierarchy that is independent of 484 the network topology. 486 o EIDs may also be structured (subnetted) in a manner suitable for 487 local routing within an Autonomous System (AS). 489 An additional LISP header MAY be prepended to packets by a TE-ITR 490 when re-routing of the path for a packet is desired. A potential 491 use-case for this would be an ISP router that needs to perform 492 Traffic Engineering for packets flowing through its network. In such 493 a situation, termed "Recursive Tunneling", an ISP transit acts as an 494 additional ITR, and the RLOC it uses for the new prepended header 495 would be either a TE-ETR within the ISP (along an intra-ISP traffic 496 engineered path) or a TE-ETR within another ISP (an inter-ISP traffic 497 engineered path, where an agreement to build such a path exists). 499 In order to avoid excessive packet overhead as well as possible 500 encapsulation loops, this document mandates that a maximum of two 501 LISP headers can be prepended to a packet. For initial LISP 502 deployments, it is assumed that two headers is sufficient, where the 503 first prepended header is used at a site for Location/Identity 504 separation and the second prepended header is used inside a service 505 provider for Traffic Engineering purposes. 507 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 508 For example, the ITR for a particular end-to-end packet exchange 509 might be the first-hop or default router within a site for the source 510 host. Similarly, the ETR might be the last-hop router directly 511 connected to the destination host. Another example, perhaps for a 512 VPN service outsourced to an ISP by a site, the ITR could be the 513 site's border router at the service provider attachment point. 514 Mixing and matching of site-operated, ISP-operated, and other Tunnel 515 Routers is allowed for maximum flexibility. 517 4.1. Packet Flow Sequence 519 This section provides an example of the unicast packet flow, 520 including also control-plane information as specified in 521 [REF_TO_RFC6833bis]. The example also assumes the following 522 conditions: 524 o Source host "host1.abc.example.com" is sending a packet to 525 "host2.xyz.example.com", exactly what host1 would do if the site 526 was not using LISP. 528 o Each site is multihomed, so each Tunnel Router has an address 529 (RLOC) assigned from the service provider address block for each 530 provider to which that particular Tunnel Router is attached. 532 o The ITR(s) and ETR(s) are directly connected to the source and 533 destination, respectively, but the source and destination can be 534 located anywhere in the LISP site. 536 o Map-Requests are sent to the mapping database system by using the 537 LISP control-plane protocol documented in [REF_to_RFC6833bis]. A 538 Map-Request is sent for an external destination when the 539 destination is not found in the forwarding table or matches a 540 default route. 542 o Map-Replies are sent on the underlying routing system topology 543 using the [REF_TO_RFC6833bis] control-plane protocol. 545 Client host1.abc.example.com wants to communicate with server 546 host2.xyz.example.com: 548 1. host1.abc.example.com wants to open a TCP connection to 549 host2.xyz.example.com. It does a DNS lookup on 550 host2.xyz.example.com. An A/AAAA record is returned. This 551 address is the destination EID. The locally assigned address of 552 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 553 packet is built and forwarded through the LISP site as a normal 554 IP packet until it reaches a LISP ITR. 556 2. The LISP ITR must be able to map the destination EID to an RLOC 557 of one of the ETRs at the destination site. The specific method 558 used to do this is not described in this example. See 559 [REF_to_RFC6833bis] for further information. 561 3. The ITR sends a LISP Map-Request as specified in 562 [REF_TO_RFC6833bis]. Map-Requests SHOULD be rate-limited. 564 4. The mapping system helps forwarding the Map-Request to the 565 corresponding ETR. When the Map-Request arrives at one of the 566 ETRs at the destination site, it will process the packet as a 567 control message. 569 5. The ETR looks at the destination EID of the Map-Request and 570 matches it against the prefixes in the ETR's configured EID-to- 571 RLOC mapping database. This is the list of EID-Prefixes the ETR 572 is supporting for the site it resides in. If there is no match, 573 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 574 returned to the ITR. 576 6. The ITR receives the Map-Reply message, parses the message (to 577 check for format validity), and stores the mapping information 578 from the packet. This information is stored in the ITR's EID-to- 579 RLOC map-cache. Note that the map-cache is an on-demand cache. 580 An ITR will manage its map-cache in such a way that optimizes for 581 its resource constraints. 583 7. Subsequent packets from host1.abc.example.com to 584 host2.xyz.example.com will have a LISP header prepended by the 585 ITR using the appropriate RLOC as the LISP header destination 586 address learned from the ETR. Note that the packet MAY be sent 587 to a different ETR than the one that returned the Map-Reply due 588 to the source site's hashing policy or the destination site's 589 Locator-Set policy. 591 8. The ETR receives these packets directly (since the destination 592 address is one of its assigned IP addresses), checks the validity 593 of the addresses, strips the LISP header, and forwards packets to 594 the attached destination host. 596 9. In order to defer the need for a mapping lookup in the reverse 597 direction, an ETR can OPTIONALLY create a cache entry that maps 598 the source EID (inner-header source IP address) to the source 599 RLOC (outer-header source IP address) in a received LISP packet. 600 Such a cache entry is termed a "gleaned" mapping and only 601 contains a single RLOC for the EID in question. More complete 602 information about additional RLOCs SHOULD be verified by sending 603 a LISP Map-Request for that EID. Both the ITR and the ETR may 604 also influence the decision the other makes in selecting an RLOC. 606 5. LISP Encapsulation Details 608 Since additional tunnel headers are prepended, the packet becomes 609 larger and can exceed the MTU of any link traversed from the ITR to 610 the ETR. It is RECOMMENDED in IPv4 that packets do not get 611 fragmented as they are encapsulated by the ITR. Instead, the packet 612 is dropped and an ICMP Too Big message is returned to the source. 614 This specification RECOMMENDS that implementations provide support 615 for one of the proposed fragmentation and reassembly schemes. Two 616 existing schemes are detailed in Section 7. 618 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 619 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 620 header is in IPv4 packet format and the outer header is in IPv6 621 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 622 is in IPv6 packet format and the outer header is in IPv4 packet 623 format). The next sub-sections illustrate packet formats for the 624 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6), but all 4 625 combinations MUST be supported. Additional types of EIDs are defined 626 in [LCAF]. 628 5.1. LISP IPv4-in-IPv4 Header Format 630 0 1 2 3 631 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 1 632 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 633 / |Version| IHL |Type of Service| Total Length | 634 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 635 | | Identification |Flags| Fragment Offset | 636 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 637 OH | Time to Live | Protocol = 17 | Header Checksum | 638 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 639 | | Source Routing Locator | 640 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 641 \ | Destination Routing Locator | 642 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 643 / | Source Port = xxxx | Dest Port = 4341 | 644 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 645 \ | UDP Length | UDP Checksum | 646 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 647 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 648 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 649 S / | Instance ID/Locator-Status-Bits | 650 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 651 / |Version| IHL |Type of Service| Total Length | 652 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 653 | | Identification |Flags| Fragment Offset | 654 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 655 IH | Time to Live | Protocol | Header Checksum | 656 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 657 | | Source EID | 658 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 659 \ | Destination EID | 660 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 662 IHL = IP-Header-Length 664 5.2. LISP IPv6-in-IPv6 Header Format 666 0 1 2 3 667 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 1 668 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 669 / |Version| Traffic Class | Flow Label | 670 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 671 | | Payload Length | Next Header=17| Hop Limit | 672 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 673 | | 674 O + + 675 u | | 676 t + Source Routing Locator + 677 e | | 678 r + + 679 | | 680 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 681 d | | 682 r + + 683 | | 684 ^ + Destination Routing Locator + 685 | | | 686 \ + + 687 \ | | 688 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 689 / | Source Port = xxxx | Dest Port = 4341 | 690 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 691 \ | UDP Length | UDP Checksum | 692 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 693 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 694 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 695 S / | Instance ID/Locator-Status-Bits | 696 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 697 / |Version| Traffic Class | Flow Label | 698 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 699 / | Payload Length | Next Header | Hop Limit | 700 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 701 | | 702 I + + 703 n | | 704 n + Source EID + 705 e | | 706 r + + 707 | | 708 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 709 d | | 710 r + + 711 | | 713 ^ + Destination EID + 714 \ | | 715 \ + + 716 \ | | 717 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 719 5.3. Tunnel Header Field Descriptions 721 Inner Header (IH): The inner header is the header on the datagram 722 received from the originating host. The source and destination IP 723 addresses are EIDs [RFC0791] [RFC2460]. 725 Outer Header: (OH) The outer header is a new header prepended by an 726 ITR. The address fields contain RLOCs obtained from the ingress 727 router's EID-to-RLOC Cache. The IP protocol number is "UDP (17)" 728 from [RFC0768]. The setting of the Don't Fragment (DF) bit 729 'Flags' field is according to rules listed in Sections 7.1 and 730 7.2. 732 UDP Header: The UDP header contains an ITR selected source port when 733 encapsulating a packet. See Section 12 for details on the hash 734 algorithm used to select a source port based on the 5-tuple of the 735 inner header. The destination port MUST be set to the well-known 736 IANA-assigned port value 4341. 