idnits 2.17.1 draft-ietf-lisp-rfc6830bis-03.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 : ---------------------------------------------------------------------------- ** The abstract seems to contain references ([I-D.ietf-lisp-RFC6833bis]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (May 2, 2017) is 2551 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC2404' is defined on line 2119, but no explicit reference was found in the text == Unused Reference: 'RFC4868' is defined on line 2146, but no explicit reference was found in the text == Unused Reference: 'RFC5496' is defined on line 2156, but no explicit reference was found in the text == Unused Reference: 'RFC6115' is defined on line 2165, but no explicit reference was found in the text == Unused Reference: 'RFC6836' is defined on line 2178, but no explicit reference was found in the text == Unused Reference: 'RFC7214' is defined on line 2188, but no explicit reference was found in the text == Unused Reference: 'RFC4107' is defined on line 2278, but no explicit reference was found in the text == Unused Reference: 'RFC6480' is defined on line 2297, but no explicit reference was found in the text == Unused Reference: 'RFC6518' is defined on line 2301, but no explicit reference was found in the text == Unused Reference: 'RFC6837' is defined on line 2322, but no explicit reference was found in the text ** Downref: Normative reference to an Experimental draft: draft-ietf-lisp-ddt (ref. 'I-D.ietf-lisp-ddt') == Outdated reference: A later version (-15) exists of draft-ietf-lisp-introduction-13 ** Downref: Normative reference to an Informational draft: draft-ietf-lisp-introduction (ref. 'I-D.ietf-lisp-introduction') == Outdated reference: A later version (-31) exists of draft-ietf-lisp-rfc6833bis-03 == Outdated reference: A later version (-29) exists of draft-ietf-lisp-sec-12 ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Downref: Normative reference to an Informational RFC: RFC 3232 ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) ** Downref: Normative reference to an Informational RFC: RFC 6115 ** Obsolete normative reference: RFC 6834 (Obsoleted by RFC 9302) ** Downref: Normative reference to an Experimental RFC: RFC 6836 ** Downref: Normative reference to an Experimental RFC: RFC 7052 ** Downref: Normative reference to an Experimental RFC: RFC 7215 ** Downref: Normative reference to an Informational RFC: RFC 7835 ** Downref: Normative reference to an Experimental RFC: RFC 8061 == Outdated reference: A later version (-02) exists of draft-farinacci-lisp-predictive-rlocs-01 == Outdated reference: A later version (-15) exists of draft-ietf-lisp-mn-00 == Outdated reference: A later version (-09) exists of draft-ietf-lisp-signal-free-multicast-03 Summary: 13 errors (**), 0 flaws (~~), 17 warnings (==), 1 comment (--). 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: November 3, 2017 D. Lewis 6 Cisco Systems 7 A. Cabellos (Ed.) 8 UPC/BarcelonaTech 9 May 2, 2017 11 The Locator/ID Separation Protocol (LISP) 12 draft-ietf-lisp-rfc6830bis-03 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. The 23 map-cache is populated by the LISP Control-Plane protocol 24 [I-D.ietf-lisp-rfc6833bis]. 26 LISP requires no change to either host protocol stacks or to underlay 27 routers and offers Traffic Engineering, multihoming and mobility, 28 among other features. 30 Status of This Memo 32 This Internet-Draft is submitted in full conformance with the 33 provisions of BCP 78 and BCP 79. 35 Internet-Drafts are working documents of the Internet Engineering 36 Task Force (IETF). Note that other groups may also distribute 37 working documents as Internet-Drafts. The list of current Internet- 38 Drafts is at http://datatracker.ietf.org/drafts/current/. 40 Internet-Drafts are draft documents valid for a maximum of six months 41 and may be updated, replaced, or obsoleted by other documents at any 42 time. It is inappropriate to use Internet-Drafts as reference 43 material or to cite them other than as "work in progress." 45 This Internet-Draft will expire on November 3, 2017. 47 Copyright Notice 49 Copyright (c) 2017 IETF Trust and the persons identified as the 50 document authors. All rights reserved. 52 This document is subject to BCP 78 and the IETF Trust's Legal 53 Provisions Relating to IETF Documents 54 (http://trustee.ietf.org/license-info) in effect on the date of 55 publication of this document. Please review these documents 56 carefully, as they describe your rights and restrictions with respect 57 to this document. Code Components extracted from this document must 58 include Simplified BSD License text as described in Section 4.e of 59 the Trust Legal Provisions and are provided without warranty as 60 described in the Simplified BSD License. 