idnits 2.17.1 draft-ietf-lisp-rfc6830bis-08.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 (January 9, 2018) is 2299 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-31) exists of draft-ietf-lisp-rfc6833bis-07 == Outdated reference: A later version (-13) exists of draft-ietf-lisp-eid-mobility-01 == Outdated reference: A later version (-15) exists of draft-ietf-lisp-introduction-13 == Outdated reference: A later version (-15) exists of draft-ietf-lisp-mn-01 == Outdated reference: A later version (-14) exists of draft-ietf-lisp-predictive-rlocs-01 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-14 == Outdated reference: A later version (-09) exists of draft-ietf-lisp-signal-free-multicast-07 == Outdated reference: A later version (-12) exists of draft-ietf-lisp-vpn-01 -- Obsolete informational reference (is this intentional?): RFC 6834 (Obsoleted by RFC 9302) Summary: 0 errors (**), 0 flaws (~~), 9 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: July 13, 2018 D. Lewis 6 Cisco Systems 7 A. Cabellos (Ed.) 8 UPC/BarcelonaTech 9 January 9, 2018 11 The Locator/ID Separation Protocol (LISP) 12 draft-ietf-lisp-rfc6830bis-08 14 Abstract 16 This document describes the data-plane protocol for the Locator/ID 17 Separation Protocol (LISP). LISP defines two namespaces, End-point 18 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. 24 LISP requires no change to either host protocol stacks or to underlay 25 routers and offers Traffic Engineering, multihoming and mobility, 26 among other features. 28 Status of This Memo 30 This Internet-Draft is submitted in full conformance with the 31 provisions of BCP 78 and BCP 79. 33 Internet-Drafts are working documents of the Internet Engineering 34 Task Force (IETF). Note that other groups may also distribute 35 working documents as Internet-Drafts. The list of current Internet- 36 Drafts is at https://datatracker.ietf.org/drafts/current/. 38 Internet-Drafts are draft documents valid for a maximum of six months 39 and may be updated, replaced, or obsoleted by other documents at any 40 time. It is inappropriate to use Internet-Drafts as reference 41 material or to cite them other than as "work in progress." 43 This Internet-Draft will expire on July 13, 2018. 45 Copyright Notice 47 Copyright (c) 2018 IETF Trust and the persons identified as the 48 document authors. All rights reserved. 50 This document is subject to BCP 78 and the IETF Trust's Legal 51 Provisions Relating to IETF Documents 52 (https://trustee.ietf.org/license-info) in effect on the date of 53 publication of this document. Please review these documents 54 carefully, as they describe your rights and restrictions with respect 55 to this document. Code Components extracted from this document must 56 include Simplified BSD License text as described in Section 4.e of 57 the Trust Legal Provisions and are provided without warranty as 58 described in the Simplified BSD License. 60 Table of Contents 62 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 63 2. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 64 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 4 65 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . 9 66 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . 11 67 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . 13 68 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 13 69 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 14 70 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . 15 71 6. LISP EID-to-RLOC Map-Cache . . . . . . . . . . . . . . . . . 19 72 7. Dealing with Large Encapsulated Packets . . . . . . . . . . . 20 73 7.1. A Stateless Solution to MTU Handling . . . . . . . . . . 20 74 7.2. A Stateful Solution to MTU Handling . . . . . . . . . . . 21 75 8. Using Virtualization and Segmentation with LISP . . . . . . . 22 76 9. Routing Locator Selection . . . . . . . . . . . . . . . . . . 23 77 10. Routing Locator Reachability . . . . . . . . . . . . . . . . 24 78 10.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . . 27 79 10.2. RLOC-Probing Algorithm . . . . . . . . . . . . . . . . . 28 80 11. EID Reachability within a LISP Site . . . . . . . . . . . . . 29 81 12. Routing Locator Hashing . . . . . . . . . . . . . . . . . . . 29 82 13. Changing the Contents of EID-to-RLOC Mappings . . . . . . . . 30 83 13.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . . 31 84 13.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . . 32 85 13.3. Database Map-Versioning . . . . . . . . . . . . . . . . 33 86 14. Multicast Considerations . . . . . . . . . . . . . . . . . . 34 87 15. Router Performance Considerations . . . . . . . . . . . . . . 35 88 16. Mobility Considerations . . . . . . . . . . . . . . . . . . . 35 89 16.1. Slow Mobility . . . . . . . . . . . . . . . . . . . . . 36 90 16.2. Fast Mobility . . . . . . . . . . . . . . . . . . . . . 36 91 16.3. LISP Mobile Node Mobility . . . . . . . . . . . . . . . 37 92 17. LISP xTR Placement and Encapsulation Methods . . . . . . . . 37 93 17.1. First-Hop/Last-Hop xTRs . . . . . . . . . . . . . . . . 38 94 17.2. Border/Edge xTRs . . . . . . . . . . . . . . . . . . . . 39 95 17.3. ISP Provider Edge (PE) xTRs . . . . . . . . . . . . . . 39 96 17.4. LISP Functionality with Conventional NATs . . . . . . . 40 97 17.5. Packets Egressing a LISP Site . . . . . . . . . . . . . 40 98 18. Traceroute Considerations . . . . . . . . . . . . . . . . . . 40 99 18.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . 41 100 18.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . 42 101 18.3. Traceroute Using Mixed Locators . . . . . . . . . . . . 42 102 19. Security Considerations . . . . . . . . . . . . . . . . . . . 43 103 20. Network Management Considerations . . . . . . . . . . . . . . 43 104 21. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 44 105 21.1. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . 44 106 22. References . . . . . . . . . . . . . . . . . . . . . . . . . 44 107 22.1. Normative References . . . . . . . . . . . . . . . . . . 44 108 22.2. Informative References . . . . . . . . . . . . . . . . . 45 109 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 50 110 Appendix B. Document Change Log . . . . . . . . . . . . . . . . 50 111 B.1. Changes to draft-ietf-lisp-rfc6830bis-08 . . . . . . . . 51 112 B.2. Changes to draft-ietf-lisp-rfc6830bis-07 . . . . . . . . 51 113 B.3. Changes to draft-ietf-lisp-rfc6830bis-06 . . . . . . . . 51 114 B.4. Changes to draft-ietf-lisp-rfc6830bis-05 . . . . . . . . 51 115 B.5. Changes to draft-ietf-lisp-rfc6830bis-04 . . . . . . . . 52 116 B.6. Changes to draft-ietf-lisp-rfc6830bis-03 . . . . . . . . 52 117 B.7. Changes to draft-ietf-lisp-rfc6830bis-02 . . . . . . . . 52 118 B.8. Changes to draft-ietf-lisp-rfc6830bis-01 . . . . . . . . 52 119 B.9. Changes to draft-ietf-lisp-rfc6830bis-00 . . . . . . . . 52 120 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 52 122 1. Introduction 124 This document describes the Locator/Identifier Separation Protocol 125 (LISP). LISP is an encapsulation protocol built around the 126 fundamental idea of separating the topological location of a network 127 attachment point from the node's identity [CHIAPPA]. As a result 128 LISP creates two namespaces: Endpoint Identifiers (EIDs), that are 129 used to identify end-hosts (e.g., nodes or Virtual Machines) and 130 routable Routing Locators (RLOCs), used to identify network 131 attachment points. LISP then defines functions for mapping between 132 the two namespaces and for encapsulating traffic originated by 133 devices using non-routable EIDs for transport across a network 134 infrastructure that routes and forwards using RLOCs. LISP 135 encapsulation uses a dynamic form of tunneling where no static 136 provisioning is required or necessary. 138 LISP is an overlay protocol that separates control from data-plane, 139 this document specifies the data-plane, how LISP-capable routers 140 (Tunnel Routers) exchange packets by encapsulating them to the 141 appropriate location. Tunnel routers are equipped with a cache, 142 called map-cache, that contains EID-to-RLOC mappings. The map-cache 143 is populated using the LISP Control-Plane protocol 144 [I-D.ietf-lisp-rfc6833bis]. 146 LISP does not require changes to either host protocol stack or to 147 underlay routers. By separating the EID from the RLOC space, LISP 148 offers native Traffic Engineering, multihoming and mobility, among 149 other features. 151 Creation of LISP was initially motivated by discussions during the 152 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 153 October 2006 (see [RFC4984]) 155 This document specifies the LISP data-plane encapsulation and other 156 LISP forwarding node functionality while [I-D.ietf-lisp-rfc6833bis] 157 specifies the LISP control plane. LISP deployment guidelines can be 158 found in [RFC7215] and [RFC6835] describes considerations for network 159 operational management. Finally, [I-D.ietf-lisp-introduction] 160 describes the LISP architecture. 162 2. Requirements Notation 164 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 165 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 166 document are to be interpreted as described in [RFC2119]. 168 3. Definition of Terms 170 Address Family Identifier (AFI): AFI is a term used to describe an 171 address encoding in a packet. An address family that pertains to 172 the data-plane. See [AFN] and [RFC3232] for details. An AFI 173 value of 0 used in this specification indicates an unspecified 174 encoded address where the length of the address is 0 octets 175 following the 16-bit AFI value of 0. 177 Anycast Address: Anycast Address is a term used in this document to 178 refer to the same IPv4 or IPv6 address configured and used on 179 multiple systems at the same time. An EID or RLOC can be an 180 anycast address in each of their own address spaces. 182 Client-side: Client-side is a term used in this document to indicate 183 a connection initiation attempt by an EID. The ITR(s) at the LISP 184 site are the first to get involved in forwarding a packet from the 185 source EID. 187 Data-Probe: A Data-Probe is a LISP-encapsulated data packet where 188 the inner-header destination address equals the outer-header 189 destination address used to trigger a Map-Reply by a decapsulating 190 ETR. In addition, the original packet is decapsulated and 191 delivered to the destination host if the destination EID is in the 192 EID-Prefix range configured on the ETR. Otherwise, the packet is 193 discarded. A Data-Probe is used in some of the mapping database 194 designs to "probe" or request a Map-Reply from an ETR; in other 195 cases, Map-Requests are used. See each mapping database design 196 for details. When using Data-Probes, by sending Map-Requests on 197 the underlying routing system, EID-Prefixes must be advertised. 199 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 200 packet where the destination address in the "outer" IP header is 201 one of its own RLOCs. The router strips the "outer" header and 202 forwards the packet based on the next IP header found. In 203 general, an ETR receives LISP-encapsulated IP packets from the 204 Internet on one side and sends decapsulated IP packets to site 205 end-systems on the other side. ETR functionality does not have to 206 be limited to a router device. A server host can be the endpoint 207 of a LISP tunnel as well. 209 EID-to-RLOC Database: The EID-to-RLOC Database is a global 210 distributed database that contains all known EID-Prefix-to-RLOC 211 mappings. Each potential ETR typically contains a small piece of 212 the database: the EID-to-RLOC mappings for the EID-Prefixes 213 "behind" the router. These map to one of the router's own 214 globally visible IP addresses. Note that there MAY be transient 215 conditions when the EID-Prefix for the site and Locator-Set for 216 each EID-Prefix may not be the same on all ETRs. This has no 217 negative implications, since a partial set of Locators can be 218 used. 220 EID-to-RLOC Map-Cache: The EID-to-RLOC map-cache is generally 221 short-lived, on-demand table in an ITR that stores, tracks, and is 222 responsible for timing out and otherwise validating EID-to-RLOC 223 mappings. This cache is distinct from the full "database" of EID- 224 to-RLOC mappings; it is dynamic, local to the ITR(s), and 225 relatively small, while the database is distributed, relatively 226 static, and much more global in scope. 228 EID-Prefix: An EID-Prefix is a power-of-two block of EIDs that are 229 allocated to a site by an address allocation authority. EID- 230 Prefixes are associated with a set of RLOC addresses. EID-Prefix 231 allocations can be broken up into smaller blocks when an RLOC set 232 is to be associated with the larger EID-Prefix block. 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 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 242 IPv6) value used in the source and destination address fields of 243 the first (most inner) LISP header of a packet. The host obtains 244 a destination EID the same way it obtains a destination address 245 today, for example, through a Domain Name System (DNS) [RFC1034] 246 lookup or Session Initiation Protocol (SIP) [RFC3261] exchange. 247 The source EID is obtained via existing mechanisms used to set a 248 host's "local" IP address. An EID used on the public Internet 249 must have the same properties as any other IP address used in that 250 manner; this means, among other things, that it must be globally 251 unique. An EID is allocated to a host from an EID-Prefix block 252 associated with the site where the host is located. An EID can be 253 used by a host to refer to other hosts. Note that EID blocks MAY 254 be assigned in a hierarchical manner, independent of the network 255 topology, to facilitate scaling of the mapping database. In 256 addition, an EID block assigned to a site MAY have site-local 257 structure (subnetting) for routing within the site; this structure 258 is not visible to the global routing system. In theory, the bit 259 string that represents an EID for one device can represent an RLOC 260 for a different device. When used in discussions with other 261 Locator/ID separation proposals, a LISP EID will be called an 262 "LEID". Throughout this document, any references to "EID" refer 263 to an LEID. 265 Ingress Tunnel Router (ITR): An ITR is a router that resides in a 266 LISP site. Packets sent by sources inside of the LISP site to 267 destinations outside of the site are candidates for encapsulation 268 by the ITR. The ITR treats the IP destination address as an EID 269 and performs an EID-to-RLOC mapping lookup. The router then 270 prepends an "outer" IP header with one of its routable RLOCs (in 271 the RLOC space) in the source address field and the result of the 272 mapping lookup in the destination address field. Note that this 273 destination RLOC MAY be an intermediate, proxy device that has 274 better knowledge of the EID-to-RLOC mapping closer to the 275 destination EID. In general, an ITR receives IP packets from site 276 end-systems on one side and sends LISP-encapsulated IP packets 277 toward the Internet on the other side. 279 Specifically, when a service provider prepends a LISP header for 280 Traffic Engineering purposes, the router that does this is also 281 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 282 on the outer destination address (the originating ITR's supplied 283 RLOC) or the inner destination address (the originating host's 284 supplied EID). 