738 UDP Checksum: The 'UDP Checksum' field SHOULD be transmitted as zero 739 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 740 [UDP-TUNNELS] [UDP-ZERO]. When a packet with a zero UDP checksum 741 is received by an ETR, the ETR MUST accept the packet for 742 decapsulation. When an ITR transmits a non-zero value for the UDP 743 checksum, it MUST send a correctly computed value in this field. 744 When an ETR receives a packet with a non-zero UDP checksum, it MAY 745 choose to verify the checksum value. If it chooses to perform 746 such verification, and the verification fails, the packet MUST be 747 silently dropped. If the ETR chooses not to perform the 748 verification, or performs the verification successfully, the 749 packet MUST be accepted for decapsulation. The handling of UDP 750 checksums for all tunneling protocols, including LISP, is under 751 active discussion within the IETF. When that discussion 752 concludes, any necessary changes will be made to align LISP with 753 the outcome of the broader discussion. 755 UDP Length: The 'UDP Length' field is set for an IPv4-encapsulated 756 packet to be the sum of the inner-header IPv4 Total Length plus 757 the UDP and LISP header lengths. For an IPv6-encapsulated packet, 758 the 'UDP Length' field is the sum of the inner-header IPv6 Payload 759 Length, the size of the IPv6 header (40 octets), and the size of 760 the UDP and LISP headers. 762 N: The N-bit is the nonce-present bit. When this bit is set to 1, 763 the low-order 24 bits of the first 32 bits of the LISP header 764 contain a Nonce. See Section 10.1 for details. Both N- and 765 V-bits MUST NOT be set in the same packet. If they are, a 766 decapsulating ETR MUST treat the 'Nonce/Map-Version' field as 767 having a Nonce value present. 769 L: The L-bit is the 'Locator-Status-Bits' field enabled bit. When 770 this bit is set to 1, the Locator-Status-Bits in the second 771 32 bits of the LISP header are in use. 773 x 1 x x 0 x x x 774 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 775 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 776 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 777 | Locator-Status-Bits | 778 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 780 E: The E-bit is the echo-nonce-request bit. This bit MUST be ignored 781 and has no meaning when the N-bit is set to 0. When the N-bit is 782 set to 1 and this bit is set to 1, an ITR is requesting that the 783 nonce value in the 'Nonce' field be echoed back in LISP- 784 encapsulated packets when the ITR is also an ETR. See 785 Section 10.1 for details. 787 V: The V-bit is the Map-Version present bit. When this bit is set to 788 1, the N-bit MUST be 0. Refer to Section 13.3 for more details. 789 This bit indicates that the LISP header is encoded in this 790 case as: 792 0 x 0 1 x x x x 793 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 794 |N|L|E|V|I|R|K|K| Source Map-Version | Dest Map-Version | 795 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 796 | Instance ID/Locator-Status-Bits | 797 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 799 I: The I-bit is the Instance ID bit. See Section 8 for more details. 800 When this bit is set to 1, the 'Locator-Status-Bits' field is 801 reduced to 8 bits and the high-order 24 bits are used as an 802 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 803 are transmitted as zero and ignored on receipt. The format of the 804 LISP header would look like this: 806 x x x x 1 x x x 807 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 808 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 809 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 810 | Instance ID | LSBs | 811 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 813 R: The R-bit is a Reserved bit for future use. It MUST be set to 0 814 on transmit and MUST be ignored on receipt. 816 KK: The KK-bits are a 2-bit field used when encapsualted packets are 817 encrypted. The field is set to 00 when the packet is not 818 encrypted. See [I.D-ietf-lisp-crypto] for further information. 820 LISP Nonce: The LISP 'Nonce' field is a 24-bit value that is 821 randomly generated by an ITR when the N-bit is set to 1. Nonce 822 generation algorithms are an implementation matter but are 823 required to generate different nonces when sending to different 824 destinations. However, the same nonce can be used for a period of 825 time to the same destination. The nonce is also used when the 826 E-bit is set to request the nonce value to be echoed by the other 827 side when packets are returned. When the E-bit is clear but the 828 N-bit is set, a remote ITR is either echoing a previously 829 requested echo-nonce or providing a random nonce. See 830 Section 10.1 for more details. 832 LISP Locator-Status-Bits (LSBs): When the L-bit is also set, the 833 'Locator-Status-Bits' field in the LISP header is set by an ITR to 834 indicate to an ETR the up/down status of the Locators in the 835 source site. Each RLOC in a Map-Reply is assigned an ordinal 836 value from 0 to n-1 (when there are n RLOCs in a mapping entry). 837 The Locator-Status-Bits are numbered from 0 to n-1 from the least 838 significant bit of the field. The field is 32 bits when the I-bit 839 is set to 0 and is 8 bits when the I-bit is set to 1. When a 840 Locator-Status-Bit is set to 1, the ITR is indicating to the ETR 841 that the RLOC associated with the bit ordinal has up status. See 842 Section 10 for details on how an ITR can determine the status of 843 the ETRs at the same site. When a site has multiple EID-Prefixes 844 that result in multiple mappings (where each could have a 845 different Locator-Set), the Locator-Status-Bits setting in an 846 encapsulated packet MUST reflect the mapping for the EID-Prefix 847 that the inner-header source EID address matches. If the LSB for 848 an anycast Locator is set to 1, then there is at least one RLOC 849 with that address, and the ETR is considered 'up'. 851 When doing ITR/PITR encapsulation: 853 o The outer-header 'Time to Live' field (or 'Hop Limit' field, in 854 the case of IPv6) SHOULD be copied from the inner-header 'Time to 855 Live' field. 857 o The outer-header 'Type of Service' field (or the 'Traffic Class' 858 field, in the case of IPv6) SHOULD be copied from the inner-header 859 'Type of Service' field (with one exception; see below). 861 When doing ETR/PETR decapsulation: 863 o The inner-header 'Time to Live' field (or 'Hop Limit' field, in 864 the case of IPv6) SHOULD be copied from the outer-header 'Time to 865 Live' field, when the Time to Live value of the outer header is 866 less than the Time to Live value of the inner header. Failing to 867 perform this check can cause the Time to Live of the inner header 868 to increment across encapsulation/decapsulation cycles. This 869 check is also performed when doing initial encapsulation, when a 870 packet comes to an ITR or PITR destined for a LISP site. 872 o The inner-header 'Type of Service' field (or the 'Traffic Class' 873 field, in the case of IPv6) SHOULD be copied from the outer-header 874 'Type of Service' field (with one exception; see below). 876 Note that if an ETR/PETR is also an ITR/PITR and chooses to re- 877 encapsulate after decapsulating, the net effect of this is that the 878 new outer header will carry the same Time to Live as the old outer 879 header minus 1. 881 Copying the Time to Live (TTL) serves two purposes: first, it 882 preserves the distance the host intended the packet to travel; 883 second, and more importantly, it provides for suppression of looping 884 packets in the event there is a loop of concatenated tunnels due to 885 misconfiguration. See Section 18.3 for TTL exception handling for 886 traceroute packets. 888 The Explicit Congestion Notification ('ECN') field occupies bits 6 889 and 7 of both the IPv4 'Type of Service' field and the IPv6 'Traffic 890 Class' field [RFC3168]. The 'ECN' field requires special treatment 891 in order to avoid discarding indications of congestion [RFC3168]. 892 ITR encapsulation MUST copy the 2-bit 'ECN' field from the inner 893 header to the outer header. Re-encapsulation MUST copy the 2-bit 894 'ECN' field from the stripped outer header to the new outer header. 895 If the 'ECN' field contains a congestion indication codepoint (the 896 value is '11', the Congestion Experienced (CE) codepoint), then ETR 897 decapsulation MUST copy the 2-bit 'ECN' field from the stripped outer 898 header to the surviving inner header that is used to forward the 899 packet beyond the ETR. These requirements preserve CE indications 900 when a packet that uses ECN traverses a LISP tunnel and becomes 901 marked with a CE indication due to congestion between the tunnel 902 endpoints. 904 6. LISP EID-to-RLOC Map-Cache 906 ITRs and PITRs maintain an on-demand cache, referred as LISP EID-to- 907 RLOC Map-Cache, that contains mappings from EID-prefixes to locator 908 sets. The cache is used to encapsulate packets from the EID space to 909 the corresponding RLOC network attachment point. 911 When an ITR/PITR receives a packet from inside of the LISP site to 912 destinations outside of the site a longest-prefix match lookup of the 913 EID is done to the map-cache. 915 When the lookup succeeds, the locator-set retrieved from the map- 916 cache is used to send the packet to the EID's topological location. 918 If the lookup fails, the ITR/PITR needs to retrieve the mapping using 919 the LISP control-plane protocol [REF_TO_RFC6833bis]. The mapping is 920 then stored in the local map-cache to forward subsequent packets 921 addressed to the same EID-prefix. 923 The map-cache is a local cache of mappings, entries are expired based 924 on the associated Time to live. In addition, entries can be updated 925 with more current information, see Section 13 for further information 926 on this. Finally, the map-cache also contains reachability 927 information about EIDs and RLOCs, and uses LISP reachability 928 information mechanisms to determine the reachability of RLOCs, see 929 Section 10 for the specific mechanisms. 931 7. Dealing with Large Encapsulated Packets 933 This section proposes two mechanisms to deal with packets that exceed 934 the path MTU between the ITR and ETR. 936 It is left to the implementor to decide if the stateless or stateful 937 mechanism should be implemented. Both or neither can be used, since 938 it is a local decision in the ITR regarding how to deal with MTU 939 issues, and sites can interoperate with differing mechanisms. 941 Both stateless and stateful mechanisms also apply to Re-encapsulating 942 and Recursive Tunneling, so any actions below referring to an ITR 943 also apply to a TE-ITR. 945 7.1. A Stateless Solution to MTU Handling 947 An ITR stateless solution to handle MTU issues is described as 948 follows: 950 1. Define H to be the size, in octets, of the outer header an ITR 951 prepends to a packet. This includes the UDP and LISP header 952 lengths. 954 2. Define L to be the size, in octets, of the maximum-sized packet 955 an ITR can send to an ETR without the need for the ITR or any 956 intermediate routers to fragment the packet. 