62 Table of Contents 64 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 65 2. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 66 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 4 67 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . 9 68 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . 11 69 5. LISP Encapsulation Details . . . . . . . . . . . . . . . . . 13 70 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 14 71 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 15 72 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . 16 73 6. LISP EID-to-RLOC Map-Cache . . . . . . . . . . . . . . . . . 20 74 7. Dealing with Large Encapsulated Packets . . . . . . . . . . . 20 75 7.1. A Stateless Solution to MTU Handling . . . . . . . . . . 21 76 7.2. A Stateful Solution to MTU Handling . . . . . . . . . . . 22 77 8. Using Virtualization and Segmentation with LISP . . . . . . . 22 78 9. Routing Locator Selection . . . . . . . . . . . . . . . . . . 23 79 10. Routing Locator Reachability . . . . . . . . . . . . . . . . 24 80 10.1. Echo Nonce Algorithm . . . . . . . . . . . . . . . . . . 27 81 10.2. RLOC-Probing Algorithm . . . . . . . . . . . . . . . . . 28 82 11. EID Reachability within a LISP Site . . . . . . . . . . . . . 29 83 12. Routing Locator Hashing . . . . . . . . . . . . . . . . . . . 30 84 13. Changing the Contents of EID-to-RLOC Mappings . . . . . . . . 31 85 13.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . . 32 86 13.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . . 32 87 13.3. Database Map-Versioning . . . . . . . . . . . . . . . . 34 88 14. Multicast Considerations . . . . . . . . . . . . . . . . . . 34 89 15. Router Performance Considerations . . . . . . . . . . . . . . 35 90 16. Mobility Considerations . . . . . . . . . . . . . . . . . . . 36 91 16.1. Slow Mobility . . . . . . . . . . . . . . . . . . . . . 36 92 16.2. Fast Mobility . . . . . . . . . . . . . . . . . . . . . 36 93 16.3. LISP Mobile Node Mobility . . . . . . . . . . . . . . . 37 94 17. LISP xTR Placement and Encapsulation Methods . . . . . . . . 38 95 17.1. First-Hop/Last-Hop xTRs . . . . . . . . . . . . . . . . 39 96 17.2. Border/Edge xTRs . . . . . . . . . . . . . . . . . . . . 39 97 17.3. ISP Provider Edge (PE) xTRs . . . . . . . . . . . . . . 40 98 17.4. LISP Functionality with Conventional NATs . . . . . . . 40 99 17.5. Packets Egressing a LISP Site . . . . . . . . . . . . . 41 100 18. Traceroute Considerations . . . . . . . . . . . . . . . . . . 41 101 18.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . 42 102 18.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . 42 103 18.3. Traceroute Using Mixed Locators . . . . . . . . . . . . 43 104 19. Security Considerations . . . . . . . . . . . . . . . . . . . 43 105 20. Network Management Considerations . . . . . . . . . . . . . . 44 106 21. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 44 107 21.1. LISP UDP Port Numbers . . . . . . . . . . . . . . . . . 44 108 22. References . . . . . . . . . . . . . . . . . . . . . . . . . 44 109 22.1. Normative References . . . . . . . . . . . . . . . . . . 44 110 22.2. Informative References . . . . . . . . . . . . . . . . . 47 111 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 51 112 Appendix B. Document Change Log . . . . . . . . . . . . . . . . 51 113 B.1. Changes to draft-ietf-lisp-rfc6830bis-03 . . . . . . . . 52 114 B.2. Changes to draft-ietf-lisp-rfc6830bis-02 . . . . . . . . 52 115 B.3. Changes to draft-ietf-lisp-rfc6830bis-01 . . . . . . . . 52 116 B.4. Changes to draft-ietf-lisp-rfc6830bis-00 . . . . . . . . 52 117 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 52 119 1. Introduction 121 This document describes the Locator/Identifier Separation Protocol 122 (LISP). LISP is an encapsulation protocol built around the 123 fundamental idea of separating the topological location of a network 124 attachment point from the node's identity [CHIAPPA]. As a result 125 LISP creates two namespaces: Endpoint Identifiers (EIDs), that are 126 used to identify end-hosts (e.g., nodes or Virtual Machines) and 127 routable Routing Locators (RLOCs), used to identify network 128 attachment points. LISP then defines functions for mapping between 129 the two namespaces and for encapsulating traffic originated by 130 devices using non-routable EIDs for transport across a network 131 infrastructure that routes and forwards using RLOCs. 