286 LISP Header: LISP header is a term used in this document to refer 287 to the outer IPv4 or IPv6 header, a UDP header, and a LISP- 288 specific 8-octet header that follow the UDP header and that an ITR 289 prepends or an ETR strips. 291 LISP Router: A LISP router is a router that performs the functions 292 of any or all of the following: ITR, ETR, RTR, Proxy-ITR (PITR), 293 or Proxy-ETR (PETR). 295 LISP Site: LISP site is a set of routers in an edge network that are 296 under a single technical administration. LISP routers that reside 297 in the edge network are the demarcation points to separate the 298 edge network from the core network. 300 Locator-Status-Bits (LSBs): Locator-Status-Bits are present in the 301 LISP header. They are used by ITRs to inform ETRs about the up/ 302 down status of all ETRs at the local site. These bits are used as 303 a hint to convey up/down router status and not path reachability 304 status. The LSBs can be verified by use of one of the Locator 305 reachability algorithms described in Section 10. 307 Negative Mapping Entry: A negative mapping entry, also known as a 308 negative cache entry, is an EID-to-RLOC entry where an EID-Prefix 309 is advertised or stored with no RLOCs. That is, the Locator-Set 310 for the EID-to-RLOC entry is empty or has an encoded Locator count 311 of 0. This type of entry could be used to describe a prefix from 312 a non-LISP site, which is explicitly not in the mapping database. 313 There are a set of well-defined actions that are encoded in a 314 Negative Map-Reply. 316 Provider-Assigned (PA) Addresses: PA addresses are an address block 317 assigned to a site by each service provider to which a site 318 connects. Typically, each block is a sub-block of a service 319 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 320 is aggregated into the larger block before being advertised into 321 the underlay network. Traditionally, IP multihoming has been 322 implemented by each multihomed site acquiring its own globally 323 visible prefix. 325 Provider-Independent (PI) Addresses: PI addresses are an address 326 block assigned from a pool where blocks are not associated with 327 any particular location in the network (e.g., from a particular 328 service provider) and are therefore not topologically aggregatable 329 in the routing system. 331 Proxy-ETR (PETR): A PETR is defined and described in [RFC6832]. A 332 PETR acts like an ETR but does so on behalf of LISP sites that 333 send packets to destinations at non-LISP sites. 335 Proxy-ITR (PITR): A PITR is defined and described in [RFC6832]. A 336 PITR acts like an ITR but does so on behalf of non-LISP sites that 337 send packets to destinations at LISP sites. 339 Recursive Tunneling: Recursive Tunneling occurs when a packet has 340 more than one LISP IP header. Additional layers of tunneling MAY 341 be employed to implement Traffic Engineering or other re-routing 342 as needed. When this is done, an additional "outer" LISP header 343 is added, and the original RLOCs are preserved in the "inner" 344 header. 346 Re-Encapsulating Tunneling Router (RTR): An RTR acts like an ETR to 347 remove a LISP header, then acts as an ITR to prepend a new LISP 348 header. This is known as Re-encapsulating Tunneling. Doing this 349 allows a packet to be re-routed by the RTR without adding the 350 overhead of additional tunnel headers. Any references to tunnels 351 in this specification refer to dynamic encapsulating tunnels; they 352 are never statically configured. When using multiple mapping 353 database systems, care must be taken to not create re- 354 encapsulation loops through misconfiguration. 356 Route-Returnability: Route-returnability is an assumption that the 357 underlying routing system will deliver packets to the destination. 358 When combined with a nonce that is provided by a sender and 359 returned by a receiver, this limits off-path data insertion. A 360 route-returnability check is verified when a message is sent with 361 a nonce, another message is returned with the same nonce, and the 362 destination of the original message appears as the source of the 363 returned message. 365 Routing Locator (RLOC): An RLOC is an IPv4 [RFC0791] or IPv6 366 [RFC8200] address of an Egress Tunnel Router (ETR). An RLOC is 367 the output of an EID-to-RLOC mapping lookup. An EID maps to one 368 or more RLOCs. Typically, RLOCs are numbered from aggregatable 369 blocks that are assigned to a site at each point to which it 370 attaches to the underlay network; where the topology is defined by 371 the connectivity of provider networks. Multiple RLOCs can be 372 assigned to the same ETR device or to multiple ETR devices at a 373 site. 375 Server-side: Server-side is a term used in this document to indicate 376 that a connection initiation attempt is being accepted for a 377 destination EID. 379 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 380 network that strips an outer LISP header for Traffic Engineering 381 purposes. 383 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 384 network that prepends an additional LISP header for Traffic 385 Engineering purposes. 387 xTR: An xTR is a reference to an ITR or ETR when direction of data 388 flow is not part of the context description. "xTR" refers to the 389 router that is the tunnel endpoint and is used synonymously with 390 the term "Tunnel Router". For example, "An xTR can be located at 391 the Customer Edge (CE) router" indicates both ITR and ETR 392 functionality at the CE router. 394 4. Basic Overview 396 One key concept of LISP is that end-systems operate the same way they 397 do today. The IP addresses that hosts use for tracking sockets and 398 connections, and for sending and receiving packets, do not change. 399 In LISP terminology, these IP addresses are called Endpoint 400 Identifiers (EIDs). 402 Routers continue to forward packets based on IP destination 403 addresses. When a packet is LISP encapsulated, these addresses are 404 referred to as Routing Locators (RLOCs). Most routers along a path 405 between two hosts will not change; they continue to perform routing/ 406 forwarding lookups on the destination addresses. For routers between 407 the source host and the ITR as well as routers from the ETR to the 408 destination host, the destination address is an EID. For the routers 409 between the ITR and the ETR, the destination address is an RLOC. 411 Another key LISP concept is the "Tunnel Router". A Tunnel Router 412 prepends LISP headers on host-originated packets and strips them 413 prior to final delivery to their destination. The IP addresses in 414 this "outer header" are RLOCs. During end-to-end packet exchange 415 between two Internet hosts, an ITR prepends a new LISP header to each 416 packet, and an ETR strips the new header. The ITR performs EID-to- 417 RLOC lookups to determine the routing path to the ETR, which has the 418 RLOC as one of its IP addresses. 420 Some basic rules governing LISP are: 422 o End-systems only send to addresses that are EIDs. They don't know 423 that addresses are EIDs versus RLOCs but assume that packets get 424 to their intended destinations. In a system where LISP is 425 deployed, LISP routers intercept EID-addressed packets and assist 426 in delivering them across the network core where EIDs cannot be 427 routed. The procedure a host uses to send IP packets does not 428 change. 430 o EIDs are typically IP addresses assigned to hosts. 432 o Other types of EID are supported by LISP, see [RFC8060] for 433 further information. 435 o LISP routers mostly deal with Routing Locator addresses. See 436 details in Section 4.1 to clarify what is meant by "mostly". 438 o RLOCs are always IP addresses assigned to routers, preferably 439 topologically oriented addresses from provider CIDR (Classless 440 Inter-Domain Routing) blocks. 442 o When a router originates packets, it MAY use as a source address 443 either an EID or RLOC. When acting as a host (e.g., when 444 terminating a transport session such as Secure SHell (SSH), 445 TELNET, or the Simple Network Management Protocol (SNMP)), it MAY 446 use an EID that is explicitly assigned for that purpose. An EID 447 that identifies the router as a host MUST NOT be used as an RLOC; 448 an EID is only routable within the scope of a site. A typical BGP 449 configuration might demonstrate this "hybrid" EID/RLOC usage where 450 a router could use its "host-like" EID to terminate iBGP sessions 451 to other routers in a site while at the same time using RLOCs to 452 terminate eBGP sessions to routers outside the site. 454 o Packets with EIDs in them are not expected to be delivered end-to- 455 end in the absence of an EID-to-RLOC mapping operation. They are 456 expected to be used locally for intra-site communication or to be 457 encapsulated for inter-site communication. 459 o EID-Prefixes are likely to be hierarchically assigned in a manner 460 that is optimized for administrative convenience and to facilitate 461 scaling of the EID-to-RLOC mapping database. 463 o EIDs MAY also be structured (subnetted) in a manner suitable for 464 local routing within an Autonomous System (AS). 466 An additional LISP header MAY be prepended to packets by a TE-ITR 467 when re-routing of the path for a packet is desired. A potential 468 use-case for this would be an ISP router that needs to perform 469 Traffic Engineering for packets flowing through its network. In such 470 a situation, termed "Recursive Tunneling", an ISP transit acts as an 471 additional ITR, and the RLOC it uses for the new prepended header 472 would be either a TE-ETR within the ISP (along an intra-ISP traffic 473 engineered path) or a TE-ETR within another ISP (an inter-ISP traffic 474 engineered path, where an agreement to build such a path exists). 476 In order to avoid excessive packet overhead as well as possible 477 encapsulation loops, this document recommends that a maximum of two 478 LISP headers can be prepended to a packet. For initial LISP 479 deployments, it is assumed that two headers is sufficient, where the 480 first prepended header is used at a site for Location/Identity 481 separation and the second prepended header is used inside a service 482 provider for Traffic Engineering purposes. 484 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 485 For example, the ITR for a particular end-to-end packet exchange 486 might be the first-hop or default router within a site for the source 487 host. Similarly, the ETR might be the last-hop router directly 488 connected to the destination host. Another example, perhaps for a 489 VPN service outsourced to an ISP by a site, the ITR could be the 490 site's border router at the service provider attachment point. 491 Mixing and matching of site-operated, ISP-operated, and other Tunnel 492 Routers is allowed for maximum flexibility. 494 4.1. Packet Flow Sequence 496 This section provides an example of the unicast packet flow, 497 including also control-plane information as specified in 498 [I-D.ietf-lisp-rfc6833bis]. The example also assumes the following 499 conditions: 501 o Source host "host1.abc.example.com" is sending a packet to 502 "host2.xyz.example.com", exactly what host1 would do if the site 503 was not using LISP. 505 o Each site is multihomed, so each Tunnel Router has an address 506 (RLOC) assigned from the service provider address block for each 507 provider to which that particular Tunnel Router is attached. 509 o The ITR(s) and ETR(s) are directly connected to the source and 510 destination, respectively, but the source and destination can be 511 located anywhere in the LISP site. 513 o A Map-Request is sent for an external destination when the 514 destination is not found in the forwarding table or matches a 515 default route. Map-Requests are sent to the mapping database 516 system by using the LISP control-plane protocol documented in 517 [I-D.ietf-lisp-rfc6833bis]. 519 o Map-Replies are sent on the underlying routing system topology 520 using the [I-D.ietf-lisp-rfc6833bis] control-plane protocol. 522 Client host1.abc.example.com wants to communicate with server 523 host2.xyz.example.com: 525 1. host1.abc.example.com wants to open a TCP connection to 526 host2.xyz.example.com. It does a DNS lookup on 527 host2.xyz.example.com. An A/AAAA record is returned. This 528 address is the destination EID. The locally assigned address of 529 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 530 packet is built and forwarded through the LISP site as a normal 531 IP packet until it reaches a LISP ITR. 533 2. The LISP ITR must be able to map the destination EID to an RLOC 534 of one of the ETRs at the destination site. The specific method 535 used to do this is not described in this example. See 536 [I-D.ietf-lisp-rfc6833bis] for further information. 538 3. The ITR sends a LISP Map-Request as specified in 539 [I-D.ietf-lisp-rfc6833bis]. Map-Requests SHOULD be rate-limited. 541 4. The mapping system helps forwarding the Map-Request to the 542 corresponding ETR. When the Map-Request arrives at one of the 543 ETRs at the destination site, it will process the packet as a 544 control message. 546 5. The ETR looks at the destination EID of the Map-Request and 547 matches it against the prefixes in the ETR's configured EID-to- 548 RLOC mapping database. This is the list of EID-Prefixes the ETR 549 is supporting for the site it resides in. If there is no match, 550 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 551 returned to the ITR. 553 6. The ITR receives the Map-Reply message, parses the message (to 554 check for format validity), and stores the mapping information 555 from the packet. This information is stored in the ITR's EID-to- 556 RLOC map-cache. Note that the map-cache is an on-demand cache. 557 An ITR will manage its map-cache in such a way that optimizes for 558 its resource constraints. 560 7. Subsequent packets from host1.abc.example.com to 561 host2.xyz.example.com will have a LISP header prepended by the 562 ITR using the appropriate RLOC as the LISP header destination 563 address learned from the ETR. Note that the packet MAY be sent 564 to a different ETR than the one that returned the Map-Reply due 565 to the source site's hashing policy or the destination site's 566 Locator-Set policy. 