958 3. Define an architectural constant S for the maximum size of a 959 packet, in octets, an ITR must receive from the source so the 960 effective MTU can be met. That is, L = S + H. 962 When an ITR receives a packet from a site-facing interface and adds H 963 octets worth of encapsulation to yield a packet size greater than L 964 octets (meaning the received packet size was greater than S octets 965 from the source), it resolves the MTU issue by first splitting the 966 original packet into 2 equal-sized fragments. A LISP header is then 967 prepended to each fragment. The size of the encapsulated fragments 968 is then (S/2 + H), which is less than the ITR's estimate of the path 969 MTU between the ITR and its correspondent ETR. 971 When an ETR receives encapsulated fragments, it treats them as two 972 individually encapsulated packets. It strips the LISP headers and 973 then forwards each fragment to the destination host of the 974 destination site. The two fragments are reassembled at the 975 destination host into the single IP datagram that was originated by 976 the source host. Note that reassembly can happen at the ETR if the 977 encapsulated packet was fragmented at or after the ITR. 979 This behavior is performed by the ITR when the source host originates 980 a packet with the 'DF' field of the IP header set to 0. When the 981 'DF' field of the IP header is set to 1, or the packet is an IPv6 982 packet originated by the source host, the ITR will drop the packet 983 when the size is greater than L and send an ICMP Too Big message to 984 the source with a value of S, where S is (L - H). 986 When the outer-header encapsulation uses an IPv4 header, an 987 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 988 can be avoided. An implementation MAY set the DF bit in such headers 989 to 0 if it has good reason to believe there are unresolvable path MTU 990 issues between the sending ITR and the receiving ETR. 992 This specification RECOMMENDS that L be defined as 1500. 994 7.2. A Stateful Solution to MTU Handling 996 An ITR stateful solution to handle MTU issues is described as follows 997 and was first introduced in [OPENLISP]: 999 1. The ITR will keep state of the effective MTU for each Locator per 1000 Map-Cache entry. The effective MTU is what the core network can 1001 deliver along the path between the ITR and ETR. 1003 2. When an IPv6-encapsulated packet, or an IPv4-encapsulated packet 1004 with the DF bit set to 1, exceeds what the core network can 1005 deliver, one of the intermediate routers on the path will send an 1006 ICMP Too Big message to the ITR. The ITR will parse the ICMP 1007 message to determine which Locator is affected by the effective 1008 MTU change and then record the new effective MTU value in the 1009 Map-Cache entry. 1011 3. When a packet is received by the ITR from a source inside of the 1012 site and the size of the packet is greater than the effective MTU 1013 stored with the Map-Cache entry associated with the destination 1014 EID the packet is for, the ITR will send an ICMP Too Big message 1015 back to the source. The packet size advertised by the ITR in the 1016 ICMP Too Big message is the effective MTU minus the LISP 1017 encapsulation length. 1019 Even though this mechanism is stateful, it has advantages over the 1020 stateless IP fragmentation mechanism, by not involving the 1021 destination host with reassembly of ITR fragmented packets. 1023 8. Using Virtualization and Segmentation with LISP 1025 FIXME: Indicate how the instance-ID is 32 bits in control-plane and 1026 24-bits in the data-plane. 1028 When multiple organizations inside of a LISP site are using private 1029 addresses [RFC1918] as EID-Prefixes, their address spaces MUST remain 1030 segregated due to possible address duplication. An Instance ID in 1031 the address encoding can aid in making the entire AFI-based address 1032 unique. See IANA Considerations (Section 21.2) for details on 1033 possible address encodings. 1035 An Instance ID can be carried in a LISP-encapsulated packet. An ITR 1036 that prepends a LISP header will copy a 24-bit value used by the LISP 1037 router to uniquely identify the address space. The value is copied 1038 to the 'Instance ID' field of the LISP header, and the I-bit is set 1039 to 1. 1041 When an ETR decapsulates a packet, the Instance ID from the LISP 1042 header is used as a table identifier to locate the forwarding table 1043 to use for the inner destination EID lookup. 1045 For example, an 802.1Q VLAN tag or VPN identifier could be used as a 1046 24-bit Instance ID. 1048 9. Routing Locator Selection 1050 Both the client-side and server-side may need control over the 1051 selection of RLOCs for conversations between them. This control is 1052 achieved by manipulating the 'Priority' and 'Weight' fields in EID- 1053 to-RLOC Map-Reply messages. Alternatively, RLOC information MAY be 1054 gleaned from received tunneled packets or EID-to-RLOC Map-Request 1055 messages. 1057 The following are different scenarios for choosing RLOCs and the 1058 controls that are available: 1060 o The server-side returns one RLOC. The client-side can only use 1061 one RLOC. The server-side has complete control of the selection. 1063 o The server-side returns a list of RLOCs where a subset of the list 1064 has the same best Priority. The client can only use the subset 1065 list according to the weighting assigned by the server-side. In 1066 this case, the server-side controls both the subset list and load- 1067 splitting across its members. The client-side can use RLOCs 1068 outside of the subset list if it determines that the subset list 1069 is unreachable (unless RLOCs are set to a Priority of 255). Some 1070 sharing of control exists: the server-side determines the 1071 destination RLOC list and load distribution while the client-side 1072 has the option of using alternatives to this list if RLOCs in the 1073 list are unreachable. 1075 o The server-side sets a Weight of 0 for the RLOC subset list. In 1076 this case, the client-side can choose how the traffic load is 1077 spread across the subset list. Control is shared by the server- 1078 side determining the list and the client determining load 1079 distribution. Again, the client can use alternative RLOCs if the 1080 server-provided list of RLOCs is unreachable. 1082 o Either side (more likely the server-side ETR) decides not to send 1083 a Map-Request. For example, if the server-side ETR does not send 1084 Map-Requests, it gleans RLOCs from the client-side ITR, giving the 1085 client-side ITR responsibility for bidirectional RLOC reachability 1086 and preferability. Server-side ETR gleaning of the client-side 1087 ITR RLOC is done by caching the inner-header source EID and the 1088 outer-header source RLOC of received packets. The client-side ITR 1089 controls how traffic is returned and can alternate using an outer- 1090 header source RLOC, which then can be added to the list the 1091 server-side ETR uses to return traffic. Since no Priority or 1092 Weights are provided using this method, the server-side ETR MUST 1093 assume that each client-side ITR RLOC uses the same best Priority 1094 with a Weight of zero. In addition, since EID-Prefix encoding 1095 cannot be conveyed in data packets, the EID-to-RLOC Cache on 1096 Tunnel Routers can grow to be very large. 1098 o A "gleaned" Map-Cache entry, one learned from the source RLOC of a 1099 received encapsulated packet, is only stored and used for a few 1100 seconds, pending verification. Verification is performed by 1101 sending a Map-Request to the source EID (the inner-header IP 1102 source address) of the received encapsulated packet. A reply to 1103 this "verifying Map-Request" is used to fully populate the Map- 1104 Cache entry for the "gleaned" EID and is stored and used for the 1105 time indicated from the 'TTL' field of a received Map-Reply. When 1106 a verified Map-Cache entry is stored, data gleaning no longer 1107 occurs for subsequent packets that have a source EID that matches 1108 the EID-Prefix of the verified entry. This "gleaning" mechanism 1109 is OPTIONAL. 1111 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1112 reachable when the R-bit for the Locator record is set to 1. When 1113 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1114 RLOC. Neither the information contained in a Map-Reply nor that 1115 stored in the mapping database system provides reachability 1116 information for RLOCs. Note that reachability is not part of the 1117 mapping system and is determined using one or more of the Routing 1118 Locator reachability algorithms described in the next section. 1120 10. Routing Locator Reachability 1122 Several mechanisms for determining RLOC reachability are currently 1123 defined: 1125 1. An ETR may examine the Locator-Status-Bits in the LISP header of 1126 an encapsulated data packet received from an ITR. If the ETR is 1127 also acting as an ITR and has traffic to return to the original 1128 ITR site, it can use this status information to help select an 1129 RLOC. 1131 2. An ITR may receive an ICMP Network Unreachable or Host 1132 Unreachable message for an RLOC it is using. This indicates that 1133 the RLOC is likely down. Note that trusting ICMP messages may 1134 not be desirable, but neither is ignoring them completely. 1135 Implementations are encouraged to follow current best practices 1136 in treating these conditions. 1138 3. An ITR that participates in the global routing system can 1139 determine that an RLOC is down if no BGP Routing Information Base 1140 (RIB) route exists that matches the RLOC IP address. 1142 4. An ITR may receive an ICMP Port Unreachable message from a 1143 destination host. This occurs if an ITR attempts to use 1144 interworking [RFC6832] and LISP-encapsulated data is sent to a 1145 non-LISP-capable site. 1147 5. An ITR may receive a Map-Reply from an ETR in response to a 1148 previously sent Map-Request. The RLOC source of the Map-Reply is 1149 likely up, since the ETR was able to send the Map-Reply to the 1150 ITR. 1152 6. When an ETR receives an encapsulated packet from an ITR, the 1153 source RLOC from the outer header of the packet is likely up. 1155 7. An ITR/ETR pair can use the Locator reachability algorithms 1156 described in this section, namely Echo-Noncing or RLOC-Probing. 1158 When determining Locator up/down reachability by examining the 1159 Locator-Status-Bits from the LISP-encapsulated data packet, an ETR 1160 will receive up-to-date status from an encapsulating ITR about 1161 reachability for all ETRs at the site. CE-based ITRs at the source 1162 site can determine reachability relative to each other using the site 1163 IGP as follows: 1165 o Under normal circumstances, each ITR will advertise a default 1166 route into the site IGP. 1168 o If an ITR fails or if the upstream link to its PE fails, its 1169 default route will either time out or be withdrawn. 1171 Each ITR can thus observe the presence or lack of a default route 1172 originated by the others to determine the Locator-Status-Bits it sets 1173 for them. 1175 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1176 Locator-Status-Bits in a LISP-encapsulated packet are numbered from 0 1177 to n-1 starting with the least significant bit. For example, if an 1178 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1179 value 2), then all ITRs at the site will clear the 3rd least 1180 significant bit (xxxx x0xx) of the 'Locator-Status-Bits' field for 1181 the packets they encapsulate. 1183 When an ETR decapsulates a packet, it will check for any change in 1184 the 'Locator-Status-Bits' field. When a bit goes from 1 to 0, the 1185 ETR, if acting also as an ITR, will refrain from encapsulating 1186 packets to an RLOC that is indicated as down. It will only resume 1187 using that RLOC if the corresponding Locator-Status-Bit returns to a 1188 value of 1. Locator-Status-Bits are associated with a Locator-Set 1189 per EID-Prefix. Therefore, when a Locator becomes unreachable, the 1190 Locator-Status-Bit that corresponds to that Locator's position in the 1191 list returned by the last Map-Reply will be set to zero for that 1192 particular EID-Prefix. 