133 LISP is an overlay protocol that separates control from data-plane, 134 this document specifies the data-plane, how LISP-capable routers 135 (Tunnel Routers) exchange packets by encapsulating them to the 136 appropriate location. Tunnel routers are equipped with a cache, 137 called map-cache, that contains EID-to-RLOC mappings. The map-cache 138 is populated using the LISP Control-Plane protocol 139 [I-D.ietf-lisp-rfc6833bis]. 141 LISP does not require changes to either host protocol stack or to 142 underlay routers. By separating the EID from the RLOC space, LISP 143 offers native Traffic Engineering, multihoming and mobility, among 144 other features. 146 Creation of LISP was initially motivated by discussions during the 147 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in 148 October 2006 (see [RFC4984]) 150 This document specifies the LISP data-plane encapsulation and other 151 LISP forwarding node functionality while [I-D.ietf-lisp-rfc6833bis] 152 specifies the LISP control plane. LISP deployment guidelines can be 153 found in [RFC7215] and [RFC6835] describes considerations for network 154 operational management. Finally, [I-D.ietf-lisp-introduction] 155 describes the LISP architecture. 157 2. Requirements Notation 159 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 160 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 161 document are to be interpreted as described in [RFC2119]. 163 3. Definition of Terms 165 Provider-Independent (PI) Addresses: PI addresses are an address 166 block assigned from a pool where blocks are not associated with 167 any particular location in the network (e.g., from a particular 168 service provider) and are therefore not topologically aggregatable 169 in the routing system. 171 Provider-Assigned (PA) Addresses: PA addresses are an address block 172 assigned to a site by each service provider to which a site 173 connects. Typically, each block is a sub-block of a service 174 provider Classless Inter-Domain Routing (CIDR) [RFC4632] block and 175 is aggregated into the larger block before being advertised into 176 the global Internet. Traditionally, IP multihoming has been 177 implemented by each multihomed site acquiring its own globally 178 visible prefix. LISP uses only topologically assigned and 179 aggregatable address blocks for RLOCs, eliminating this 180 demonstrably non-scalable practice. 182 Routing Locator (RLOC): An RLOC is an IPv4 [RFC0791] or IPv6 183 [RFC2460] address of an Egress Tunnel Router (ETR). An RLOC is 184 the output of an EID-to-RLOC mapping lookup. An EID maps to one 185 or more RLOCs. Typically, RLOCs are numbered from topologically 186 aggregatable blocks that are assigned to a site at each point to 187 which it attaches to the global Internet; where the topology is 188 defined by the connectivity of provider networks, RLOCs can be 189 thought of as PA addresses. Multiple RLOCs can be assigned to the 190 same ETR device or to multiple ETR devices at a site. 192 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for 193 IPv6) value used in the source and destination address fields of 194 the first (most inner) LISP header of a packet. The host obtains 195 a destination EID the same way it obtains a destination address 196 today, for example, through a Domain Name System (DNS) [RFC1034] 197 lookup or Session Initiation Protocol (SIP) [RFC3261] exchange. 198 The source EID is obtained via existing mechanisms used to set a 199 host's "local" IP address. An EID used on the public Internet 200 must have the same properties as any other IP address used in that 201 manner; this means, among other things, that it must be globally 202 unique. An EID is allocated to a host from an EID-Prefix block 203 associated with the site where the host is located. An EID can be 204 used by a host to refer to other hosts. EIDs MUST NOT be used as 205 LISP RLOCs. Note that EID blocks MAY be assigned in a 206 hierarchical manner, independent of the network topology, to 207 facilitate scaling of the mapping database. In addition, an EID 208 block assigned to a site may have site-local structure 209 (subnetting) for routing within the site; this structure is not 210 visible to the global routing system. In theory, the bit string 211 that represents an EID for one device can represent an RLOC for a 212 different device. As the architecture is realized, if a given bit 213 string is both an RLOC and an EID, it must refer to the same 214 entity in both cases. When used in discussions with other 215 