568 8. The ETR receives these packets directly (since the destination 569 address is one of its assigned IP addresses), checks the validity 570 of the addresses, strips the LISP header, and forwards packets to 571 the attached destination host. 573 9. In order to defer the need for a mapping lookup in the reverse 574 direction, an ETR can OPTIONALLY create a cache entry that maps 575 the source EID (inner-header source IP address) to the source 576 RLOC (outer-header source IP address) in a received LISP packet. 577 Such a cache entry is termed a "gleaned" mapping and only 578 contains a single RLOC for the EID in question. More complete 579 information about additional RLOCs SHOULD be verified by sending 580 a LISP Map-Request for that EID. Both the ITR and the ETR MAY 581 also influence the decision the other makes in selecting an RLOC. 583 5. LISP Encapsulation Details 585 Since additional tunnel headers are prepended, the packet becomes 586 larger and can exceed the MTU of any link traversed from the ITR to 587 the ETR. It is RECOMMENDED in IPv4 that packets do not get 588 fragmented as they are encapsulated by the ITR. Instead, the packet 589 is dropped and an ICMP Unreachable/Fragmentation-Needed message is 590 returned to the source. 592 In the case when fragmentation is needed, this specification 593 RECOMMENDS that implementations provide support for one of the 594 proposed fragmentation and reassembly schemes. Two existing schemes 595 are detailed in Section 7. 597 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 598 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 599 header is in IPv4 packet format and the outer header is in IPv6 600 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 601 is in IPv6 packet format and the outer header is in IPv4 packet 602 format). The next sub-sections illustrate packet formats for the 603 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6), but all 4 604 combinations MUST be supported. Additional types of EIDs are defined 605 in [RFC8060]. 607 5.1. LISP IPv4-in-IPv4 Header Format 608 0 1 2 3 609 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 610 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 611 / |Version| IHL | DSCP |ECN| Total Length | 612 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 613 | | Identification |Flags| Fragment Offset | 614 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 615 OH | Time to Live | Protocol = 17 | Header Checksum | 616 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 617 | | Source Routing Locator | 618 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 619 \ | Destination Routing Locator | 620 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 621 / | Source Port = xxxx | Dest Port = 4341 | 622 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 623 \ | UDP Length | UDP Checksum | 624 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 625 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 626 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 627 S / | Instance ID/Locator-Status-Bits | 628 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 629 / |Version| IHL | DSCP |ECN| Total Length | 630 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 631 | | Identification |Flags| Fragment Offset | 632 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 633 IH | Time to Live | Protocol | Header Checksum | 634 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 635 | | Source EID | 636 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 637 \ | Destination EID | 638 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 640 IHL = IP-Header-Length 642 5.2. LISP IPv6-in-IPv6 Header Format 644 0 1 2 3 645 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 1 646 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 647 / |Version| DSCP |ECN| Flow Label | 648 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 649 | | Payload Length | Next Header=17| Hop Limit | 650 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 651 | | 652 O + + 653 u | | 654 t + Source Routing Locator + 655 e | | 656 r + + 657 | | 658 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 659 d | | 660 r + + 661 | | 662 ^ + Destination Routing Locator + 663 | | | 664 \ + + 665 \ | | 666 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 667 / | Source Port = xxxx | Dest Port = 4341 | 668 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 669 \ | UDP Length | UDP Checksum | 670 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 671 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 672 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 673 S / | Instance ID/Locator-Status-Bits | 674 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 675 / |Version| DSCP |ECN| Flow Label | 676 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 677 / | Payload Length | Next Header | Hop Limit | 678 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 679 | | 680 I + + 681 n | | 682 n + Source EID + 683 e | | 684 r + + 685 | | 686 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 687 d | | 688 r + + 689 | | 690 ^ + Destination EID + 691 \ | | 692 \ + + 693 \ | | 694 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 696 5.3. Tunnel Header Field Descriptions 698 Inner Header (IH): The inner header is the header on the 699 datagram received from the originating host [RFC0791] [RFC8200] 700 [RFC2474]. The source and destination IP addresses are EIDs. 702 Outer Header: (OH) The outer header is a new header prepended by an 703 ITR. The address fields contain RLOCs obtained from the ingress 704 router's EID-to-RLOC Cache. The IP protocol number is "UDP (17)" 705 from [RFC0768]. The setting of the Don't Fragment (DF) bit 706 'Flags' field is according to rules listed in Sections 7.1 and 707 7.2. 709 UDP Header: The UDP header contains an ITR selected source port when 710 encapsulating a packet. See Section 12 for details on the hash 711 algorithm used to select a source port based on the 5-tuple of the 712 inner header. The destination port MUST be set to the well-known 713 IANA-assigned port value 4341. 715 UDP Checksum: The 'UDP Checksum' field SHOULD be transmitted as zero 716 by an ITR for either IPv4 [RFC0768] and IPv6 encapsulation 717 [RFC6935] [RFC6936]. When a packet with a zero UDP checksum is 718 received by an ETR, the ETR MUST accept the packet for 719 decapsulation. When an ITR transmits a non-zero value for the UDP 720 checksum, it MUST send a correctly computed value in this field. 721 When an ETR receives a packet with a non-zero UDP checksum, it MAY 722 choose to verify the checksum value. If it chooses to perform 723 such verification, and the verification fails, the packet MUST be 724 silently dropped. If the ETR chooses not to perform the 725 verification, or performs the verification successfully, the 726 packet MUST be accepted for decapsulation. The handling of UDP 727 zero checksums over IPv6 for all tunneling protocols, including 728 LISP, is subject to the applicability statement in [RFC6936]. 730 UDP Length: The 'UDP Length' field is set for an IPv4-encapsulated 731 packet to be the sum of the inner-header IPv4 Total Length plus 732 the UDP and LISP header lengths. For an IPv6-encapsulated packet, 733 the 'UDP Length' field is the sum of the inner-header IPv6 Payload 734 Length, the size of the IPv6 header (40 octets), and the size of 735 the UDP and LISP headers. 737 N: The N-bit is the nonce-present bit. When this bit is set to 1, 738 the low-order 24 bits of the first 32 bits of the LISP header 739 contain a Nonce. See Section 10.1 for details. Both N- and 740 V-bits MUST NOT be set in the same packet. If they are, a 741 decapsulating ETR MUST treat the 'Nonce/Map-Version' field as 742 having a Nonce value present. 744 L: The L-bit is the 'Locator-Status-Bits' field enabled bit. When 745 this bit is set to 1, the Locator-Status-Bits in the second 746 32 bits of the LISP header are in use. 748 x 1 x x 0 x x x 749 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 750 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 751 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 752 | Locator-Status-Bits | 753 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 755 E: The E-bit is the echo-nonce-request bit. This bit MUST be ignored 756 and has no meaning when the N-bit is set to 0. When the N-bit is 757 set to 1 and this bit is set to 1, an ITR is requesting that the 758 nonce value in the 'Nonce' field be echoed back in LISP- 759 encapsulated packets when the ITR is also an ETR. See 760 Section 10.1 for details. 762 V: The V-bit is the Map-Version present bit. When this bit is set to 763 1, the N-bit MUST be 0. Refer to Section 13.3 for more details. 764 This bit indicates that the LISP header is encoded in this 765 case as: 767 0 x 0 1 x x x x 768 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 769 |N|L|E|V|I|R|K|K| Source Map-Version | Dest Map-Version | 770 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 771 | Instance ID/Locator-Status-Bits | 772 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 774 I: The I-bit is the Instance ID bit. See Section 8 for more details. 775 When this bit is set to 1, the 'Locator-Status-Bits' field is 776 reduced to 8 bits and the high-order 24 bits are used as an 777 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 778 are transmitted as zero and ignored on receipt. The format of the 779 LISP header would look like this: 781 x x x x 1 x x x 782 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 783 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 784 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 785 | Instance ID | LSBs | 786 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 788 R: The R-bit is a Reserved bit for future use. It MUST be set to 0 789 on transmit and MUST be ignored on receipt. 791 KK: The KK-bits are a 2-bit field used when encapsualted packets are 792 encrypted. The field is set to 00 when the packet is not 793 encrypted. See [RFC8061] for further information. 795 LISP Nonce: The LISP 'Nonce' field is a 24-bit value that is 796 randomly generated by an ITR when the N-bit is set to 1. Nonce 797 generation algorithms are an implementation matter but are 798 required to generate different nonces when sending to different 799 destinations. However, the same nonce can be used for a period of 800 time to the same destination. The nonce is also used when the 801 E-bit is set to request the nonce value to be echoed by the other 802 side when packets are returned. When the E-bit is clear but the 803 N-bit is set, a remote ITR is either echoing a previously 804 requested echo-nonce or providing a random nonce. See 805 Section 10.1 for more details. 807 LISP Locator-Status-Bits (LSBs): When the L-bit is also set, the 808 'Locator-Status-Bits' field in the LISP header is set by an ITR to 809 indicate to an ETR the up/down status of the Locators in the 810 source site. Each RLOC in a Map-Reply is assigned an ordinal 811 value from 0 to n-1 (when there are n RLOCs in a mapping entry). 812 The Locator-Status-Bits are numbered from 0 to n-1 from the least 813 significant bit of the field. The field is 32 bits when the I-bit 814 is set to 0 and is 8 bits when the I-bit is set to 1. When a 815 Locator-Status-Bit is set to 1, the ITR is indicating to the ETR 816 that the RLOC associated with the bit ordinal has up status. See 817 Section 10 for details on how an ITR can determine the status of 818 the ETRs at the same site. When a site has multiple EID-Prefixes 819 that result in multiple mappings (where each could have a 820 different Locator-Set), the Locator-Status-Bits setting in an 821 encapsulated packet MUST reflect the mapping for the EID-Prefix 822 that the inner-header source EID address matches. If the LSB for 823 an anycast Locator is set to 1, then there is at least one RLOC 824 with that address, and the ETR is considered 'up'. 826 When doing ITR/PITR encapsulation: 828 o The outer-header 'Time to Live' field (or 'Hop Limit' field, in 829 the case of IPv6) SHOULD be copied from the inner-header 'Time to 830 Live' field. 832 o The outer-header 'Type of Service' field (or the 'Traffic Class' 833 field, in the case of IPv6) SHOULD be copied from the inner-header 834 'Type of Service' field (with one exception; see below). 836 When doing ETR/PETR decapsulation: 838 o The inner-header 'Time to Live' field (or 'Hop Limit' field, in 839 the case of IPv6) SHOULD be copied from the outer-header 'Time to 840 Live' field, when the Time to Live value of the outer header is 841 less than the Time to Live value of the inner header. Failing to 842 perform this check can cause the Time to Live of the inner header 843 to increment across encapsulation/decapsulation cycles. This 844 check is also performed when doing initial encapsulation, when a 845 packet comes to an ITR or PITR destined for a LISP site. 847 o The inner-header 'Type of Service' field (or the 'Traffic Class' 848 field, in the case of IPv6) SHOULD be copied from the outer-header 849 'Type of Service' field (with one exception; see below). 851 Note that if an ETR/PETR is also an ITR/PITR and chooses to re- 852 encapsulate after decapsulating, the net effect of this is that the 853 new outer header will carry the same Time to Live as the old outer 854 header minus 1. 856 Copying the Time to Live (TTL) serves two purposes: first, it 857 preserves the distance the host intended the packet to travel; 858 second, and more importantly, it provides for suppression of looping 859 packets in the event there is a loop of concatenated tunnels due to 860 misconfiguration. See Section 18.3 for TTL exception handling for 861 traceroute packets. 863 The Explicit Congestion Notification ('ECN') field occupies bits 6 864 and 7 of both the IPv4 'Type of Service' field and the IPv6 'Traffic 865 Class' field [RFC3168]. The 'ECN' field requires special treatment 866 in order to avoid discarding indications of congestion [RFC3168]. An 867 ITR/PITR encapsulation MUST copy the 2-bit 'ECN' field from the inner 868 header to the outer header. Re-encapsulation MUST copy the 2-bit 869 'ECN' field from the stripped outer header to the new outer header. 870 If the 'ECN' field contains a congestion indication codepoint (the 871 value is '11', the Congestion Experienced (CE) codepoint), then ETR/ 872 PETR decapsulation MUST copy the 2-bit 'ECN' field from the stripped 873 outer header to the surviving inner header that is used to forward 874 the packet beyond the ETR. These requirements preserve CE 875 indications when a packet that uses ECN traverses a LISP tunnel and 876 becomes marked with a CE indication due to congestion between the 877 tunnel endpoints. 