1194 When ITRs at the site are not deployed in CE routers, the IGP can 1195 still be used to determine the reachability of Locators, provided 1196 they are injected into the IGP. This is typically done when a /32 1197 address is configured on a loopback interface. 1199 When ITRs receive ICMP Network Unreachable or Host Unreachable 1200 messages as a method to determine unreachability, they will refrain 1201 from using Locators that are described in Locator lists of Map- 1202 Replies. However, using this approach is unreliable because many 1203 network operators turn off generation of ICMP Destination Unreachable 1204 messages. 1206 If an ITR does receive an ICMP Network Unreachable or Host 1207 Unreachable message, it MAY originate its own ICMP Destination 1208 Unreachable message destined for the host that originated the data 1209 packet the ITR encapsulated. 1211 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1212 locator address from a Locator-Set in a mapping entry matches a 1213 prefix. If it does not find one and BGP is running in the Default- 1214 Free Zone (DFZ), it can decide to not use the Locator even though the 1215 Locator-Status-Bits indicate that the Locator is up. In this case, 1216 the path from the ITR to the ETR that is assigned the Locator is not 1217 available. More details are in [LOC-ID-ARCH]. 1219 Optionally, an ITR can send a Map-Request to a Locator, and if a Map- 1220 Reply is returned, reachability of the Locator has been determined. 1221 Obviously, sending such probes increases the number of control 1222 messages originated by Tunnel Routers for active flows, so Locators 1223 are assumed to be reachable when they are advertised. 1225 This assumption does create a dependency: Locator unreachability is 1226 detected by the receipt of ICMP Host Unreachable messages. When a 1227 Locator has been determined to be unreachable, it is not used for 1228 active traffic; this is the same as if it were listed in a Map-Reply 1229 with Priority 255. 1231 The ITR can test the reachability of the unreachable Locator by 1232 sending periodic Requests. Both Requests and Replies MUST be rate- 1233 limited. Locator reachability testing is never done with data 1234 packets, since that increases the risk of packet loss for end-to-end 1235 sessions. 1237 When an ETR decapsulates a packet, it knows that it is reachable from 1238 the encapsulating ITR because that is how the packet arrived. In 1239 most cases, the ETR can also reach the ITR but cannot assume this to 1240 be true, due to the possibility of path asymmetry. In the presence 1241 of unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD 1242 NOT use the lack of return traffic as an indication that the ETR is 1243 unreachable. Instead, it MUST use an alternate mechanism to 1244 determine reachability. 1246 10.1. Echo Nonce Algorithm 1248 When data flows bidirectionally between Locators from different 1249 sites, a data-plane mechanism called "nonce echoing" can be used to 1250 determine reachability between an ITR and ETR. When an ITR wants to 1251 solicit a nonce echo, it sets the N- and E-bits and places a 24-bit 1252 nonce [RFC4086] in the LISP header of the next encapsulated data 1253 packet. 1255 When this packet is received by the ETR, the encapsulated packet is 1256 forwarded as normal. When the ETR next sends a data packet to the 1257 ITR, it includes the nonce received earlier with the N-bit set and 1258 E-bit cleared. The ITR sees this "echoed nonce" and knows that the 1259 path to and from the ETR is up. 1261 The ITR will set the E-bit and N-bit for every packet it sends while 1262 in the echo-nonce-request state. The time the ITR waits to process 1263 the echoed nonce before it determines the path is unreachable is 1264 variable and is a choice left for the implementation. 1266 If the ITR is receiving packets from the ETR but does not see the 1267 nonce echoed while being in the echo-nonce-request state, then the 1268 path to the ETR is unreachable. This decision may be overridden by 1269 other Locator reachability algorithms. Once the ITR determines that 1270 the path to the ETR is down, it can switch to another Locator for 1271 that EID-Prefix. 1273 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 1274 be implementing both ITR and ETR functionality for the echo nonce 1275 mechanism to operate. 1277 The ITR and ETR may both go into the echo-nonce-request state at the 1278 same time. The number of packets sent or the time during which echo 1279 nonce requests are sent is an implementation-specific setting. 1280 However, when an ITR is in the echo-nonce-request state, it can echo 1281 the ETR's nonce in the next set of packets that it encapsulates and 1282 subsequently continue sending echo-nonce-request packets. 1284 This mechanism does not completely solve the forward path 1285 reachability problem, as traffic may be unidirectional. That is, the 1286 ETR receiving traffic at a site may not be the same device as an ITR 1287 that transmits traffic from that site, or the site-to-site traffic is 1288 unidirectional so there is no ITR returning traffic. 1290 The echo-nonce algorithm is bilateral. That is, if one side sets the 1291 E-bit and the other side is not enabled for echo-noncing, then the 1292 echoing of the nonce does not occur and the requesting side may 1293 erroneously consider the Locator unreachable. An ITR SHOULD only set 1294 the E-bit in an encapsulated data packet when it knows the ETR is 1295 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 1296 Reply message. 1298 Note that other Locator reachability mechanisms are being researched 1299 and can be used to compliment or even override the echo nonce 1300 algorithm. See the next section for an example of control-plane 1301 probing. 1303 10.2. RLOC-Probing Algorithm 1305 RLOC-Probing is a method that an ITR or PITR can use to determine the 1306 reachability status of one or more Locators that it has cached in a 1307 Map-Cache entry. The probe-bit of the Map-Request and Map-Reply 1308 messages is used for RLOC-Probing. 1310 RLOC-Probing is done in the control plane on a timer basis, where an 1311 ITR or PITR will originate a Map-Request destined to a locator 1312 address from one of its own locator addresses. A Map-Request used as 1313 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or to 1314 the mapping database system as one would when soliciting mapping 1315 data. The EID record encoded in the Map-Request is the EID-Prefix of 1316 the Map-Cache entry cached by the ITR or PITR. The ITR may include a 1317 mapping data record for its own database mapping information that 1318 contains the local EID-Prefixes and RLOCs for its site. RLOC-probes 1319 are sent periodically using a jittered timer interval. 1321 When an ETR receives a Map-Request message with the probe-bit set, it 1322 returns a Map-Reply with the probe-bit set. The source address of 1323 the Map-Reply is set according to the procedure described in 1324 [REF_TO_RFC6833bis]. The Map-Reply SHOULD contain mapping data for 1325 the EID-Prefix contained in the Map-Request. This provides the 1326 opportunity for the ITR or PITR that sent the RLOC-probe to get 1327 mapping updates if there were changes to the ETR's database mapping 1328 entries. 1330 There are advantages and disadvantages of RLOC-Probing. The greatest 1331 benefit of RLOC-Probing is that it can handle many failure scenarios 1332 allowing the ITR to determine when the path to a specific Locator is 1333 reachable or has become unreachable, thus providing a robust 1334 mechanism for switching to using another Locator from the cached 1335 Locator. RLOC-Probing can also provide rough Round-Trip Time (RTT) 1336 estimates between a pair of Locators, which can be useful for network 1337 management purposes as well as for selecting low delay paths. The 1338 major disadvantage of RLOC-Probing is in the number of control 1339 messages required and the amount of bandwidth used to obtain those 1340 benefits, especially if the requirement for failure detection times 1341 is very small. 1343 Continued research and testing will attempt to characterize the 1344 tradeoffs of failure detection times versus message overhead. 1346 11. EID Reachability within a LISP Site 1348 A site may be multihomed using two or more ETRs. The hosts and 1349 infrastructure within a site will be addressed using one or more EID- 1350 Prefixes that are mapped to the RLOCs of the relevant ETRs in the 1351 mapping system. One possible failure mode is for an ETR to lose 1352 reachability to one or more of the EID-Prefixes within its own site. 1353 When this occurs when the ETR sends Map-Replies, it can clear the 1354 R-bit associated with its own Locator. And when the ETR is also an 1355 ITR, it can clear its Locator-Status-Bit in the encapsulation data 1356 header. 1358 It is recognized that there are no simple solutions to the site 1359 partitioning problem because it is hard to know which part of the 1360 EID-Prefix range is partitioned and which Locators can reach any sub- 1361 ranges of the EID-Prefixes. This problem is under investigation with 1362 the expectation that experiments will tell us more. Note that this 1363 is not a new problem introduced by the LISP architecture. The 1364 problem exists today when a multihomed site uses BGP to advertise its 1365 reachability upstream. 1367 12. Routing Locator Hashing 1369 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 1370 a requesting ITR, the Locator-Set for the EID-Prefix may contain 1371 different Priority values for each locator address. When more than 1372 one best Priority Locator exists, the ITR can decide how to load- 1373 share traffic against the corresponding Locators. 1375 The following hash algorithm may be used by an ITR to select a 1376 Locator for a packet destined to an EID for the EID-to-RLOC mapping: 1378 1. Either a source and destination address hash or the traditional 1379 5-tuple hash can be used. The traditional 5-tuple hash includes 1380 the source and destination addresses; source and destination TCP, 1381 UDP, or Stream Control Transmission Protocol (SCTP) port numbers; 1382 and the IP protocol number field or IPv6 next-protocol fields of 1383 a packet that a host originates from within a LISP site. When a 1384 packet is not a TCP, UDP, or SCTP packet, the source and 1385 destination addresses only from the header are used to compute 1386 the hash. 1388 2. Take the hash value and divide it by the number of Locators 1389 stored in the Locator-Set for the EID-to-RLOC mapping. 1391 3. The remainder will yield a value of 0 to "number of Locators 1392 minus 1". Use the remainder to select the Locator in the 1393 Locator-Set. 1395 Note that when a packet is LISP encapsulated, the source port number 1396 in the outer UDP header needs to be set. Selecting a hashed value 1397 allows core routers that are attached to Link Aggregation Groups 1398 (LAGs) to load-split the encapsulated packets across member links of 1399 such LAGs. Otherwise, core routers would see a single flow, since 1400 packets have a source address of the ITR, for packets that are 1401 originated by different EIDs at the source site. A suggested setting 1402 for the source port number computed by an ITR is a 5-tuple hash 1403 function on the inner header, as described above. 