Locator/ID separation proposals, a LISP EID will be called an 216 "LEID". Throughout this document, any references to "EID" refer 217 to an LEID. 219 EID-Prefix: An EID-Prefix is a power-of-two block of EIDs that are 220 allocated to a site by an address allocation authority. EID- 221 Prefixes are associated with a set of RLOC addresses that make up 222 a "database mapping". EID-Prefix allocations can be broken up 223 into smaller blocks when an RLOC set is to be associated with the 224 larger EID-Prefix block. A globally routed address block (whether 225 PI or PA) is not inherently an EID-Prefix. A globally routed 226 address block MAY be used by its assignee as an EID block. The 227 converse is not supported. That is, a site that receives an 228 explicitly allocated EID-Prefix may not use that EID-Prefix as a 229 globally routed prefix. This would require coordination and 230 cooperation with the entities managing the mapping infrastructure. 231 Once this has been done, that block could be removed from the 232 globally routed IP system, if other suitable transition and access 233 mechanisms are in place. Discussion of such transition and access 234 mechanisms can be found in [RFC6832] and [RFC7215]. 236 End-system: An end-system is an IPv4 or IPv6 device that originates 237 packets with a single IPv4 or IPv6 header. The end-system 238 supplies an EID value for the destination address field of the IP 239 header when communicating globally (i.e., outside of its routing 240 domain). An end-system can be a host computer, a switch or router 241 device, or any network appliance. 243 Ingress Tunnel Router (ITR): An ITR is a router that resides in a 244 LISP site. Packets sent by sources inside of the LISP site to 245 destinations outside of the site are candidates for encapsulation 246 by the ITR. The ITR treats the IP destination address as an EID 247 and performs an EID-to-RLOC mapping lookup. The router then 248 prepends an "outer" IP header with one of its globally routable 249 RLOCs in the source address field and the result of the mapping 250 lookup in the destination address field. Note that this 251 destination RLOC MAY be an intermediate, proxy device that has 252 better knowledge of the EID-to-RLOC mapping closer to the 253 destination EID. In general, an ITR receives IP packets from site 254 end-systems on one side and sends LISP-encapsulated IP packets 255 toward the Internet on the other side. 257 Specifically, when a service provider prepends a LISP header for 258 Traffic Engineering purposes, the router that does this is also 259 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 260 on the outer destination address (the originating ITR's supplied 261 RLOC) or the inner destination address (the originating host's 262 supplied EID). 264 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider 265 network that prepends an additional LISP header for Traffic 266 Engineering purposes. 268 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP 269 packet where the destination address in the "outer" IP header is 270 one of its own RLOCs. The router strips the "outer" header and 271 forwards the packet based on the next IP header found. In 272 general, an ETR receives LISP-encapsulated IP packets from the 273 Internet on one side and sends decapsulated IP packets to site 274 end-systems on the other side. ETR functionality does not have to 275 be limited to a router device. A server host can be the endpoint 276 of a LISP tunnel as well. 278 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider 279 network that strips an outer LISP header for Traffic Engineering 280 purposes. 282 xTR: An xTR is a reference to an ITR or ETR when direction of data 283 flow is not part of the context description. "xTR" refers to the 284 router that is the tunnel endpoint and is used synonymously with 285 the term "Tunnel Router". For example, "An xTR can be located at 286 the Customer Edge (CE) router" indicates both ITR and ETR 287 functionality at the CE router. 289 LISP Router: A LISP router is a router that performs the functions 290 of any or all of the following: ITR, ETR, Proxy-ITR (PITR), or 291 Proxy-ETR (PETR). 293 EID-to-RLOC Map-Cache: The EID-to-RLOC map-cache is a short-lived, 294 on-demand table in an ITR that stores, tracks, and is responsible 295 for timing out and otherwise validating EID-to-RLOC mappings. 296 This cache is distinct from the full "database" of EID-to-RLOC 297 mappings; it is dynamic, local to the ITR(s), and relatively 298 small, while the database is distributed, relatively static, and 299 much more global in scope. 