879 6. LISP EID-to-RLOC Map-Cache 881 ITRs and PITRs maintain an on-demand cache, referred as LISP EID-to- 882 RLOC Map-Cache, that contains mappings from EID-prefixes to locator 883 sets. The cache is used to encapsulate packets from the EID space to 884 the corresponding RLOC network attachment point. 886 When an ITR/PITR receives a packet from inside of the LISP site to 887 destinations outside of the site a longest-prefix match lookup of the 888 EID is done to the map-cache. 890 When the lookup succeeds, the locator-set retrieved from the map- 891 cache is used to send the packet to the EID's topological location. 893 If the lookup fails, the ITR/PITR needs to retrieve the mapping using 894 the LISP control-plane protocol [I-D.ietf-lisp-rfc6833bis]. The 895 mapping is then stored in the local map-cache to forward subsequent 896 packets addressed to the same EID-prefix. 898 The map-cache is a local cache of mappings, entries are expired based 899 on the associated Time to live. In addition, entries can be updated 900 with more current information, see Section 13 for further information 901 on this. Finally, the map-cache also contains reachability 902 information about EIDs and RLOCs, and uses LISP reachability 903 information mechanisms to determine the reachability of RLOCs, see 904 Section 10 for the specific mechanisms. 906 7. Dealing with Large Encapsulated Packets 908 This section proposes two mechanisms to deal with packets that exceed 909 the path MTU between the ITR and ETR. 911 It is left to the implementor to decide if the stateless or stateful 912 mechanism SHOULD be implemented. Both or neither can be used, since 913 it is a local decision in the ITR regarding how to deal with MTU 914 issues, and sites can interoperate with differing mechanisms. 916 Both stateless and stateful mechanisms also apply to Re-encapsulating 917 and Recursive Tunneling, so any actions below referring to an ITR 918 also apply to a TE-ITR. 920 7.1. A Stateless Solution to MTU Handling 922 An ITR stateless solution to handle MTU issues is described as 923 follows: 925 1. Define H to be the size, in octets, of the outer header an ITR 926 prepends to a packet. This includes the UDP and LISP header 927 lengths. 929 2. Define L to be the size, in octets, of the maximum-sized packet 930 an ITR can send to an ETR without the need for the ITR or any 931 intermediate routers to fragment the packet. 933 3. Define an architectural constant S for the maximum size of a 934 packet, in octets, an ITR MUST receive from the source so the 935 effective MTU can be met. That is, L = S + H. 937 When an ITR receives a packet from a site-facing interface and adds H 938 octets worth of encapsulation to yield a packet size greater than L 939 octets (meaning the received packet size was greater than S octets 940 from the source), it resolves the MTU issue by first splitting the 941 original packet into 2 equal-sized fragments. A LISP header is then 942 prepended to each fragment. The size of the encapsulated fragments 943 is then (S/2 + H), which is less than the ITR's estimate of the path 944 MTU between the ITR and its correspondent ETR. 946 When an ETR receives encapsulated fragments, it treats them as two 947 individually encapsulated packets. It strips the LISP headers and 948 then forwards each fragment to the destination host of the 949 destination site. The two fragments are reassembled at the 950 destination host into the single IP datagram that was originated by 951 the source host. Note that reassembly can happen at the ETR if the 952 encapsulated packet was fragmented at or after the ITR. 954 This behavior is performed by the ITR when the source host originates 955 a packet with the 'DF' field of the IP header set to 0. When the 956 'DF' field of the IP header is set to 1, or the packet is an IPv6 957 packet originated by the source host, the ITR will drop the packet 958 when the size is greater than L and send an ICMP Unreachable/ 959 Fragmentation-Needed message to the source with a value of S, where S 960 is (L - H). 962 When the outer-header encapsulation uses an IPv4 header, an 963 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 964 can be avoided. An implementation MAY set the DF bit in such headers 965 to 0 if it has good reason to believe there are unresolvable path MTU 966 issues between the sending ITR and the receiving ETR. 968 This specification RECOMMENDS that L be defined as 1500. 970 7.2. A Stateful Solution to MTU Handling 972 An ITR stateful solution to handle MTU issues is described as follows 973 and was first introduced in [OPENLISP]: 975 1. The ITR will keep state of the effective MTU for each Locator per 976 Map-Cache entry. The effective MTU is what the core network can 977 deliver along the path between the ITR and ETR. 979 2. When an IPv6-encapsulated packet, or an IPv4-encapsulated packet 980 with the DF bit set to 1, exceeds what the core network can 981 deliver, one of the intermediate routers on the path will send an 982 ICMP Unreachable/Fragmentation-Needed message to the ITR. The 983 ITR will parse the ICMP message to determine which Locator is 984 affected by the effective MTU change and then record the new 985 effective MTU value in the Map-Cache entry. 987 3. When a packet is received by the ITR from a source inside of the 988 site and the size of the packet is greater than the effective MTU 989 stored with the Map-Cache entry associated with the destination 990 EID the packet is for, the ITR will send an ICMP Unreachable/ 991 Fragmentation-Needed message back to the source. The packet size 992 advertised by the ITR in the ICMP Unreachable/Fragmentation- 993 Needed message is the effective MTU minus the LISP encapsulation 994 length. 996 Even though this mechanism is stateful, it has advantages over the 997 stateless IP fragmentation mechanism, by not involving the 998 destination host with reassembly of ITR fragmented packets. 1000 8. Using Virtualization and Segmentation with LISP 1002 There are several cases where segregation is needed at the EID level. 1003 For instance, this is the case for deployments containing overlapping 1004 addresses, traffic isolation policies or multi-tenant virtualization. 1005 For these and other scenarios where segregation is needed, Instance 1006 IDs are used. 1008 An Instance ID can be carried in a LISP-encapsulated packet. An ITR 1009 that prepends a LISP header will copy a 24-bit value used by the LISP 1010 router to uniquely identify the address space. The value is copied 1011 to the 'Instance ID' field of the LISP header, and the I-bit is set 1012 to 1. 1014 When an ETR decapsulates a packet, the Instance ID from the LISP 1015 header is used as a table identifier to locate the forwarding table 1016 to use for the inner destination EID lookup. 1018 For example, an 802.1Q VLAN tag or VPN identifier could be used as a 1019 24-bit Instance ID. See [I-D.ietf-lisp-vpn] for LISP VPN use-case 1020 details. 1022 The Instance ID that is stored in the mapping database when LISP-DDT 1023 [RFC8111] is used is 32 bits in length. That means the control-plane 1024 can store more instances than a given data-plane can use. Multiple 1025 data-planes can use the same 32-bit space as long as the low-order 24 1026 bits don't overlap among xTRs. 1028 9. Routing Locator Selection 1030 Both the client-side and server-side MAY need control over the 1031 selection of RLOCs for conversations between them. This control is 1032 achieved by manipulating the 'Priority' and 'Weight' fields in EID- 1033 to-RLOC Map-Reply messages. Alternatively, RLOC information MAY be 1034 gleaned from received tunneled packets or EID-to-RLOC Map-Request 1035 messages. 1037 The following are different scenarios for choosing RLOCs and the 1038 controls that are available: 1040 o The server-side returns one RLOC. The client-side can only use 1041 one RLOC. The server-side has complete control of the selection. 1043 o The server-side returns a list of RLOCs where a subset of the list 1044 has the same best Priority. The client can only use the subset 1045 list according to the weighting assigned by the server-side. In 1046 this case, the server-side controls both the subset list and load- 1047 splitting across its members. The client-side can use RLOCs 1048 outside of the subset list if it determines that the subset list 1049 is unreachable (unless RLOCs are set to a Priority of 255). Some 1050 sharing of control exists: the server-side determines the 1051 destination RLOC list and load distribution while the client-side 1052 has the option of using alternatives to this list if RLOCs in the 1053 list are unreachable. 1055 o The server-side sets a Weight of 0 for the RLOC subset list. In 1056 this case, the client-side can choose how the traffic load is 1057 spread across the subset list. Control is shared by the server- 1058 side determining the list and the client determining load 1059 distribution. Again, the client can use alternative RLOCs if the 1060 server-provided list of RLOCs is unreachable. 1062 o Either side (more likely the server-side ETR) decides not to send 1063 a Map-Request. For example, if the server-side ETR does not send 1064 Map-Requests, it gleans RLOCs from the client-side ITR, giving the 1065 client-side ITR responsibility for bidirectional RLOC reachability 1066 and preferability. Server-side ETR gleaning of the client-side 1067 ITR RLOC is done by caching the inner-header source EID and the 1068 outer-header source RLOC of received packets. The client-side ITR 1069 controls how traffic is returned and can alternate using an outer- 1070 header source RLOC, which then can be added to the list the 1071 server-side ETR uses to return traffic. Since no Priority or 1072 Weights are provided using this method, the server-side ETR MUST 1073 assume that each client-side ITR RLOC uses the same best Priority 1074 with a Weight of zero. In addition, since EID-Prefix encoding 1075 cannot be conveyed in data packets, the EID-to-RLOC Cache on 1076 Tunnel Routers can grow to be very large. 1078 o A "gleaned" Map-Cache entry, one learned from the source RLOC of a 1079 received encapsulated packet, is only stored and used for a few 1080 seconds, pending verification. Verification is performed by 1081 sending a Map-Request to the source EID (the inner-header IP 1082 source address) of the received encapsulated packet. A reply to 1083 this "verifying Map-Request" is used to fully populate the Map- 1084 Cache entry for the "gleaned" EID and is stored and used for the 1085 time indicated from the 'TTL' field of a received Map-Reply. When 1086 a verified Map-Cache entry is stored, data gleaning no longer 1087 occurs for subsequent packets that have a source EID that matches 1088 the EID-Prefix of the verified entry. This "gleaning" mechanism 1089 is OPTIONAL. 1091 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1092 reachable when the R-bit for the Locator record is set to 1. When 1093 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1094 RLOC. Neither the information contained in a Map-Reply nor that 1095 stored in the mapping database system provides reachability 1096 information for RLOCs. Note that reachability is not part of the 1097 mapping system and is determined using one or more of the Routing 1098 Locator reachability algorithms described in the next section. 1100 10. Routing Locator Reachability 1102 Several mechanisms for determining RLOC reachability are currently 1103 defined: 1105 1. An ETR MAY examine the Locator-Status-Bits in the LISP header of 1106 an encapsulated data packet received from an ITR. If the ETR is 1107 also acting as an ITR and has traffic to return to the original 1108 ITR site, it can use this status information to help select an 1109 RLOC. 1111 2. An ITR MAY receive an ICMP Network Unreachable or Host 1112 Unreachable message for an RLOC it is using. This indicates that 1113 the RLOC is likely down. Note that trusting ICMP messages may 1114 not be desirable, but neither is ignoring them completely. 1115 Implementations are encouraged to follow current best practices 1116 in treating these conditions. 1118 3. An ITR that participates in the global routing system can 1119 determine that an RLOC is down if no BGP Routing Information Base 1120 (RIB) route exists that matches the RLOC IP address. 1122 4. An ITR MAY receive an ICMP Port Unreachable message from a 1123 destination host. This occurs if an ITR attempts to use 1124 interworking [RFC6832] and LISP-encapsulated data is sent to a 1125 non-LISP-capable site. 1127 5. An ITR MAY receive a Map-Reply from an ETR in response to a 1128 previously sent Map-Request. The RLOC source of the Map-Reply is 1129 likely up, since the ETR was able to send the Map-Reply to the 1130 ITR. 1132 6. When an ETR receives an encapsulated packet from an ITR, the 1133 source RLOC from the outer header of the packet is likely up. 1135 7. An ITR/ETR pair can use the Locator reachability algorithms 1136 described in this section, namely Echo-Noncing or RLOC-Probing. 1138 When determining Locator up/down reachability by examining the 1139 Locator-Status-Bits from the LISP-encapsulated data packet, an ETR 1140 will receive up-to-date status from an encapsulating ITR about 1141 reachability for all ETRs at the site. CE-based ITRs at the source 1142 site can determine reachability relative to each other using the site 1143 IGP as follows: 1145 o Under normal circumstances, each ITR will advertise a default 1146 route into the site IGP. 1148 o If an ITR fails or if the upstream link to its PE fails, its 1149 default route will either time out or be withdrawn. 1151 Each ITR can thus observe the presence or lack of a default route 1152 originated by the others to determine the Locator-Status-Bits it sets 1153 for them. 1155 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1156 Locator-Status-Bits in a LISP-encapsulated packet are numbered from 0 1157 to n-1 starting with the least significant bit. For example, if an 1158 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1159 value 2), then all ITRs at the site will clear the 3rd least 1160 significant bit (xxxx x0xx) of the 'Locator-Status-Bits' field for 1161 the packets they encapsulate. 1163 When an ETR decapsulates a packet, it will check for any change in 1164 the 'Locator-Status-Bits' field. When a bit goes from 1 to 0, the 1165 ETR, if acting also as an ITR, will refrain from encapsulating 1166 packets to an RLOC that is indicated as down. It will only resume 1167 using that RLOC if the corresponding Locator-Status-Bit returns to a 1168 value of 1. Locator-Status-Bits are associated with a Locator-Set 1169 per EID-Prefix. Therefore, when a Locator becomes unreachable, the 1170 Locator-Status-Bit that corresponds to that Locator's position in the 1171 list returned by the last Map-Reply will be set to zero for that 1172 particular EID-Prefix. Refer to Section 19 for security related 1173 issues regarding Locator-Status-Bits. 1175 When ITRs at the site are not deployed in CE routers, the IGP can 1176 still be used to determine the reachability of Locators, provided 1177 they are injected into the IGP. This is typically done when a /32 1178 address is configured on a loopback interface. 