1405 Many core router implementations use a 5-tuple hash to decide how to 1406 balance packet load across members of a LAG. The 5-tuple hash 1407 includes the source and destination addresses of the packet and the 1408 source and destination ports when the protocol number in the packet 1409 is TCP or UDP. For this reason, UDP encoding is used for LISP 1410 encapsulation. 1412 13. Changing the Contents of EID-to-RLOC Mappings 1414 Since the LISP architecture uses a caching scheme to retrieve and 1415 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 1416 date mapping is to re-request the mapping. However, the ITRs do not 1417 know when the mappings change, and the ETRs do not keep track of 1418 which ITRs requested its mappings. For scalability reasons, we want 1419 to maintain this approach but need to provide a way for ETRs to 1420 change their mappings and inform the sites that are currently 1421 communicating with the ETR site using such mappings. 1423 When adding a new Locator record in lexicographic order to the end of 1424 a Locator-Set, it is easy to update mappings. We assume that new 1425 mappings will maintain the same Locator ordering as the old mapping 1426 but will just have new Locators appended to the end of the list. So, 1427 some ITRs can have a new mapping while other ITRs have only an old 1428 mapping that is used until they time out. When an ITR has only an 1429 old mapping but detects bits set in the Locator-Status-Bits that 1430 correspond to Locators beyond the list it has cached, it simply 1431 ignores them. However, this can only happen for locator addresses 1432 that are lexicographically greater than the locator addresses in the 1433 existing Locator-Set. 1435 When a Locator record is inserted in the middle of a Locator-Set, to 1436 maintain lexicographic order, the SMR procedure in Section 13.2 is 1437 used to inform ITRs and PITRs of the new Locator-Status-Bit mappings. 1439 When a Locator record is removed from a Locator-Set, ITRs that have 1440 the mapping cached will not use the removed Locator because the xTRs 1441 will set the Locator-Status-Bit to 0. So, even if the Locator is in 1442 the list, it will not be used. For new mapping requests, the xTRs 1443 can set the Locator AFI to 0 (indicating an unspecified address), as 1444 well as setting the corresponding Locator-Status-Bit to 0. This 1445 forces ITRs with old or new mappings to avoid using the removed 1446 Locator. 1448 If many changes occur to a mapping over a long period of time, one 1449 will find empty record slots in the middle of the Locator-Set and new 1450 records appended to the Locator-Set. At some point, it would be 1451 useful to compact the Locator-Set so the Locator-Status-Bit settings 1452 can be efficiently packed. 1454 We propose here three approaches for Locator-Set compaction: one 1455 operational mechanism and two protocol mechanisms. The operational 1456 approach uses a clock sweep method. The protocol approaches use the 1457 concept of Solicit-Map-Requests and Map-Versioning. 1459 13.1. Clock Sweep 1461 The clock sweep approach uses planning in advance and the use of 1462 count-down TTLs to time out mappings that have already been cached. 1463 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So, 1464 there is a 24-hour window to time out old mappings. The following 1465 clock sweep procedure is used: 1467 1. 24 hours before a mapping change is to take effect, a network 1468 administrator configures the ETRs at a site to start the clock 1469 sweep window. 1471 2. During the clock sweep window, ETRs continue to send Map-Reply 1472 messages with the current (unchanged) mapping records. The TTL 1473 for these mappings is set to 1 hour. 1475 3. 24 hours later, all previous cache entries will have timed out, 1476 and any active cache entries will time out within 1 hour. During 1477 this 1-hour window, the ETRs continue to send Map-Reply messages 1478 with the current (unchanged) mapping records with the TTL set to 1479 1 minute. 1481 4. At the end of the 1-hour window, the ETRs will send Map-Reply 1482 messages with the new (changed) mapping records. So, any active 1483 caches can get the new mapping contents right away if not cached, 1484 or in 1 minute if they had the mapping cached. The new mappings 1485 are cached with a TTL equal to the TTL in the Map-Reply. 1487 13.2. Solicit-Map-Request (SMR) 1489 Soliciting a Map-Request is a selective way for ETRs, at the site 1490 where mappings change, to control the rate they receive requests for 1491 Map-Reply messages. SMRs are also used to tell remote ITRs to update 1492 the mappings they have cached. 1494 Since the ETRs don't keep track of remote ITRs that have cached their 1495 mappings, they do not know which ITRs need to have their mappings 1496 updated. As a result, an ETR will solicit Map-Requests (called an 1497 SMR message) from those sites to which it has been sending 1498 encapsulated data for the last minute. In particular, an ETR will 1499 send an SMR to an ITR to which it has recently sent encapsulated 1500 data. 1502 An SMR message is simply a bit set in a Map-Request message. An ITR 1503 or PITR will send a Map-Request when they receive an SMR message. 1504 Both the SMR sender and the Map-Request responder MUST rate-limit 1505 these messages. Rate-limiting can be implemented as a global rate- 1506 limiter or one rate-limiter per SMR destination. 1508 The following procedure shows how an SMR exchange occurs when a site 1509 is doing Locator-Set compaction for an EID-to-RLOC mapping: 1511 1. When the database mappings in an ETR change, the ETRs at the site 1512 begin to send Map-Requests with the SMR bit set for each Locator 1513 in each Map-Cache entry the ETR caches. 1515 2. A remote ITR that receives the SMR message will schedule sending 1516 a Map-Request message to the source locator address of the SMR 1517 message or to the mapping database system. A newly allocated 1518 random nonce is selected, and the EID-Prefix used is the one 1519 copied from the SMR message. If the source Locator is the only 1520 Locator in the cached Locator-Set, the remote ITR SHOULD send a 1521 Map-Request to the database mapping system just in case the 1522 single Locator has changed and may no longer be reachable to 1523 accept the Map-Request. 1525 3. The remote ITR MUST rate-limit the Map-Request until it gets a 1526 Map-Reply while continuing to use the cached mapping. When 1527 Map-Versioning as described in Section 13.3 is used, an SMR 1528 sender can detect if an ITR is using the most up-to-date database 1529 mapping. 1531 4. The ETRs at the site with the changed mapping will reply to the 1532 Map-Request with a Map-Reply message that has a nonce from the 1533 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate- 1534 limited. This is important to avoid Map-Reply implosion. 1536 5. The ETRs at the site with the changed mapping record the fact 1537 that the site that sent the Map-Request has received the new 1538 mapping data in the Map-Cache entry for the remote site so the 1539 Locator-Status-Bits are reflective of the new mapping for packets 1540 going to the remote site. The ETR then stops sending SMR 1541 messages. 1543 Experimentation is in progress to determine the appropriate rate- 1544 limit parameters. 1546 For security reasons, an ITR MUST NOT process unsolicited Map- 1547 Replies. To avoid Map-Cache entry corruption by a third party, a 1548 sender of an SMR-based Map-Request MUST be verified. If an ITR 1549 receives an SMR-based Map-Request and the source is not in the 1550 Locator-Set for the stored Map-Cache entry, then the responding Map- 1551 Request MUST be sent with an EID destination to the mapping database 1552 system. Since the mapping database system is a more secure way to 1553 reach an authoritative ETR, it will deliver the Map-Request to the 1554 authoritative source of the mapping data. 1556 When an ITR receives an SMR-based Map-Request for which it does not 1557 have a cached mapping for the EID in the SMR message, it MAY not send 1558 an SMR-invoked Map-Request. This scenario can occur when an ETR 1559 sends SMR messages to all Locators in the Locator-Set it has stored 1560 in its map-cache but the remote ITRs that receive the SMR may not be 1561 sending packets to the site. There is no point in updating the ITRs 1562 until they need to send, in which case they will send Map-Requests to 1563 obtain a Map-Cache entry. 1565 13.3. Database Map-Versioning 1567 When there is unidirectional packet flow between an ITR and ETR, and 1568 the EID-to-RLOC mappings change on the ETR, it needs to inform the 1569 ITR so encapsulation to a removed Locator can stop and can instead be 1570 started to a new Locator in the Locator-Set. 1572 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 1573 Number. This is known as the Destination Map-Version Number. ITRs 1574 include the Destination Map-Version Number in packets they 1575 encapsulate to the site. When an ETR decapsulates a packet and 1576 detects that the Destination Map-Version Number is less than the 1577 current version for its mapping, the SMR procedure described in 1578 Section 13.2 occurs. 1580 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 1581 Version Number. This is known as the Source Map-Version Number. 1582 When an ETR decapsulates a packet and detects that the Source Map- 1583 Version Number is greater than the last Map-Version Number sent in a 1584 Map-Reply from the ITR's site, the ETR will send a Map-Request to one 1585 of the ETRs for the source site. 1587 A Map-Version Number is used as a sequence number per EID-Prefix, so 1588 values that are greater are considered to be more recent. A value of 1589 0 for the Source Map-Version Number or the Destination Map-Version 1590 Number conveys no versioning information, and an ITR does no 1591 comparison with previously received Map-Version Numbers. 1593 A Map-Version Number can be included in Map-Register messages as 1594 well. This is a good way for the Map-Server to assure that all ETRs 1595 for a site registering to it will be synchronized according to Map- 1596 Version Number. 1598 See [RFC6834] for a more detailed analysis and description of 1599 Database Map-Versioning. 1601 14. Multicast Considerations 1603 A multicast group address, as defined in the original Internet 1604 architecture, is an identifier of a grouping of topologically 1605 independent receiver host locations. The address encoding itself 1606 does not determine the location of the receiver(s). The multicast 1607 routing protocol, and the network-based state the protocol creates, 1608 determine where the receivers are located. 1610 In the context of LISP, a multicast group address is both an EID and 1611 a Routing Locator. Therefore, no specific semantic or action needs 1612 to be taken for a destination address, as it would appear in an IP 1613 header. Therefore, a group address that appears in an inner IP 1614 header built by a source host will be used as the destination EID. 1615 The outer IP header (the destination Routing Locator address), 1616 prepended by a LISP router, can use the same group address as the 1617 destination Routing Locator, use a multicast or unicast Routing 1618 Locator obtained from a Mapping System lookup, or use other means to 1619 determine the group address mapping. 1621 With respect to the source Routing Locator address, the ITR prepends 1622 its own IP address as the source address of the outer IP header. 1623 Just like it would if the destination EID was a unicast address. 1624 This source Routing Locator address, like any other Routing Locator 1625 address, MUST be globally routable. 1627 There are two approaches for LISP-Multicast, one that uses native 1628 multicast routing in the underlay with no support from the Mapping 1629 System and the other that uses only unicast routing in the underlay 1630 with support from the Mapping System. See [RFC6831] and 1631 [I.D-ietf-lisp-signal-free-multicast], respectively, for details. 1632 Details for LISP-Multicast and interworking with non-LISP sites are 1633 described in [RFC6831] and [RFC6832]. 1635 15. Router Performance Considerations 1637 LISP is designed to be very "hardware-based forwarding friendly". A 1638 few implementation techniques can be used to incrementally implement 1639 LISP: 1641 o When a tunnel-encapsulated packet is received by an ETR, the outer 1642 destination address may not be the address of the router. This 1643 makes it challenging for the control plane to get packets from the 1644 hardware. This may be mitigated by creating special Forwarding 1645 Information Base (FIB) entries for the EID-Prefixes of EIDs served 1646 by the ETR (those for which the router provides an RLOC 1647 translation). These FIB entries are marked with a flag indicating 1648 that control-plane processing should be performed. The forwarding 1649 logic of testing for particular IP protocol number values is not 1650 necessary. There are a few proven cases where no changes to 1651 existing deployed hardware were needed to support the LISP data- 1652 plane. 1654 o On an ITR, prepending a new IP header consists of adding more 1655 octets to a MAC rewrite string and prepending the string as part 1656 of the outgoing encapsulation procedure. Routers that support 1657 Generic Routing Encapsulation (GRE) tunneling [RFC2784] or 6to4 1658 tunneling [RFC3056] may already support this action. 1660 o A packet's source address or interface the packet was received on 1661 can be used to select VRF (Virtual Routing/Forwarding). The VRF's 1662 routing table can be used to find EID-to-RLOC mappings. 1664 For performance issues related to map-cache management, see 1665 Section 19. 1667 16. Mobility Considerations 1669 There are several kinds of mobility, of which only some might be of 1670 concern to LISP. Essentially, they are as follows. 1672 16.1. Slow Mobility 1674 A site wishes to change its attachment points to the Internet, and 1675 its LISP Tunnel Routers will have new RLOCs when it changes upstream 1676 providers. Changes in EID-to-RLOC mappings for sites are expected to 1677 be handled by configuration, outside of LISP. 1679 An individual endpoint wishes to move but is not concerned about 1680 maintaining session continuity. Renumbering is involved. LISP can 1681 help with the issues surrounding renumbering [RFC4192] [LISA96] by 1682 decoupling the address space used by a site from the address spaces 1683 used by its ISPs [RFC4984]. 1685 16.2. Fast Mobility 1687 Fast endpoint mobility occurs when an endpoint moves relatively 1688 rapidly, changing its IP-layer network attachment point. Maintenance 1689 of session continuity is a goal. This is where the Mobile IPv4 1690 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used and 1691 primarily where interactions with LISP need to be explored, such as 1692 the mechanisms in [I-D.portoles-lisp-eid-mobility] when the EID moves 1693 but the RLOC is in the network infrastructure. 1695 In LISP, one possibility is to "glean" information. When a packet 1696 arrives, the ETR could examine the EID-to-RLOC mapping and use that 1697 mapping for all outgoing traffic to that EID. It can do this after 1698 performing a route-returnability check, to ensure that the new 1699 network location does have an internal route to that endpoint. 1700 However, this does not cover the case where an ITR (the node assigned 1701 the RLOC) at the mobile-node location has been compromised. 1703 Mobile IP packet exchange is designed for an environment in which all 1704 routing information is disseminated before packets can be forwarded. 1705 In order to allow the Internet to grow to support expected future 1706 use, we are moving to an environment where some information may have 1707 to be obtained after packets are in flight. Modifications to IP 1708 mobility should be considered in order to optimize the behavior of 1709 the overall system. Anything that decreases the number of new EID- 1710 to-RLOC mappings needed when a node moves, or maintains the validity 1711 of an EID-to-RLOC mapping for a longer time, is useful. 1713 In addition to endpoints, a network can be mobile, possibly changing 1714 xTRs. A "network" can be as small as a single router and as large as 1715 a whole site. This is different from site mobility in that it is 1716 fast and possibly short-lived, but different from endpoint mobility 1717 in that a whole prefix is changing RLOCs. However, the mechanisms 1718 are the same, and there is no new overhead in LISP. A map request 1719 for any endpoint will return a binding for the entire mobile prefix. 1721 If mobile networks become a more common occurrence, it may be useful 1722 to revisit the design of the mapping service and allow for dynamic 1723 updates of the database. 1725 The issue of interactions between mobility and LISP needs to be 1726 explored further. Specific improvements to the entire system will 1727 depend on the details of mapping mechanisms. Mapping mechanisms 1728 should be evaluated on how well they support session continuity for 1729 mobile nodes. See [REF_TO_I-D.farinacci-lisp-predictive-rlocs] for 1730 more recent mechanisms which can provide near-zero packet loss during 1731 handoffs. 1733 16.3. LISP Mobile Node Mobility 1735 A mobile device can use the LISP infrastructure to achieve mobility 1736 by implementing the LISP encapsulation and decapsulation functions 1737 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 1738 node" can use topologically independent EID IP addresses that are not 1739 advertised into and do not impose a cost on the global routing 1740 system. These EIDs are maintained at the edges of the mapping system 1741 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to 1742 only the correspondents of the LISP mobile node. 1744 Refer to [LISP-MN] for more details for when the EID and RLOC are co- 1745 located in the roaming node. 1747 17. LISP xTR Placement and Encapsulation Methods 1749 [FIXME: The Authors/Ed. agree on adding further scenarios.] 1751 This section will explore how and where ITRs and ETRs can be placed 1752 in the networkand will discuss the pros and cons of each scenario. 1753 For a more detailed networkd design deployment recommendation, refer 1754 to [RFC7215]. 1756 There are two basic deployment tradeoffs to consider: centralized 1757 versus distributed caches; and flat, Recursive, or Re-encapsulating 1758 Tunneling. When deciding on centralized versus distributed caching, 1759 the following issues should be considered: 1761 o Are the xTRs spread out so that the caches are spread across all 1762 the memories of each router? A centralized cache is when an ITR 1763 keeps a cache for all the EIDs it is encapsulating to. The packet 1764 takes a direct path to the destination Locator. A distributed 1765 cache is when an ITR needs help from other Re-Encapsulating Tunnel 1766 Routers (RTRs) because it does not store all the cache entries for 1767 the EIDs it is encapsulating to. So, the packet takes a path 1768 through RTRs that have a different set of cache entries. 1770 o Should management "touch points" be minimized by only choosing a 1771 few xTRs, just enough for redundancy? 1773 o In general, using more ITRs doesn't increase management load, 1774 since caches are built and stored dynamically. On the other hand, 1775 using more ETRs does require more management, since EID-Prefix-to- 1776 RLOC mappings need to be explicitly configured. 1778 When deciding on flat, Recursive, or Re-Encapsulating Tunneling, the 1779 following issues should be considered: 1781 o Flat tunneling implements a single encapsulation path between the 1782 source site and destination site. This generally offers better 1783 paths between sources and destinations with a single encapsulation 1784 path. 1786 o Recursive Tunneling is when encapsulated traffic is again further 1787 encapsulated in another tunnel, either to implement VPNs or to 1788 perform Traffic Engineering. When doing VPN-based tunneling, the 1789 site has some control, since the site is prepending a new 1790 encapsulation header. In the case of TE-based tunneling, the site 1791 may have control if it is prepending a new tunnel header, but if 1792 the site's ISP is doing the TE, then the site has no control. 1793 Recursive Tunneling generally will result in suboptimal paths but 1794 with the benefit of steering traffic to parts of the network that 1795 have more resources available. 1797 o The technique of Re-Encapsulation ensures that packets only 1798 require one encapsulation header. So, if a packet needs to be re- 1799 routed, it is first decapsulated by the RTR and then Re- 1800 Encapsulated with a new encapsulation header using a new RLOC. 1802 The next sub-sections will examine where xTRs and RTRs can reside in 1803 the network. 1805 17.1. First-Hop/Last-Hop xTRs 1807 By locating xTRs close to hosts, the EID-Prefix set is at the 1808 granularity of an IP subnet. So, at the expense of more EID-Prefix- 1809 to-RLOC sets for the site, the caches in each xTR can remain 1810 relatively small. But caches always depend on the number of non- 1811 aggregated EID destination flows active through these xTRs. 1813 With more xTRs doing encapsulation, the increase in control traffic 1814 grows as well: since the EID granularity is greater, more Map- 1815 Requests and Map-Replies are traveling between more routers. 1817 The advantage of placing the caches and databases at these stub 1818 routers is that the products deployed in this part of the network 1819 have better price-memory ratios than their core router counterparts. 1820 Memory is typically less expensive in these devices, and fewer routes 1821 are stored (only IGP routes). These devices tend to have excess 1822 capacity, both for forwarding and routing states. 1824 LISP functionality can also be deployed in edge switches. These 1825 devices generally have layer-2 ports facing hosts and layer-3 ports 1826 facing the Internet. Spare capacity is also often available in these 1827 devices. 1829 17.2. Border/Edge xTRs 1831 Using Customer Edge (CE) routers for xTR placement allows the EID 1832 space associated with a site to be reachable via a small set of RLOCs 1833 assigned to the CE-based xTRs for that site. 1835 This offers the opposite benefit of the first-hop/last-hop xTR 1836 scenario: the number of mapping entries and network management touch 1837 points is reduced, allowing better scaling. 1839 One disadvantage is that fewer network resources are used to reach 1840 host endpoints, thereby centralizing the point-of-failure domain and 1841 creating network choke points at the CE xTR. 1843 Note that more than one CE xTR at a site can be configured with the 1844 same IP address. In this case, an RLOC is an anycast address. This 1845 allows resilience between the CE xTRs. That is, if a CE xTR fails, 1846 traffic is automatically routed to the other xTRs using the same 1847 anycast address. However, this comes with the disadvantage where the 1848 site cannot control the entrance point when the anycast route is 1849 advertised out from all border routers. Another disadvantage of 1850 using anycast Locators is the limited advertisement scope of /32 (or 1851 /128 for IPv6) routes. 