1180 When ITRs receive ICMP Network Unreachable or Host Unreachable 1181 messages as a method to determine unreachability, they will refrain 1182 from using Locators that are described in Locator lists of Map- 1183 Replies. However, using this approach is unreliable because many 1184 network operators turn off generation of ICMP Destination Unreachable 1185 messages. 1187 If an ITR does receive an ICMP Network Unreachable or Host 1188 Unreachable message, it MAY originate its own ICMP Destination 1189 Unreachable message destined for the host that originated the data 1190 packet the ITR encapsulated. 1192 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1193 locator address from a Locator-Set in a mapping entry matches a 1194 prefix. If it does not find one and BGP is running in the Default- 1195 Free Zone (DFZ), it can decide to not use the Locator even though the 1196 Locator-Status-Bits indicate that the Locator is up. In this case, 1197 the path from the ITR to the ETR that is assigned the Locator is not 1198 available. More details are in [I-D.meyer-loc-id-implications]. 1200 Optionally, an ITR can send a Map-Request to a Locator, and if a Map- 1201 Reply is returned, reachability of the Locator has been determined. 1202 Obviously, sending such probes increases the number of control 1203 messages originated by Tunnel Routers for active flows, so Locators 1204 are assumed to be reachable when they are advertised. 1206 This assumption does create a dependency: Locator unreachability is 1207 detected by the receipt of ICMP Host Unreachable messages. When a 1208 Locator has been determined to be unreachable, it is not used for 1209 active traffic; this is the same as if it were listed in a Map-Reply 1210 with Priority 255. 1212 The ITR can test the reachability of the unreachable Locator by 1213 sending periodic Requests. Both Requests and Replies MUST be rate- 1214 limited. Locator reachability testing is never done with data 1215 packets, since that increases the risk of packet loss for end-to-end 1216 sessions. 1218 When an ETR decapsulates a packet, it knows that it is reachable from 1219 the encapsulating ITR because that is how the packet arrived. In 1220 most cases, the ETR can also reach the ITR but cannot assume this to 1221 be true, due to the possibility of path asymmetry. In the presence 1222 of unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD 1223 NOT use the lack of return traffic as an indication that the ETR is 1224 unreachable. Instead, it MUST use an alternate mechanism to 1225 determine reachability. 1227 10.1. Echo Nonce Algorithm 1229 When data flows bidirectionally between Locators from different 1230 sites, a data-plane mechanism called "nonce echoing" can be used to 1231 determine reachability between an ITR and ETR. When an ITR wants to 1232 solicit a nonce echo, it sets the N- and E-bits and places a 24-bit 1233 nonce [RFC4086] in the LISP header of the next encapsulated data 1234 packet. 1236 When this packet is received by the ETR, the encapsulated packet is 1237 forwarded as normal. When the ETR next sends a data packet to the 1238 ITR, it includes the nonce received earlier with the N-bit set and 1239 E-bit cleared. The ITR sees this "echoed nonce" and knows that the 1240 path to and from the ETR is up. 1242 The ITR will set the E-bit and N-bit for every packet it sends while 1243 in the echo-nonce-request state. The time the ITR waits to process 1244 the echoed nonce before it determines the path is unreachable is 1245 variable and is a choice left for the implementation. 1247 If the ITR is receiving packets from the ETR but does not see the 1248 nonce echoed while being in the echo-nonce-request state, then the 1249 path to the ETR is unreachable. This decision MAY be overridden by 1250 other Locator reachability algorithms. Once the ITR determines that 1251 the path to the ETR is down, it can switch to another Locator for 1252 that EID-Prefix. 1254 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 1255 be implementing both ITR and ETR functionality for the echo nonce 1256 mechanism to operate. 1258 The ITR and ETR MAY both go into the echo-nonce-request state at the 1259 same time. The number of packets sent or the time during which echo 1260 nonce requests are sent is an implementation-specific setting. 1261 However, when an ITR is in the echo-nonce-request state, it can echo 1262 the ETR's nonce in the next set of packets that it encapsulates and 1263 subsequently continue sending echo-nonce-request packets. 1265 This mechanism does not completely solve the forward path 1266 reachability problem, as traffic may be unidirectional. That is, the 1267 ETR receiving traffic at a site MAY not be the same device as an ITR 1268 that transmits traffic from that site, or the site-to-site traffic is 1269 unidirectional so there is no ITR returning traffic. 1271 The echo-nonce algorithm is bilateral. That is, if one side sets the 1272 E-bit and the other side is not enabled for echo-noncing, then the 1273 echoing of the nonce does not occur and the requesting side may 1274 erroneously consider the Locator unreachable. An ITR SHOULD only set 1275 the E-bit in an encapsulated data packet when it knows the ETR is 1276 enabled for echo-noncing. This is conveyed by the E-bit in the RLOC- 1277 probe Map-Reply message. 1279 Note other Locator Reachability mechanisms can be used to compliment 1280 or even override the echo nonce algorithm. See the next section for 1281 an example of control-plane probing. 1283 10.2. RLOC-Probing Algorithm 1285 RLOC-Probing is a method that an ITR or PITR can use to determine the 1286 reachability status of one or more Locators that it has cached in a 1287 Map-Cache entry. The probe-bit of the Map-Request and Map-Reply 1288 messages is used for RLOC-Probing. 1290 RLOC-Probing is done in the control plane on a timer basis, where an 1291 ITR or PITR will originate a Map-Request destined to a locator 1292 address from one of its own locator addresses. A Map-Request used as 1293 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or to 1294 the mapping database system as one would when soliciting mapping 1295 data. The EID record encoded in the Map-Request is the EID-Prefix of 1296 the Map-Cache entry cached by the ITR or PITR. The ITR MAY include a 1297 mapping data record for its own database mapping information that 1298 contains the local EID-Prefixes and RLOCs for its site. RLOC-probes 1299 are sent periodically using a jittered timer interval. 1301 When an ETR receives a Map-Request message with the probe-bit set, it 1302 returns a Map-Reply with the probe-bit set. The source address of 1303 the Map-Reply is set according to the procedure described in 1304 [I-D.ietf-lisp-rfc6833bis]. The Map-Reply SHOULD contain mapping 1305 data for the EID-Prefix contained in the Map-Request. This provides 1306 the opportunity for the ITR or PITR that sent the RLOC-probe to get 1307 mapping updates if there were changes to the ETR's database mapping 1308 entries. 1310 There are advantages and disadvantages of RLOC-Probing. The greatest 1311 benefit of RLOC-Probing is that it can handle many failure scenarios 1312 allowing the ITR to determine when the path to a specific Locator is 1313 reachable or has become unreachable, thus providing a robust 1314 mechanism for switching to using another Locator from the cached 1315 Locator. RLOC-Probing can also provide rough Round-Trip Time (RTT) 1316 estimates between a pair of Locators, which can be useful for network 1317 management purposes as well as for selecting low delay paths. The 1318 major disadvantage of RLOC-Probing is in the number of control 1319 messages required and the amount of bandwidth used to obtain those 1320 benefits, especially if the requirement for failure detection times 1321 is very small. 1323 11. EID Reachability within a LISP Site 1325 A site MAY be multihomed using two or more ETRs. The hosts and 1326 infrastructure within a site will be addressed using one or more EID- 1327 Prefixes that are mapped to the RLOCs of the relevant ETRs in the 1328 mapping system. One possible failure mode is for an ETR to lose 1329 reachability to one or more of the EID-Prefixes within its own site. 1330 When this occurs when the ETR sends Map-Replies, it can clear the 1331 R-bit associated with its own Locator. And when the ETR is also an 1332 ITR, it can clear its Locator-Status-Bit in the encapsulation data 1333 header. 1335 It is recognized that there are no simple solutions to the site 1336 partitioning problem because it is hard to know which part of the 1337 EID-Prefix range is partitioned and which Locators can reach any sub- 1338 ranges of the EID-Prefixes. Note that this is not a new problem 1339 introduced by the LISP architecture. The problem exists today when a 1340 multihomed site uses BGP to advertise its reachability upstream. 1342 12. Routing Locator Hashing 1344 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message 1345 that is stored in the map-cache of a requesting ITR, the Locator-Set 1346 for the EID-Prefix MAY contain different Priority and Weight values 1347 for each locator address. When more than one best Priority Locator 1348 exists, the ITR can decide how to load-share traffic against the 1349 corresponding Locators. 1351 The following hash algorithm MAY be used by an ITR to select a 1352 Locator for a packet destined to an EID for the EID-to-RLOC mapping: 1354 1. Either a source and destination address hash or the traditional 1355 5-tuple hash can be used. The traditional 5-tuple hash includes 1356 the source and destination addresses; source and destination TCP, 1357 UDP, or Stream Control Transmission Protocol (SCTP) port numbers; 1358 and the IP protocol number field or IPv6 next-protocol fields of 1359 a packet that a host originates from within a LISP site. When a 1360 packet is not a TCP, UDP, or SCTP packet, the source and 1361 destination addresses only from the header are used to compute 1362 the hash. 1364 2. Take the hash value and divide it by the number of Locators 1365 stored in the Locator-Set for the EID-to-RLOC mapping. 1367 3. The remainder will yield a value of 0 to "number of Locators 1368 minus 1". Use the remainder to select the Locator in the 1369 Locator-Set. 1371 Note that when a packet is LISP encapsulated, the source port number 1372 in the outer UDP header needs to be set. Selecting a hashed value 1373 allows core routers that are attached to Link Aggregation Groups 1374 (LAGs) to load-split the encapsulated packets across member links of 1375 such LAGs. Otherwise, core routers would see a single flow, since 1376 packets have a source address of the ITR, for packets that are 1377 originated by different EIDs at the source site. A suggested setting 1378 for the source port number computed by an ITR is a 5-tuple hash 1379 function on the inner header, as described above. 1381 Many core router implementations use a 5-tuple hash to decide how to 1382 balance packet load across members of a LAG. The 5-tuple hash 1383 includes the source and destination addresses of the packet and the 1384 source and destination ports when the protocol number in the packet 1385 is TCP or UDP. For this reason, UDP encoding is used for LISP 1386 encapsulation. 1388 13. Changing the Contents of EID-to-RLOC Mappings 1390 Since the LISP architecture uses a caching scheme to retrieve and 1391 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 1392 date mapping is to re-request the mapping. However, the ITRs do not 1393 know when the mappings change, and the ETRs do not keep track of 1394 which ITRs requested its mappings. For scalability reasons, it is 1395 desirable to maintain this approach but need to provide a way for 1396 ETRs to change their mappings and inform the sites that are currently 1397 communicating with the ETR site using such mappings. 1399 When adding a new Locator record in lexicographic order to the end of 1400 a Locator-Set, it is easy to update mappings. We assume that new 1401 mappings will maintain the same Locator ordering as the old mapping 1402 but will just have new Locators appended to the end of the list. So, 1403 some ITRs can have a new mapping while other ITRs have only an old 1404 mapping that is used until they time out. When an ITR has only an 1405 old mapping but detects bits set in the Locator-Status-Bits that 1406 correspond to Locators beyond the list it has cached, it simply 1407 ignores them. However, this can only happen for locator addresses 1408 that are lexicographically greater than the locator addresses in the 1409 existing Locator-Set. 1411 When a Locator record is inserted in the middle of a Locator-Set, to 1412 maintain lexicographic order, the SMR procedure in Section 13.2 is 1413 used to inform ITRs and PITRs of the new Locator-Status-Bit mappings. 1415 When a Locator record is removed from a Locator-Set, ITRs that have 1416 the mapping cached will not use the removed Locator because the xTRs 1417 will set the Locator-Status-Bit to 0. So, even if the Locator is in 1418 the list, it will not be used. For new mapping requests, the xTRs 1419 can set the Locator AFI to 0 (indicating an unspecified address), as 1420 well as setting the corresponding Locator-Status-Bit to 0. This 1421 forces ITRs with old or new mappings to avoid using the removed 1422 Locator. 1424 If many changes occur to a mapping over a long period of time, one 1425 will find empty record slots in the middle of the Locator-Set and new 1426 records appended to the Locator-Set. At some point, it would be 1427 useful to compact the Locator-Set so the Locator-Status-Bit settings 1428 can be efficiently packed. 1430 We propose here three approaches for Locator-Set compaction: one 1431 operational mechanism and two protocol mechanisms. The operational 1432 approach uses a clock sweep method. The protocol approaches use the 1433 concept of Solicit-Map-Requests and Map-Versioning. 1435 13.1. Clock Sweep 1437 The clock sweep approach uses planning in advance and the use of 1438 count-down TTLs to time out mappings that have already been cached. 1439 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So, 1440 there is a 24-hour window to time out old mappings. The following 1441 clock sweep procedure is used: 1443 1. 