1853 17.3. ISP Provider Edge (PE) xTRs 1855 The use of ISP PE routers as xTRs is not the typical deployment 1856 scenario envisioned in this specification. This section attempts to 1857 capture some of the reasoning behind this preference for implementing 1858 LISP on CE routers. 1860 The use of ISP PE routers for xTR placement gives an ISP, rather than 1861 a site, control over the location of the ETRs. That is, the ISP can 1862 decide whether the xTRs are in the destination site (in either CE 1863 xTRs or last-hop xTRs within a site) or at other PE edges. The 1864 advantage of this case is that two encapsulation headers can be 1865 avoided. By having the PE be the first router on the path to 1866 encapsulate, it can choose a TE path first, and the ETR can 1867 decapsulate and Re-Encapsulate for a new encapsuluation path to the 1868 destination end site. 1870 An obvious disadvantage is that the end site has no control over 1871 where its packets flow or over the RLOCs used. Other disadvantages 1872 include difficulty in synchronizing path liveness updates between CE 1873 and PE routers. 1875 As mentioned in earlier sections, a combination of these scenarios is 1876 possible at the expense of extra packet header overhead; if both site 1877 and provider want control, then Recursive or Re-Encapsulating Tunnels 1878 are used. 1880 17.4. LISP Functionality with Conventional NATs 1882 LISP routers can be deployed behind Network Address Translator (NAT) 1883 devices to provide the same set of packet services hosts have today 1884 when they are addressed out of private address space. 1886 It is important to note that a locator address in any LISP control 1887 message MUST be a globally routable address and therefore SHOULD NOT 1888 contain [RFC1918] addresses. If a LISP xTR is configured with 1889 private RLOC addresses, they MUST be used only in the outer IP header 1890 so the NAT device can translate properly. Otherwise, EID addresses 1891 MUST be translated before encapsulation is performed when LISP VPNs 1892 are not in use. Both NAT translation and LISP encapsulation 1893 functions could be co-located in the same device. 1895 17.5. Packets Egressing a LISP Site 1897 When a LISP site is using two ITRs for redundancy, the failure of one 1898 ITR will likely shift outbound traffic to the second. This second 1899 ITR's cache may not be populated with the same EID-to-RLOC mapping 1900 entries as the first. If this second ITR does not have these 1901 mappings, traffic will be dropped while the mappings are retrieved 1902 from the mapping system. The retrieval of these messages may 1903 increase the load of requests being sent into the mapping system. 1904 Deployment and experimentation will determine whether this issue 1905 requires more attention. 1907 18. Traceroute Considerations 1909 When a source host in a LISP site initiates a traceroute to a 1910 destination host in another LISP site, it is highly desirable for it 1911 to see the entire path. Since packets are encapsulated from the ITR 1912 to the ETR, the hop across the tunnel could be viewed as a single 1913 hop. However, LISP traceroute will provide the entire path so the 1914 user can see 3 distinct segments of the path from a source LISP host 1915 to a destination LISP host: 1917 Segment 1 (in source LISP site based on EIDs): 1919 source host ---> first hop ... next hop ---> ITR 1921 Segment 2 (in the core network based on RLOCs): 1923 ITR ---> next hop ... next hop ---> ETR 1925 Segment 3 (in the destination LISP site based on EIDs): 1927 ETR ---> next hop ... last hop ---> destination host 1929 For segment 1 of the path, ICMP Time Exceeded messages are returned 1930 in the normal manner as they are today. The ITR performs a TTL 1931 decrement and tests for 0 before encapsulating. Therefore, the ITR's 1932 hop is seen by the traceroute source as having an EID address (the 1933 address of the site-facing interface). 1935 For segment 2 of the path, ICMP Time Exceeded messages are returned 1936 to the ITR because the TTL decrement to 0 is done on the outer 1937 header, so the destinations of the ICMP messages are the ITR RLOC 1938 address and the source RLOC address of the encapsulated traceroute 1939 packet. The ITR looks inside of the ICMP payload to inspect the 1940 traceroute source so it can return the ICMP message to the address of 1941 the traceroute client and also retain the core router IP address in 1942 the ICMP message. This is so the traceroute client can display the 1943 core router address (the RLOC address) in the traceroute output. The 1944 ETR returns its RLOC address and responds to the TTL decrement to 0, 1945 as the previous core routers did. 1947 For segment 3, the next-hop router downstream from the ETR will be 1948 decrementing the TTL for the packet that was encapsulated, sent into 1949 the core, decapsulated by the ETR, and forwarded because it isn't the 1950 final destination. If the TTL is decremented to 0, any router on the 1951 path to the destination of the traceroute, including the next-hop 1952 router or destination, will send an ICMP Time Exceeded message to the 1953 source EID of the traceroute client. The ICMP message will be 1954 encapsulated by the local ITR and sent back to the ETR in the 1955 originated traceroute source site, where the packet will be delivered 1956 to the host. 1958 18.1. IPv6 Traceroute 1960 IPv6 traceroute follows the procedure described above, since the 1961 entire traceroute data packet is included in the ICMP Time Exceeded 1962 message payload. Therefore, only the ITR needs to pay special 1963 attention to forwarding ICMP messages back to the traceroute source. 1965 18.2. IPv4 Traceroute 1967 For IPv4 traceroute, we cannot follow the above procedure, since IPv4 1968 ICMP Time Exceeded messages only include the invoking IP header and 1969 8 octets that follow the IP header. Therefore, when a core router 1970 sends an IPv4 Time Exceeded message to an ITR, all the ITR has in the 1971 ICMP payload is the encapsulated header it prepended, followed by a 1972 UDP header. The original invoking IP header, and therefore the 1973 identity of the traceroute source, is lost. 1975 The solution we propose to solve this problem is to cache traceroute 1976 IPv4 headers in the ITR and to match them up with corresponding IPv4 1977 Time Exceeded messages received from core routers and the ETR. The 1978 ITR will use a circular buffer for caching the IPv4 and UDP headers 1979 of traceroute packets. It will select a 16-bit number as a key to 1980 find them later when the IPv4 Time Exceeded messages are received. 1981 When an ITR encapsulates an IPv4 traceroute packet, it will use the 1982 16-bit number as the UDP source port in the encapsulating header. 1983 When the ICMP Time Exceeded message is returned to the ITR, the UDP 1984 header of the encapsulating header is present in the ICMP payload, 1985 thereby allowing the ITR to find the cached headers for the 1986 traceroute source. The ITR puts the cached headers in the payload 1987 and sends the ICMP Time Exceeded message to the traceroute source 1988 retaining the source address of the original ICMP Time Exceeded 1989 message (a core router or the ETR of the site of the traceroute 1990 destination). 1992 The signature of a traceroute packet comes in two forms. The first 1993 form is encoded as a UDP message where the destination port is 1994 inspected for a range of values. The second form is encoded as an 1995 ICMP message where the IP identification field is inspected for a 1996 well-known value. 1998 18.3. Traceroute Using Mixed Locators 2000 When either an IPv4 traceroute or IPv6 traceroute is originated and 2001 the ITR encapsulates it in the other address family header, one 2002 cannot get all 3 segments of the traceroute. Segment 2 of the 2003 traceroute cannot be conveyed to the traceroute source, since it is 2004 expecting addresses from intermediate hops in the same address format 2005 for the type of traceroute it originated. Therefore, in this case, 2006 segment 2 will make the tunnel look like one hop. All the ITR has to 2007 do to make this work is to not copy the inner TTL to the outer, 2008 encapsulating header's TTL when a traceroute packet is encapsulated 2009 using an RLOC from a different address family. This will cause no 2010 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2012 19. Security Considerations 2014 FIXME: ToDo 2016 Security considerations for LISP are discussed in [LISP-THREATS], in 2017 addition [LISP-SEC] provides authentication and integrity to LISP 2018 mappings. 2020 20. Network Management Considerations 2022 Considerations for network management tools exist so the LISP 2023 protocol suite can be operationally managed. These mechanisms can be 2024 found in [LISP-MIB] and [RFC6835]. 2026 21. IANA Considerations 2028 This section provides guidance to the Internet Assigned Numbers 2029 Authority (IANA) regarding registration of values related to the LISP 2030 specification, in accordance with BCP 26 [RFC5226]. 2032 There are four namespaces (listed in the sub-sections below) in LISP 2033 that have been registered. 2035 o LISP IANA registry allocations should not be made for purposes 2036 unrelated to LISP routing or transport protocols. 2038 o The following policies are used here with the meanings defined in 2039 BCP 26: "Specification Required", "IETF Review", "Experimental 2040 Use", and "First Come First Served". 2042 21.1. LISP ACT and Flag Fields 2044 New ACT values [REF_TO_RFC6833bis] can be allocated through IETF 2045 review or IESG approval. Four values have already been allocated by 2046 this specification ([REF_TO_RFC6833bis]. 2048 In addition, LISP has a number of flag fields and reserved fields, 2049 such as the LISP header flags field (Section 5.3). New bits for 2050 flags in these fields can be implemented after IETF review or IESG 2051 approval, but these need not be managed by IANA. 2053 21.2. LISP Address Type Codes 2055 LISP Address [LCAF] type codes have a range from 0 to 255. New type 2056 codes MUST be allocated consecutively, starting at 0. Type Codes 2057 0-127 are to be assigned by IETF review or IESG approval. 2059 Type Codes 128-255 are available according to the [RFC5226] First 2060 Come First Served policy. 2062 This registry, initially empty, is constructed for future use in 2063 experimental work related to LISP Canonical Address Format (LCAF) 2064 values. See [LCAF] for details of other possible unapproved address 2065 encodings. The unapproved LCAF encodings are an area for further 2066 study and experimentation. 2068 21.3. LISP UDP Port Numbers 2070 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2071 lisp-data and lisp-control operation, respectively. IANA has updated 2072 the description for UDP ports 4341 and 4342 as follows: 2074 lisp-data 4341 udp LISP Data Packets 2075 lisp-control 4342 udp LISP Control Packets 2077 21.4. LISP Key ID Numbers 2079 The following Key ID values are defined by this specification as used 2080 in any packet type that references a 'Key ID' field: 2082 Name Number Defined in 2083 ----------------------------------------------- 2084 None 0 n/a 2085 HMAC-SHA-1-96 1 [RFC2404] 2086 HMAC-SHA-256-128 2 [RFC4868] 2088 Number values are in the range of 0 to 65535. The allocation of 2089 values is on a first come first served basis. 2091 22. References 2093 22.1. Normative References 2095 [I.D-ietf-lisp-crypto] 2096 Farinacci, D. and B. Weiss, "LISP Data-Plane 2097 Confidentiality", October 2016. 2099 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2100 DOI 10.17487/RFC0768, August 1980, 2101 . 2103 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2104 DOI 10.17487/RFC0791, September 1981, 2105 . 2107 [RFC1918] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G., 2108 and E. Lear, "Address Allocation for Private Internets", 2109 BCP 5, RFC 1918, DOI 10.17487/RFC1918, February 1996, 2110 . 2112 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2113 Requirement Levels", BCP 14, RFC 2119, 2114 DOI 10.17487/RFC2119, March 1997, 2115 . 2117 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 2118 ESP and AH", RFC 2404, DOI 10.17487/RFC2404, November 2119 1998, . 2121 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2122 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 2123 December 1998, . 2125 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2126 of Explicit Congestion Notification (ECN) to IP", 2127 RFC 3168, DOI 10.17487/RFC3168, September 2001, 2128 . 2130 [RFC3232] Reynolds, J., Ed., "Assigned Numbers: RFC 1700 is Replaced 2131 by an On-line Database", RFC 3232, DOI 10.17487/RFC3232, 2132 January 2002, . 2134 [RFC4086] Eastlake 3rd, D., Schiller, J., and S. Crocker, 2135 "Randomness Requirements for Security", BCP 106, RFC 4086, 2136 DOI 10.17487/RFC4086, June 2005, 2137 . 2139 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 2140 (CIDR): The Internet Address Assignment and Aggregation 2141 Plan", BCP 122, RFC 4632, DOI 10.17487/RFC4632, August 2142 2006, . 2144 [RFC4868] Kelly, S. and S. Frankel, "Using HMAC-SHA-256, HMAC-SHA- 2145 384, and HMAC-SHA-512 with IPsec", RFC 4868, 2146 DOI 10.17487/RFC4868, May 2007, 2147 . 2149 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2150 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2151 DOI 10.17487/RFC5226, May 2008, 2152 . 2154 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 2155 Forwarding (RPF) Vector TLV", RFC 5496, 2156 DOI 10.17487/RFC5496, March 2009, 2157 . 2159 [RFC5944] Perkins, C., Ed., "IP Mobility Support for IPv4, Revised", 2160 RFC 5944, DOI 10.17487/RFC5944, November 2010, 2161 . 2163 [RFC6115] Li, T., Ed., "Recommendation for a Routing Architecture", 2164 RFC 6115, DOI 10.17487/RFC6115, February 2011, 2165 . 2167 [RFC6275] Perkins, C., Ed., Johnson, D., and J. Arkko, "Mobility 2168 Support in IPv6", RFC 6275, DOI 10.17487/RFC6275, July 2169 2011, . 2171 [RFC6833] Farinacci, D. and V. Fuller, "Locator/ID Separation 2172 Protocol (LISP) Map-Server Interface", RFC 6833, January 2173 2013. 2175 [RFC6834] Iannone, L., Saucez, D., and O. Bonaventure, "Locator/ID 2176 Separation Protocol (LISP) Map-Versioning", RFC 6834, 2177 January 2013. 2179 [RFC6836] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis, 2180 "Locator/ID Separation Protocol Alternative Logical 2181 Topology (LISP+ALT)", RFC 6836, January 2013. 2183 [RFC7215] Jakab, L., Cabellos, A., Coras, F., Domingo-Pascual, J., 2184 and D. Lewis, "Locator/Identifier Separation Protocol 2185 (LISP) Network Element Deployment Considerations", 2186 RFC 6834, April 2014. 2188 22.2. Informative References 2190 [AFI] IANA, "Address Family Numbers", November 2007, 2191 . 2193 [BGP-SEC] Lepinski, M. and S. Turner, "An Overview of BGPSEC", Work 2194 in Progress, May 2012. 2196 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 2197 Enhancement to the Internet Architecture", 1999, 2198 . 2200 [CONS] Brim, S., Chiappa, N., Farinacci, D., Fuller, V., Lewis, 2201 D., and D. Meyer, "LISP-CONS: A Content distribution 2202 Overlay Network Service for LISP", Work in Progress, April 2203 2008. 2205 [EMACS] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID 2206 Mappings Multicast Across Cooperating Systems for LISP", 2207 Work in Progress, November 2007. 2209 [I-D.portoles-lisp-eid-mobility] 2210 Portoles, M., Ashtaputre, V., Moreno, V., Maino, F., and 2211 D. Farinacci, "LISP L2/L3 EID Mobility Using a Unified 2212 Control Plane", Work in Progress, April 2016. 2214 [I.D-ietf-lisp-signal-free-multicast] 2215 Moreno, V. and D. Farinacci, "Signal-Free LISP Multicast", 2216 Work in Progress, October 2016. 2218 [LCAF] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 2219 Address Format (LCAF)", Work in Progress, January 2013. 2221 [LISA96] Lear, E., Tharp, D., Katinsky, J., and J. Coffin, 2222 "Renumbering: Threat or Menace?", Usenix Tenth System 2223 Administration Conference (LISA 96), October 1996. 2225 [LISP-DDT] 2226 Fuller, V., Lewis, D., Ermagan, V., Jain, A., and A. 2227 Smirnov, "LISP Delegated Database Tree", Work in Progress, 2228 April 2015. 2230 [LISP-INTRO] 2231 Cabellos, A. and D. Saucez, "An Architectural Introduction 2232 to the Locator/ID Separation Protocol (LISP)", Work 2233 in Progress, April 2015. 2235 [LISP-MIB] 2236 Schudel, G., Jain, A., and V. Moreno, "LISP MIB", Work 2237 in Progress, January 2013. 2239 [LISP-MN] Farinacci, D., Lewis, D., Meyer, D., and C. White, "LISP 2240 Mobile Node", Work in Progress, October 2012. 2242 [LISP-SEC] 2243 Maino, F., Ermagan, V., Cabellos, A., Saucez, D., and O. 2244 Bonaventure, "LISP-Security (LISP-SEC)", Work in Progress, 2245 October 2012. 2247 [LISP-THREATS] 2248 Saucez, D., Iannone, L., and O. Bonaventure, "LISP Threats 2249 Analysis", Work in Progress, January 2016. 2251 [LOC-ID-ARCH] 2252 Meyer, D. and D. Lewis, "Architectural Implications of 2253 Locator/ID Separation", Work in Progress, January 2009. 2255 [OPENLISP] 2256 Iannone, L., Saucez, D., and O. Bonaventure, "OpenLISP 2257 Implementation Report", Work in Progress, July 2008. 2259 [RADIR] Narten, T., "On the Scalability of Internet Routing", Work 2260 in Progress, February 2010. 2262 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2263 STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, 2264 . 2266 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2267 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2268 DOI 10.17487/RFC2784, March 2000, 2269 . 2271 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 2272 via IPv4 Clouds", RFC 3056, DOI 10.17487/RFC3056, February 2273 2001, . 2275 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2276 A., Peterson, J., Sparks, R., Handley, M., and E. 2277 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2278 DOI 10.17487/RFC3261, June 2002, 2279 . 2281 [RFC4107] Bellovin, S. and R. Housley, "Guidelines for Cryptographic 2282 Key Management", BCP 107, RFC 4107, DOI 10.17487/RFC4107, 2283 June 2005, . 2285 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2286 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2287 DOI 10.17487/RFC4192, September 2005, 2288 . 2290 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2291 Optimization for Mobile IPv6", RFC 4866, 2292 DOI 10.17487/RFC4866, May 2007, 2293 . 2295 [RFC4984] Meyer, D., Ed., Zhang, L., Ed., and K. Fall, Ed., "Report 2296 from the IAB Workshop on Routing and Addressing", 2297 RFC 4984, DOI 10.17487/RFC4984, September 2007, 2298 . 2300 [RFC6480] Lepinski, M. and S. Kent, "An Infrastructure to Support 2301 Secure Internet Routing", RFC 6480, DOI 10.17487/RFC6480, 2302 February 2012, . 2304 [RFC6518] Lebovitz, G. and M. Bhatia, "Keying and Authentication for 2305 Routing Protocols (KARP) Design Guidelines", RFC 6518, 2306 DOI 10.17487/RFC6518, February 2012, 2307 . 2309 [RFC6831] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, "The 2310 Locator/ID Separation Protocol (LISP) for Multicast 2311 Environments", RFC 6831, January 2013. 2313 [RFC6832] Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 2314 "Interworking between Locator/ID Separation Protocol 2315 (LISP) and Non-LISP Sites", RFC 6832, January 2013. 2317 [RFC6835] Farinacci, D. and D. Meyer, "The Locator/ID Separation 2318 Protocol Internet Groper (LIG)", RFC 6835, January 2013. 2320 [RFC6837] Lear, E., "NERD: A Not-so-novel Endpoint ID (EID) to 2321 Routing Locator (RLOC) Database", RFC 6837, January 2013. 2323 [UDP-TUNNELS] 2324 Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2325 UDP Checksums for Tunneled Packets", Work in Progress, 2326 January 2013. 2328 [UDP-ZERO] 2329 Fairhurst, G. and M. Westerlund, "Applicability Statement 2330 for the use of IPv6 UDP Datagrams with Zero Checksums", 2331 Work in Progress, December 2012. 2333 Appendix A. Acknowledgments 2335 An initial thank you goes to Dave Oran for planting the seeds for the 2336 initial ideas for LISP. His consultation continues to provide value 2337 to the LISP authors. 2339 A special and appreciative thank you goes to Noel Chiappa for 2340 providing architectural impetus over the past decades on separation 2341 of location and identity, as well as detailed reviews of the LISP 2342 architecture and documents, coupled with enthusiasm for making LISP a 2343 practical and incremental transition for the Internet. 2345 The authors would like to gratefully acknowledge many people who have 2346 contributed discussions and ideas to the making of this proposal. 2347 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 2348 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 2349 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 2350 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 2351 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 2352 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 2353 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 2354 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 2355 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 2356 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 2357 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 2358 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 2359 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 2360 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 2361 Ermagan, Fabio Maino, Victor Moreno, Chris White, Clarence Filsfils, 2362 Alia Atlas, Florin Coras and Alberto Rodriguez. 2364 This work originated in the Routing Research Group (RRG) of the IRTF. 2365 An individual submission was converted into the IETF LISP working 2366 group document that became this RFC. 2368 The LISP working group would like to give a special thanks to Jari 2369 Arkko, the Internet Area AD at the time that the set of LISP 2370 documents were being prepared for IESG last call, and for his 2371 meticulous reviews and detailed commentaries on the 7 working group 2372 last call documents progressing toward experimental RFCs. 2374 Appendix B. Document Change Log 2376 [RFC Editor: Please delete this section on publication as RFC.] 2378 B.1. Changes to draft-ietf-lisp-rfc6830bis-00 2380 o Posted December 2016. 2382 o Created working group document from draft-farinacci-lisp 2383 -rfc6830-00 individual submission. No other changes made. 2385 Authors' Addresses 2387 Dino Farinacci 2388 Cisco Systems 2389 Tasman Drive 2390 San Jose, CA 95134 2391 USA 2393 EMail: farinacci@gmail.com 2395 Vince Fuller 2396 Cisco Systems 2397 Tasman Drive 2398 San Jose, CA 95134 2399 USA 2401 EMail: vaf@vaf.net 2403 Dave Meyer 2404 Cisco Systems 2405 170 Tasman Drive 2406 San Jose, CA 2407 USA 2409 EMail: dmm@1-4-5.net 2411 Darrel Lewis 2412 Cisco Systems 2413 170 Tasman Drive 2414 San Jose, CA 2415 USA 2417 EMail: darlewis@cisco.com 2418 Albert Cabellos 2419 UPC/BarcelonaTech 2420 Campus Nord, C. Jordi Girona 1-3 2421 Barcelona, Catalunya 2422 Spain 2424 EMail: acabello@ac.upc.edu