24 hours before a mapping change is to take effect, a network 1444 administrator configures the ETRs at a site to start the clock 1445 sweep window. 1447 2. During the clock sweep window, ETRs continue to send Map-Reply 1448 messages with the current (unchanged) mapping records. The TTL 1449 for these mappings is set to 1 hour. 1451 3. 24 hours later, all previous cache entries will have timed out, 1452 and any active cache entries will time out within 1 hour. During 1453 this 1-hour window, the ETRs continue to send Map-Reply messages 1454 with the current (unchanged) mapping records with the TTL set to 1455 1 minute. 1457 4. At the end of the 1-hour window, the ETRs will send Map-Reply 1458 messages with the new (changed) mapping records. So, any active 1459 caches can get the new mapping contents right away if not cached, 1460 or in 1 minute if they had the mapping cached. The new mappings 1461 are cached with a TTL equal to the TTL in the Map-Reply. 1463 13.2. Solicit-Map-Request (SMR) 1465 Soliciting a Map-Request is a selective way for ETRs, at the site 1466 where mappings change, to control the rate they receive requests for 1467 Map-Reply messages. SMRs are also used to tell remote ITRs to update 1468 the mappings they have cached. 1470 Since the ETRs don't keep track of remote ITRs that have cached their 1471 mappings, they do not know which ITRs need to have their mappings 1472 updated. As a result, an ETR will solicit Map-Requests (called an 1473 SMR message) from those sites to which it has been sending 1474 encapsulated data for the last minute. In particular, an ETR will 1475 send an SMR to an ITR to which it has recently sent encapsulated 1476 data. This can only occur when both ITR and ETR functionality reside 1477 in the same router. 1479 An SMR message is simply a bit set in a Map-Request message. An ITR 1480 or PITR will send a Map-Request when they receive an SMR message. 1481 Both the SMR sender and the Map-Request responder MUST rate-limit 1482 these messages. Rate-limiting can be implemented as a global rate- 1483 limiter or one rate-limiter per SMR destination. 1485 The following procedure shows how an SMR exchange occurs when a site 1486 is doing Locator-Set compaction for an EID-to-RLOC mapping: 1488 1. When the database mappings in an ETR change, the ETRs at the site 1489 begin to send Map-Requests with the SMR bit set for each Locator 1490 in each Map-Cache entry the ETR caches. 1492 2. A remote ITR that receives the SMR message will schedule sending 1493 a Map-Request message to the source locator address of the SMR 1494 message or to the mapping database system. A newly allocated 1495 random nonce is selected, and the EID-Prefix used is the one 1496 copied from the SMR message. If the source Locator is the only 1497 Locator in the cached Locator-Set, the remote ITR SHOULD send a 1498 Map-Request to the database mapping system just in case the 1499 single Locator has changed and may no longer be reachable to 1500 accept the Map-Request. 1502 3. The remote ITR MUST rate-limit the Map-Request until it gets a 1503 Map-Reply while continuing to use the cached mapping. When 1504 Map-Versioning as described in Section 13.3 is used, an SMR 1505 sender can detect if an ITR is using the most up-to-date database 1506 mapping. 1508 4. The ETRs at the site with the changed mapping will reply to the 1509 Map-Request with a Map-Reply message that has a nonce from the 1510 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate- 1511 limited. This is important to avoid Map-Reply implosion. 1513 5. The ETRs at the site with the changed mapping record the fact 1514 that the site that sent the Map-Request has received the new 1515 mapping data in the Map-Cache entry for the remote site so the 1516 Locator-Status-Bits are reflective of the new mapping for packets 1517 going to the remote site. The ETR then stops sending SMR 1518 messages. 1520 For security reasons, an ITR MUST NOT process unsolicited Map- 1521 Replies. To avoid Map-Cache entry corruption by a third party, a 1522 sender of an SMR-based Map-Request MUST be verified. If an ITR 1523 receives an SMR-based Map-Request and the source is not in the 1524 Locator-Set for the stored Map-Cache entry, then the responding Map- 1525 Request MUST be sent with an EID destination to the mapping database 1526 system. Since the mapping database system is a more secure way to 1527 reach an authoritative ETR, it will deliver the Map-Request to the 1528 authoritative source of the mapping data. 1530 When an ITR receives an SMR-based Map-Request for which it does not 1531 have a cached mapping for the EID in the SMR message, it may not send 1532 an SMR-invoked Map-Request. This scenario can occur when an ETR 1533 sends SMR messages to all Locators in the Locator-Set it has stored 1534 in its map-cache but the remote ITRs that receive the SMR may not be 1535 sending packets to the site. There is no point in updating the ITRs 1536 until they need to send, in which case they will send Map-Requests to 1537 obtain a Map-Cache entry. 1539 13.3. Database Map-Versioning 1541 When there is unidirectional packet flow between an ITR and ETR, and 1542 the EID-to-RLOC mappings change on the ETR, it needs to inform the 1543 ITR so encapsulation to a removed Locator can stop and can instead be 1544 started to a new Locator in the Locator-Set. 1546 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 1547 Number. This is known as the Destination Map-Version Number. ITRs 1548 include the Destination Map-Version Number in packets they 1549 encapsulate to the site. When an ETR decapsulates a packet and 1550 detects that the Destination Map-Version Number is less than the 1551 current version for its mapping, the SMR procedure described in 1552 Section 13.2 occurs. 1554 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 1555 Version Number. This is known as the Source Map-Version Number. 1556 When an ETR decapsulates a packet and detects that the Source Map- 1557 Version Number is greater than the last Map-Version Number sent in a 1558 Map-Reply from the ITR's site, the ETR will send a Map-Request to one 1559 of the ETRs for the source site. 1561 A Map-Version Number is used as a sequence number per EID-Prefix, so 1562 values that are greater are considered to be more recent. A value of 1563 0 for the Source Map-Version Number or the Destination Map-Version 1564 Number conveys no versioning information, and an ITR does no 1565 comparison with previously received Map-Version Numbers. 1567 A Map-Version Number can be included in Map-Register messages as 1568 well. This is a good way for the Map-Server to assure that all ETRs 1569 for a site registering to it will be synchronized according to Map- 1570 Version Number. 1572 See [RFC6834] for a more detailed analysis and description of 1573 Database Map-Versioning. 1575 14. Multicast Considerations 1577 A multicast group address, as defined in the original Internet 1578 architecture, is an identifier of a grouping of topologically 1579 independent receiver host locations. The address encoding itself 1580 does not determine the location of the receiver(s). The multicast 1581 routing protocol, and the network-based state the protocol creates, 1582 determine where the receivers are located. 1584 In the context of LISP, a multicast group address is both an EID and 1585 a Routing Locator. Therefore, no specific semantic or action needs 1586 to be taken for a destination address, as it would appear in an IP 1587 header. Therefore, a group address that appears in an inner IP 1588 header built by a source host will be used as the destination EID. 1589 The outer IP header (the destination Routing Locator address), 1590 prepended by a LISP router, can use the same group address as the 1591 destination Routing Locator, use a multicast or unicast Routing 1592 Locator obtained from a Mapping System lookup, or use other means to 1593 determine the group address mapping. 1595 With respect to the source Routing Locator address, the ITR prepends 1596 its own IP address as the source address of the outer IP header. 1597 Just like it would if the destination EID was a unicast address. 1598 This source Routing Locator address, like any other Routing Locator 1599 address, MUST be globally routable. 1601 There are two approaches for LISP-Multicast, one that uses native 1602 multicast routing in the underlay with no support from the Mapping 1603 System and the other that uses only unicast routing in the underlay 1604 with support from the Mapping System. See [RFC6831] and 1605 [I-D.ietf-lisp-signal-free-multicast], respectively, for details. 1606 Details for LISP-Multicast and interworking with non-LISP sites are 1607 described in [RFC6831] and [RFC6832]. 1609 15. Router Performance Considerations 1611 LISP is designed to be very "hardware-based forwarding friendly". A 1612 few implementation techniques can be used to incrementally implement 1613 LISP: 1615 o When a tunnel-encapsulated packet is received by an ETR, the outer 1616 destination address may not be the address of the router. This 1617 makes it challenging for the control plane to get packets from the 1618 hardware. This may be mitigated by creating special Forwarding 1619 Information Base (FIB) entries for the EID-Prefixes of EIDs served 1620 by the ETR (those for which the router provides an RLOC 1621 translation). These FIB entries are marked with a flag indicating 1622 that control-plane processing SHOULD be performed. The forwarding 1623 logic of testing for particular IP protocol number values is not 1624 necessary. There are a few proven cases where no changes to 1625 existing deployed hardware were needed to support the LISP data- 1626 plane. 1628 o On an ITR, prepending a new IP header consists of adding more 1629 octets to a MAC rewrite string and prepending the string as part 1630 of the outgoing encapsulation procedure. Routers that support 1631 Generic Routing Encapsulation (GRE) tunneling [RFC2784] or 6to4 1632 tunneling [RFC3056] may already support this action. 1634 o A packet's source address or interface the packet was received on 1635 can be used to select VRF (Virtual Routing/Forwarding). The VRF's 1636 routing table can be used to find EID-to-RLOC mappings. 1638 For performance issues related to map-cache management, see 1639 Section 19. 1641 16. Mobility Considerations 1643 There are several kinds of mobility, of which only some might be of 1644 concern to LISP. Essentially, they are as follows. 1646 16.1. Slow Mobility 1648 A site wishes to change its attachment points to the Internet, and 1649 its LISP Tunnel Routers will have new RLOCs when it changes upstream 1650 providers. Changes in EID-to-RLOC mappings for sites are expected to 1651 be handled by configuration, outside of LISP. 1653 An individual endpoint wishes to move but is not concerned about 1654 maintaining session continuity. Renumbering is involved. LISP can 1655 help with the issues surrounding renumbering [RFC4192] [LISA96] by 1656 decoupling the address space used by a site from the address spaces 1657 used by its ISPs [RFC4984]. 1659 16.2. Fast Mobility 1661 Fast endpoint mobility occurs when an endpoint moves relatively 1662 rapidly, changing its IP-layer network attachment point. Maintenance 1663 of session continuity is a goal. This is where the Mobile IPv4 1664 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used and 1665 primarily where interactions with LISP need to be explored, such as 1666 the mechanisms in [I-D.ietf-lisp-eid-mobility] when the EID moves but 1667 the RLOC is in the network infrastructure. 1669 In LISP, one possibility is to "glean" information. When a packet 1670 arrives, the ETR could examine the EID-to-RLOC mapping and use that 1671 mapping for all outgoing traffic to that EID. It can do this after 1672 performing a route-returnability check, to ensure that the new 1673 network location does have an internal route to that endpoint. 1674 However, this does not cover the case where an ITR (the node assigned 1675 the RLOC) at the mobile-node location has been compromised. 1677 Mobile IP packet exchange is designed for an environment in which all 1678 routing information is disseminated before packets can be forwarded. 1679 In order to allow the Internet to grow to support expected future 1680 use, we are moving to an environment where some information may have 1681 to be obtained after packets are in flight. Modifications to IP 1682 mobility should be considered in order to optimize the behavior of 1683 the overall system. Anything that decreases the number of new EID- 1684 to-RLOC mappings needed when a node moves, or maintains the validity 1685 of an EID-to-RLOC mapping for a longer time, is useful. 1687 In addition to endpoints, a network can be mobile, possibly changing 1688 xTRs. A "network" can be as small as a single router and as large as 1689 a whole site. This is different from site mobility in that it is 1690 fast and possibly short-lived, but different from endpoint mobility 1691 in that a whole prefix is changing RLOCs. However, the mechanisms 1692 are the same, and there is no new overhead in LISP. A map request 1693 for any endpoint will return a binding for the entire mobile prefix. 1695 If mobile networks become a more common occurrence, it may be useful 1696 to revisit the design of the mapping service and allow for dynamic 1697 updates of the database. 1699 The issue of interactions between mobility and LISP needs to be 1700 explored further. Specific improvements to the entire system will 1701 depend on the details of mapping mechanisms. Mapping mechanisms 1702 should be evaluated on how well they support session continuity for 1703 mobile nodes. See [I-D.ietf-lisp-predictive-rlocs] for more recent 1704 mechanisms which can provide near-zero packet loss during handoffs. 1706 16.3. LISP Mobile Node Mobility 1708 A mobile device can use the LISP infrastructure to achieve mobility 1709 by implementing the LISP encapsulation and decapsulation functions 1710 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 1711 node" can use topologically independent EID IP addresses that are not 1712 advertised into and do not impose a cost on the global routing 1713 system. These EIDs are maintained at the edges of the mapping system 1714 in LISP Map-Servers and Map-Resolvers) and are provided on demand to 1715 only the correspondents of the LISP mobile node. 1717 Refer to [I-D.ietf-lisp-mn] for more details for when the EID and 1718 RLOC are co-located in the roaming node. 1720 17. LISP xTR Placement and Encapsulation Methods 1722 This section will explore how and where ITRs and ETRs can be placed 1723 in the network and will discuss the pros and cons of each scenario. 1724 For a more detailed networkd design deployment recommendation, refer 1725 to [RFC7215]. 1727 There are two basic deployment tradeoffs to consider: centralized 1728 versus distributed caches; and flat, Recursive, or Re-encapsulating 1729 Tunneling. When deciding on centralized versus distributed caching, 1730 the following issues SHOULD be considered: 1732 o Are the xTRs spread out so that the caches are spread across all 1733 the memories of each router? A centralized cache is when an ITR 1734 keeps a cache for all the EIDs it is encapsulating to. The packet 1735 takes a direct path to the destination Locator. A distributed 1736 cache is when an ITR needs help from other Re-Encapsulating Tunnel 1737 Routers (RTRs) because it does not store all the cache entries for 1738 the EIDs it is encapsulating to. So, the packet takes a path 1739 through RTRs that have a different set of cache entries. 1741 o Should management "touch points" be minimized by only choosing a 1742 few xTRs, just enough for redundancy? 1744 o In general, using more ITRs doesn't increase management load, 1745 since caches are built and stored dynamically. On the other hand, 1746 using more ETRs does require more management, since EID-Prefix-to- 1747 RLOC mappings need to be explicitly configured. 1749 When deciding on flat, Recursive, or Re-Encapsulating Tunneling, the 1750 following issues SHOULD be considered: 1752 o Flat tunneling implements a single encapsulation path between the 1753 source site and destination site. This generally offers better 1754 paths between sources and destinations with a single encapsulation 1755 path. 1757 o Recursive Tunneling is when encapsulated traffic is again further 1758 encapsulated in another tunnel, either to implement VPNs or to 1759 perform Traffic Engineering. When doing VPN-based tunneling, the 1760 site has some control, since the site is prepending a new 1761 encapsulation header. In the case of TE-based tunneling, the site 1762 MAY have control if it is prepending a new tunnel header, but if 1763 the site's ISP is doing the TE, then the site has no control. 1764 Recursive Tunneling generally will result in suboptimal paths but 1765 with the benefit of steering traffic to parts of the network that 1766 have more resources available. 1768 o The technique of Re-Encapsulation ensures that packets only 1769 require one encapsulation header. So, if a packet needs to be re- 1770 routed, it is first decapsulated by the RTR and then Re- 1771 Encapsulated with a new encapsulation header using a new RLOC. 1773 The next sub-sections will examine where xTRs and RTRs can reside in 1774 the network. 1776 17.1. First-Hop/Last-Hop xTRs 1778 By locating xTRs close to hosts, the EID-Prefix set is at the 1779 granularity of an IP subnet. So, at the expense of more EID-Prefix- 1780 to-RLOC sets for the site, the caches in each xTR can remain 1781 relatively small. But caches always depend on the number of non- 1782 aggregated EID destination flows active through these xTRs. 1784 With more xTRs doing encapsulation, the increase in control traffic 1785 grows as well: since the EID granularity is greater, more Map- 1786 Requests and Map-Replies are traveling between more routers. 1788 The advantage of placing the caches and databases at these stub 1789 routers is that the products deployed in this part of the network 1790 have better price-memory ratios than their core router counterparts. 1791 Memory is typically less expensive in these devices, and fewer routes 1792 are stored (only IGP routes). These devices tend to have excess 1793 capacity, both for forwarding and routing states. 1795 LISP functionality can also be deployed in edge switches. These 1796 devices generally have layer-2 ports facing hosts and layer-3 ports 1797 facing the Internet. Spare capacity is also often available in these 1798 devices. 1800 17.2. Border/Edge xTRs 1802 Using Customer Edge (CE) routers for xTR placement allows the EID 1803 space associated with a site to be reachable via a small set of RLOCs 1804 assigned to the CE-based xTRs for that site. 1806 This offers the opposite benefit of the first-hop/last-hop xTR 1807 scenario: the number of mapping entries and network management touch 1808 points is reduced, allowing better scaling. 1810 One disadvantage is that fewer network resources are used to reach 1811 host endpoints, thereby centralizing the point-of-failure domain and 1812 creating network choke points at the CE xTR. 1814 Note that more than one CE xTR at a site can be configured with the 1815 same IP address. In this case, an RLOC is an anycast address. This 1816 allows resilience between the CE xTRs. That is, if a CE xTR fails, 1817 traffic is automatically routed to the other xTRs using the same 1818 anycast address. However, this comes with the disadvantage where the 1819 site cannot control the entrance point when the anycast route is 1820 advertised out from all border routers. Another disadvantage of 1821 using anycast Locators is the limited advertisement scope of /32 (or 1822 /128 for IPv6) routes. 1824 17.3. ISP Provider Edge (PE) xTRs 1826 The use of ISP PE routers as xTRs is not the typical deployment 1827 scenario envisioned in this specification. This section attempts to 1828 capture some of the reasoning behind this preference for implementing 1829 LISP on CE routers. 1831 The use of ISP PE routers for xTR placement gives an ISP, rather than 1832 a site, control over the location of the ETRs. That is, the ISP can 1833 decide whether the xTRs are in the destination site (in either CE 1834 xTRs or last-hop xTRs within a site) or at other PE edges. The 1835 advantage of this case is that two encapsulation headers can be 1836 avoided. By having the PE be the first router on the path to 1837 encapsulate, it can choose a TE path first, and the ETR can 1838 decapsulate and Re-Encapsulate for a new encapsuluation path to the 1839 destination end site. 1841 An obvious disadvantage is that the end site has no control over 1842 where its packets flow or over the RLOCs used. Other disadvantages 1843 include difficulty in synchronizing path liveness updates between CE 1844 and PE routers. 1846 As mentioned in earlier sections, a combination of these scenarios is 1847 possible at the expense of extra packet header overhead; if both site 1848 and provider want control, then Recursive or Re-Encapsulating Tunnels 1849 are used. 1851 17.4. LISP Functionality with Conventional NATs 1853 LISP routers can be deployed behind Network Address Translator (NAT) 1854 devices to provide the same set of packet services hosts have today 1855 when they are addressed out of private address space. 1857 It is important to note that a locator address in any LISP control 1858 message MUST be a routable address and therefore [RFC1918] addresses 1859 SHOULD only be presence when running in a local environment. When a 1860 LISP xTR is configured with private RLOC addresses and resides behind 1861 a NAT device and desires to communicate on the Internet, the private 1862 addresses MUST be used only in the outer IP header so the NAT device 1863 can translate properly. Otherwise, EID addresses MUST be translated 1864 before encapsulation is performed when LISP VPNs are not in use. 1865 Both NAT translation and LISP encapsulation functions could be co- 1866 located in the same device. 1868 17.5. Packets Egressing a LISP Site 1870 When a LISP site is using two ITRs for redundancy, the failure of one 1871 ITR will likely shift outbound traffic to the second. This second 1872 ITR's cache MAY not be populated with the same EID-to-RLOC mapping 1873 entries as the first. If this second ITR does not have these 1874 mappings, traffic will be dropped while the mappings are retrieved 1875 from the mapping system. The retrieval of these messages may 1876 increase the load of requests being sent into the mapping system. 1878 18. Traceroute Considerations 1880 When a source host in a LISP site initiates a traceroute to a 1881 destination host in another LISP site, it is highly desirable for it 1882 to see the entire path. Since packets are encapsulated from the ITR 1883 to the ETR, the hop across the tunnel could be viewed as a single 1884 hop. However, LISP traceroute will provide the entire path so the 1885 user can see 3 distinct segments of the path from a source LISP host 1886 to a destination LISP host: 1888 Segment 1 (in source LISP site based on EIDs): 1890 source host ---> first hop ... next hop ---> ITR 1892 Segment 2 (in the core network based on RLOCs): 1894 ITR ---> next hop ... next hop ---> ETR 1896 Segment 3 (in the destination LISP site based on EIDs): 1898 ETR ---> next hop ... last hop ---> destination host 1900 For segment 1 of the path, ICMP Time Exceeded messages are returned 1901 in the normal manner as they are today. The ITR performs a TTL 1902 decrement and tests for 0 before encapsulating. Therefore, the ITR's 1903 hop is seen by the traceroute source as having an EID address (the 1904 address of the site-facing interface). 1906 For segment 2 of the path, ICMP Time Exceeded messages are returned 1907 to the ITR because the TTL decrement to 0 is done on the outer 1908 header, so the destinations of the ICMP messages are the ITR RLOC 1909 address and the source RLOC address of the encapsulated traceroute 1910 packet. The ITR looks inside of the ICMP payload to inspect the 1911 traceroute source so it can return the ICMP message to the address of 1912 the traceroute client and also retain the core router IP address in 1913 the ICMP message. This is so the traceroute client can display the 1914 core router address (the RLOC address) in the traceroute output. The 1915 ETR returns its RLOC address and responds to the TTL decrement to 0, 1916 as the previous core routers did. 1918 For segment 3, the next-hop router downstream from the ETR will be 1919 decrementing the TTL for the packet that was encapsulated, sent into 1920 the core, decapsulated by the ETR, and forwarded because it isn't the 1921 final destination. If the TTL is decremented to 0, any router on the 1922 path to the destination of the traceroute, including the next-hop 1923 router or destination, will send an ICMP Time Exceeded message to the 1924 source EID of the traceroute client. The ICMP message will be 1925 encapsulated by the local ITR and sent back to the ETR in the 1926 originated traceroute source site, where the packet will be delivered 1927 to the host. 1929 18.1. IPv6 Traceroute 1931 IPv6 traceroute follows the procedure described above, since the 1932 entire traceroute data packet is included in the ICMP Time Exceeded 1933 message payload. Therefore, only the ITR needs to pay special 1934 attention to forwarding ICMP messages back to the traceroute source. 1936 18.2. IPv4 Traceroute 1938 For IPv4 traceroute, we cannot follow the above procedure, since IPv4 1939 ICMP Time Exceeded messages only include the invoking IP header and 1940 8 octets that follow the IP header. Therefore, when a core router 1941 sends an IPv4 Time Exceeded message to an ITR, all the ITR has in the 1942 ICMP payload is the encapsulated header it prepended, followed by a 1943 UDP header. The original invoking IP header, and therefore the 1944 identity of the traceroute source, is lost. 1946 The solution we propose to solve this problem is to cache traceroute 1947 IPv4 headers in the ITR and to match them up with corresponding IPv4 1948 Time Exceeded messages received from core routers and the ETR. The 1949 ITR will use a circular buffer for caching the IPv4 and UDP headers 1950 of traceroute packets. It will select a 16-bit number as a key to 1951 find them later when the IPv4 Time Exceeded messages are received. 1952 When an ITR encapsulates an IPv4 traceroute packet, it will use the 1953 16-bit number as the UDP source port in the encapsulating header. 1954 When the ICMP Time Exceeded message is returned to the ITR, the UDP 1955 header of the encapsulating header is present in the ICMP payload, 1956 thereby allowing the ITR to find the cached headers for the 1957 traceroute source. The ITR puts the cached headers in the payload 1958 and sends the ICMP Time Exceeded message to the traceroute source 1959 retaining the source address of the original ICMP Time Exceeded 1960 message (a core router or the ETR of the site of the traceroute 1961 destination). 1963 The signature of a traceroute packet comes in two forms. The first 1964 form is encoded as a UDP message where the destination port is 1965 inspected for a range of values. The second form is encoded as an 1966 ICMP message where the IP identification field is inspected for a 1967 well-known value. 1969 18.3. Traceroute Using Mixed Locators 1971 When either an IPv4 traceroute or IPv6 traceroute is originated and 1972 the ITR encapsulates it in the other address family header, one 1973 cannot get all 3 segments of the traceroute. Segment 2 of the 1974 traceroute cannot be conveyed to the traceroute source, since it is 1975 expecting addresses from intermediate hops in the same address format 1976 for the type of traceroute it originated. Therefore, in this case, 1977 segment 2 will make the tunnel look like one hop. All the ITR has to 1978 do to make this work is to not copy the inner TTL to the outer, 1979 encapsulating header's TTL when a traceroute packet is encapsulated 1980 using an RLOC from a different address family. This will cause no 1981 TTL decrement to 0 to occur in core routers between the ITR and ETR. 1983 19. Security Considerations 1985 Security considerations for LISP are discussed in [RFC7833], in 1986 addition [I-D.ietf-lisp-sec] provides authentication and integrity to 1987 LISP mappings. 1989 A complete LISP threat analysis can be found in [RFC7835], in what 1990 follows we provide a summary. 1992 The optional mechanisms of gleaning is offered to directly obtain a 1993 mapping from the LISP encapsulated packets. Specifically, an xTR can 1994 learn the EID-to-RLOC mapping by inspecting the source RLOC and 1995 source EID of an encapsulated packet, and insert this new mapping 1996 into its map-cache. An off-path attacker can spoof the source EID 1997 address to divert the traffic sent to the victim's spoofed EID. If 1998 the attacker spoofs the source RLOC, it can mount a DoS attack by 1999 redirecting traffic to the spoofed victim;s RLOC, potentially 2000 overloading it. 2002 The LISP Data-Plane defines several mechanisms to monitor RLOC data- 2003 plane reachability, in this context Locator-Status Bits, Nonce- 2004 Present and Echo-Nonce bits of the LISP encapsulation header can be 2005 manipulated by an attacker to mount a DoS attack. An off-path 2006 attacker able to spoof the RLOC of a victim's xTR can manipulate such 2007 mechanisms to declare a set of RLOCs unreachable. This can be used 2008 also, for instance, to declare only one RLOC reachable with the aim 2009 of overload it. 2011 Map-Versioning is a data-plane mechanism used to signal a peering xTR 2012 that a local EID-to-RLOC mapping has been updated, so that the 2013 peering xTR uses LISP Control-Plane signaling message to retrieve a 2014 fresh mapping. This can be used by an attacker to forge the map- 2015 versioning field of a LISP encapsulated header and force an excessive 2016 amount of signaling between xTRs that may overload them. 2018 Most of the attack vectors can be mitigated with careful deployment 2019 and configuration, information learned opportunistically (such as LSB 2020 or gleaning) SHOULD be verified with other reachability mechanisms. 2021 In addition, systematic rate-limitation and filtering is an effective 2022 technique to mitigate attacks that aim to overload the control-plane. 2024 20. Network Management Considerations 2026 Considerations for network management tools exist so the LISP 2027 protocol suite can be operationally managed. These mechanisms can be 2028 found in [RFC7052] and [RFC6835]. 2030 21. IANA Considerations 2032 This section provides guidance to the Internet Assigned Numbers 2033 Authority (IANA) regarding registration of values related to this 2034 data-plane LISP specification, in accordance with BCP 26 [RFC8126]. 2036 21.1. LISP UDP Port Numbers 2038 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2039 lisp-data and lisp-control operation, respectively. IANA has updated 2040 the description for UDP ports 4341 and 4342 as follows: 2042 lisp-data 4341 udp LISP Data Packets 2043 lisp-control 4342 udp LISP Control Packets 2045 22. References 2047 22.1. Normative References 2049 [I-D.ietf-lisp-rfc6833bis] 2050 Fuller, V., Farinacci, D., and A. Cabellos-Aparicio, 2051 "Locator/ID Separation Protocol (LISP) Control-Plane", 2052 draft-ietf-lisp-rfc6833bis-07 (work in progress), December 2053 2017. 2055 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2056 DOI 10.17487/RFC0768, August 1980, 2057 . 2059 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2060 DOI 10.17487/RFC0791, September 1981, 2061 . 2063 [RFC1918] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G., 2064 and E. Lear, "Address Allocation for Private Internets", 2065 BCP 5, RFC 1918, DOI 10.17487/RFC1918, February 1996, 2066 . 2068 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2069 Requirement Levels", BCP 14, RFC 2119, 2070 DOI 10.17487/RFC2119, March 1997, 2071 . 2073 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 2074 "Definition of the Differentiated Services Field (DS 2075 Field) in the IPv4 and IPv6 Headers", RFC 2474, 2076 DOI 10.17487/RFC2474, December 1998, 2077 . 2079 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2080 of Explicit Congestion Notification (ECN) to IP", 2081 RFC 3168, DOI 10.17487/RFC3168, September 2001, 2082 . 2084 [RFC4086] Eastlake 3rd, D., Schiller, J., and S. Crocker, 2085 "Randomness Requirements for Security", BCP 106, RFC 4086, 2086 DOI 10.17487/RFC4086, June 2005, 2087 . 2089 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 2090 (CIDR): The Internet Address Assignment and Aggregation 2091 Plan", BCP 122, RFC 4632, DOI 10.17487/RFC4632, August 2092 2006, . 2094 [RFC5944] Perkins, C., Ed., "IP Mobility Support for IPv4, Revised", 2095 RFC 5944, DOI 10.17487/RFC5944, November 2010, 2096 . 2098 [RFC6275] Perkins, C., Ed., Johnson, D., and J. Arkko, "Mobility 2099 Support in IPv6", RFC 6275, DOI 10.17487/RFC6275, July 2100 2011, . 2102 [RFC7833] Howlett, J., Hartman, S., and A. Perez-Mendez, Ed., "A 2103 RADIUS Attribute, Binding, Profiles, Name Identifier 2104 Format, and Confirmation Methods for the Security 2105 Assertion Markup Language (SAML)", RFC 7833, 2106 DOI 10.17487/RFC7833, May 2016, 2107 . 2109 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 2110 Writing an IANA Considerations Section in RFCs", BCP 26, 2111 RFC 8126, DOI 10.17487/RFC8126, June 2017, 2112 . 2114 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2115 (IPv6) Specification", STD 86, RFC 8200, 2116 DOI 10.17487/RFC8200, July 2017, 2117 . 2119 22.2. Informative References 2121 [AFN] IANA, "Address Family Numbers", August 2016, 2122 . 2124 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed", 2125 1999, 2126 . 2128 [I-D.ietf-lisp-eid-mobility] 2129 Portoles-Comeras, M., Ashtaputre, V., Moreno, V., Maino, 2130 F., and D. Farinacci, "LISP L2/L3 EID Mobility Using a 2131 Unified Control Plane", draft-ietf-lisp-eid-mobility-01 2132 (work in progress), November 2017. 2134 [I-D.ietf-lisp-introduction] 2135 Cabellos-Aparicio, A. and D. Saucez, "An Architectural 2136 Introduction to the Locator/ID Separation Protocol 2137 (LISP)", draft-ietf-lisp-introduction-13 (work in 2138 progress), April 2015. 2140 [I-D.ietf-lisp-mn] 2141 Farinacci, D., Lewis, D., Meyer, D., and C. White, "LISP 2142 Mobile Node", draft-ietf-lisp-mn-01 (work in progress), 2143 October 2017. 2145 [I-D.ietf-lisp-predictive-rlocs] 2146 Farinacci, D. and P. Pillay-Esnault, "LISP Predictive 2147 RLOCs", draft-ietf-lisp-predictive-rlocs-01 (work in 2148 progress), November 2017. 2150 [I-D.ietf-lisp-sec] 2151 Maino, F., Ermagan, V., Cabellos-Aparicio, A., and D. 2152 Saucez, "LISP-Security (LISP-SEC)", draft-ietf-lisp-sec-14 2153 (work in progress), October 2017. 2155 [I-D.ietf-lisp-signal-free-multicast] 2156 Moreno, V. and D. Farinacci, "Signal-Free LISP Multicast", 2157 draft-ietf-lisp-signal-free-multicast-07 (work in 2158 progress), November 2017. 2160 [I-D.ietf-lisp-vpn] 2161 Moreno, V. and D. Farinacci, "LISP Virtual Private 2162 Networks (VPNs)", draft-ietf-lisp-vpn-01 (work in 2163 progress), November 2017. 2165 [I-D.meyer-loc-id-implications] 2166 Meyer, D. and D. Lewis, "Architectural Implications of 2167 Locator/ID Separation", draft-meyer-loc-id-implications-01 2168 (work in progress), January 2009. 2170 [LISA96] Lear, E., Tharp, D., Katinsky, J., and J. Coffin, 2171 "Renumbering: Threat or Menace?", Usenix Tenth System 2172 Administration Conference (LISA 96), October 1996. 2174 [OPENLISP] 2175 Iannone, L., Saucez, D., and O. Bonaventure, "OpenLISP 2176 Implementation Report", Work in Progress, July 2008. 2178 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2179 STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, 2180 . 2182 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2183 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2184 DOI 10.17487/RFC2784, March 2000, 2185 . 2187 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 2188 via IPv4 Clouds", RFC 3056, DOI 10.17487/RFC3056, February 2189 2001, . 2191 [RFC3232] Reynolds, J., Ed., "Assigned Numbers: RFC 1700 is Replaced 2192 by an On-line Database", RFC 3232, DOI 10.17487/RFC3232, 2193 January 2002, . 2195 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2196 A., Peterson, J., Sparks, R., Handley, M., and E. 2197 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2198 DOI 10.17487/RFC3261, June 2002, 2199 . 2201 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2202 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2203 DOI 10.17487/RFC4192, September 2005, 2204 . 2206 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2207 Optimization for Mobile IPv6", RFC 4866, 2208 DOI 10.17487/RFC4866, May 2007, 2209 . 2211 [RFC4984] Meyer, D., Ed., Zhang, L., Ed., and K. Fall, Ed., "Report 2212 from the IAB Workshop on Routing and Addressing", 2213 RFC 4984, DOI 10.17487/RFC4984, September 2007, 2214 . 2216 [RFC6831] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, "The 2217 Locator/ID Separation Protocol (LISP) for Multicast 2218 Environments", RFC 6831, DOI 10.17487/RFC6831, January 2219 2013, . 2221 [RFC6832] Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 2222 "Interworking between Locator/ID Separation Protocol 2223 (LISP) and Non-LISP Sites", RFC 6832, 2224 DOI 10.17487/RFC6832, January 2013, 2225 . 2227 [RFC6834] Iannone, L., Saucez, D., and O. Bonaventure, "Locator/ID 2228 Separation Protocol (LISP) Map-Versioning", RFC 6834, 2229 DOI 10.17487/RFC6834, January 2013, 2230 . 2232 [RFC6835] Farinacci, D. and D. Meyer, "The Locator/ID Separation 2233 Protocol Internet Groper (LIG)", RFC 6835, 2234 DOI 10.17487/RFC6835, January 2013, 2235 . 2237 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2238 UDP Checksums for Tunneled Packets", RFC 6935, 2239 DOI 10.17487/RFC6935, April 2013, 2240 . 2242 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 2243 for the Use of IPv6 UDP Datagrams with Zero Checksums", 2244 RFC 6936, DOI 10.17487/RFC6936, April 2013, 2245 . 2247 [RFC7052] Schudel, G., Jain, A., and V. Moreno, "Locator/ID 2248 Separation Protocol (LISP) MIB", RFC 7052, 2249 DOI 10.17487/RFC7052, October 2013, 2250 . 2252 [RFC7215] Jakab, L., Cabellos-Aparicio, A., Coras, F., Domingo- 2253 Pascual, J., and D. Lewis, "Locator/Identifier Separation 2254 Protocol (LISP) Network Element Deployment 2255 Considerations", RFC 7215, DOI 10.17487/RFC7215, April 2256 2014, . 2258 [RFC7835] Saucez, D., Iannone, L., and O. Bonaventure, "Locator/ID 2259 Separation Protocol (LISP) Threat Analysis", RFC 7835, 2260 DOI 10.17487/RFC7835, April 2016, 2261 . 2263 [RFC8060] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 2264 Address Format (LCAF)", RFC 8060, DOI 10.17487/RFC8060, 2265 February 2017, . 2267 [RFC8061] Farinacci, D. and B. Weis, "Locator/ID Separation Protocol 2268 (LISP) Data-Plane Confidentiality", RFC 8061, 2269 DOI 10.17487/RFC8061, February 2017, 2270 . 2272 [RFC8111] Fuller, V., Lewis, D., Ermagan, V., Jain, A., and A. 2273 Smirnov, "Locator/ID Separation Protocol Delegated 2274 Database Tree (LISP-DDT)", RFC 8111, DOI 10.17487/RFC8111, 2275 May 2017, . 2277 Appendix A. Acknowledgments 2279 An initial thank you goes to Dave Oran for planting the seeds for the 2280 initial ideas for LISP. His consultation continues to provide value 2281 to the LISP authors. 2283 A special and appreciative thank you goes to Noel Chiappa for 2284 providing architectural impetus over the past decades on separation 2285 of location and identity, as well as detailed reviews of the LISP 2286 architecture and documents, coupled with enthusiasm for making LISP a 2287 practical and incremental transition for the Internet. 2289 The authors would like to gratefully acknowledge many people who have 2290 contributed discussions and ideas to the making of this proposal. 2291 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 2292 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 2293 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 2294 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 2295 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 2296 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 2297 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 2298 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 2299 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 2300 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 2301 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 2302 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 2303 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 2304 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 2305 Ermagan, Fabio Maino, Victor Moreno, Chris White, Clarence Filsfils, 2306 Alia Atlas, Florin Coras and Alberto Rodriguez. 2308 This work originated in the Routing Research Group (RRG) of the IRTF. 2309 An individual submission was converted into the IETF LISP working 2310 group document that became this RFC. 2312 The LISP working group would like to give a special thanks to Jari 2313 Arkko, the Internet Area AD at the time that the set of LISP 2314 documents were being prepared for IESG last call, and for his 2315 meticulous reviews and detailed commentaries on the 7 working group 2316 last call documents progressing toward standards-track RFCs. 2318 Appendix B. Document Change Log 2320 [RFC Editor: Please delete this section on publication as RFC.] 2322 B.1. Changes to draft-ietf-lisp-rfc6830bis-08 2324 o Posted January 2018. 2326 o Remove references to research work for any protocol mechanisms. 2328 o Document scanned to make sure it is RFC 2119 compliant. 2330 o Made changes to reflect comments from document WG shepherd Luigi 2331 Iannone. 2333 o Ran IDNITs on the document. 2335 B.2. Changes to draft-ietf-lisp-rfc6830bis-07 2337 o Posted November 2017. 2339 o Rephrase how Instance-IDs are used and don't refer to [RFC1918] 2340 addresses. 2342 B.3. Changes to draft-ietf-lisp-rfc6830bis-06 2344 o Posted October 2017. 2346 o Put RTR definition before it is used. 2348 o Rename references that are now working group drafts. 2350 o Remove "EIDs MUST NOT be used as used by a host to refer to other 2351 hosts. Note that EID blocks MAY LISP RLOCs". 2353 o Indicate what address-family can appear in data packets. 2355 o ETRs may, rather than will, be the ones to send Map-Replies. 2357 o Recommend, rather than mandate, max encapsulation headers to 2. 2359 o Reference VPN draft when introducing Instance-ID. 2361 o Indicate that SMRs can be sent when ITR/ETR are in the same node. 2363 o Clarify when private addreses can be used. 2365 B.4. Changes to draft-ietf-lisp-rfc6830bis-05 2367 o Posted August 2017. 2369 o Make it clear that a Reencapsulating Tunnel Router is an RTR. 2371 B.5. Changes to draft-ietf-lisp-rfc6830bis-04 2373 o Posted July 2017. 2375 o Changed reference of IPv6 RFC2460 to RFC8200. 2377 o Indicate that the applicability statement for UDP zero checksums 2378 over IPv6 adheres to RFC6936. 2380 B.6. Changes to draft-ietf-lisp-rfc6830bis-03 2382 o Posted May 2017. 2384 o Move the control-plane related codepoints in the IANA 2385 Considerations section to RFC6833bis. 2387 B.7. Changes to draft-ietf-lisp-rfc6830bis-02 2389 o Posted April 2017. 2391 o Reflect some editorial comments from Damien Sausez. 2393 B.8. Changes to draft-ietf-lisp-rfc6830bis-01 2395 o Posted March 2017. 2397 o Include references to new RFCs published. 2399 o Change references from RFC6833 to RFC6833bis. 2401 o Clarified LCAF text in the IANA section. 2403 o Remove references to "experimental". 2405 B.9. Changes to draft-ietf-lisp-rfc6830bis-00 2407 o Posted December 2016. 2409 o Created working group document from draft-farinacci-lisp 2410 -rfc6830-00 individual submission. No other changes made. 2412 Authors' Addresses 2413 Dino Farinacci 2414 Cisco Systems 2415 Tasman Drive 2416 San Jose, CA 95134 2417 USA 2419 EMail: farinacci@gmail.com 2421 Vince Fuller 2422 Cisco Systems 2423 Tasman Drive 2424 San Jose, CA 95134 2425 USA 2427 EMail: vince.fuller@gmail.com 2429 Dave Meyer 2430 Cisco Systems 2431 170 Tasman Drive 2432 San Jose, CA 2433 USA 2435 EMail: dmm@1-4-5.net 2437 Darrel Lewis 2438 Cisco Systems 2439 170 Tasman Drive 2440 San Jose, CA 2441 USA 2443 EMail: darlewis@cisco.com 2445 Albert Cabellos 2446 UPC/BarcelonaTech 2447 Campus Nord, C. Jordi Girona 1-3 2448 Barcelona, Catalunya 2449 Spain 2451 EMail: acabello@ac.upc.edu