301 EID-to-RLOC Database: The EID-to-RLOC Database is a global 302 distributed database that contains all known EID-Prefix-to-RLOC 303 mappings. Each potential ETR typically contains a small piece of 304 the database: the EID-to-RLOC mappings for the EID-Prefixes 305 "behind" the router. These map to one of the router's own 306 globally visible IP addresses. The same database mapping entries 307 MUST be configured on all ETRs for a given site. In a steady 308 state, the EID-Prefixes for the site and the Locator-Set for each 309 EID-Prefix MUST be the same on all ETRs. Procedures to enforce 310 and/or verify this are outside the scope of this document. Note 311 that there MAY be transient conditions when the EID-Prefix for the 312 site and Locator-Set for each EID-Prefix may not be the same on 313 all ETRs. This has no negative implications, since a partial set 314 of Locators can be used. 316 Recursive Tunneling: Recursive Tunneling occurs when a packet has 317 more than one LISP IP header. Additional layers of tunneling MAY 318 be employed to implement Traffic Engineering or other re-routing 319 as needed. When this is done, an additional "outer" LISP header 320 is added, and the original RLOCs are preserved in the "inner" 321 header. Any references to tunnels in this specification refer to 322 dynamic encapsulating tunnels; they are never statically 323 configured. 325 Re-encapsulating Tunnels: Re-encapsulating Tunneling occurs when an 326 ETR removes a LISP header, then acts as an ITR to prepend another 327 LISP header. Doing this allows a packet to be re-routed by the 328 re-encapsulating router without adding the overhead of additional 329 tunnel headers. Any references to tunnels in this specification 330 refer to dynamic encapsulating tunnels; they are never statically 331 configured. When using multiple mapping database systems, care 332 must be taken to not create re-encapsulation loops through 333 misconfiguration. 335 LISP Header: LISP header is a term used in this document to refer 336 to the outer IPv4 or IPv6 header, a UDP header, and a LISP- 337 specific 8-octet header that follow the UDP header and that an ITR 338 prepends or an ETR strips. 340 Address Family Identifier (AFI): AFI is a term used to describe an 341 address encoding in a packet. An address family currently 342 pertains to an IPv4 or IPv6 address. See [AFN] and [RFC3232] for 343 details. An AFI value of 0 used in this specification indicates 344 an unspecified encoded address where the length of the address is 345 0 octets following the 16-bit AFI value of 0. 347 Negative Mapping Entry: A negative mapping entry, also known as a 348 negative cache entry, is an EID-to-RLOC entry where an EID-Prefix 349 is advertised or stored with no RLOCs. That is, the Locator-Set 350 for the EID-to-RLOC entry is empty or has an encoded Locator count 351 of 0. This type of entry could be used to describe a prefix from 352 a non-LISP site, which is explicitly not in the mapping database. 353 There are a set of well-defined actions that are encoded in a 354 Negative Map-Reply. 356 Data-Probe: A Data-Probe is a LISP-encapsulated data packet where 357 the inner-header destination address equals the outer-header 358 destination address used to trigger a Map-Reply by a decapsulating 359 ETR. In addition, the original packet is decapsulated and 360 delivered to the destination host if the destination EID is in the 361 EID-Prefix range configured on the ETR. Otherwise, the packet is 362 discarded. A Data-Probe is used in some of the mapping database 363 designs to "probe" or request a Map-Reply from an ETR; in other 364 cases, Map-Requests are used. See each mapping database design 365 for details. When using Data-Probes, by sending Map-Requests on 366 the underlying routing system, EID-Prefixes must be advertised. 367 However, this is discouraged if the core is to scale by having 368 less EID-Prefixes stored in the core router's routing tables. 370 Proxy-ITR (PITR): A PITR is defined and described in [RFC6832]. A 371 PITR acts like an ITR but does so on behalf of non-LISP sites that 372 send packets to destinations at LISP sites. 374 Proxy-ETR (PETR): A PETR is defined and described in [RFC6832]. A 375 PETR acts like an ETR but does so on behalf of LISP sites that 376 send packets to destinations at non-LISP sites. 378 Route-returnability: Route-returnability is an assumption that the 379 underlying routing system will deliver packets to the destination. 380 When combined with a nonce that is provided by a sender and 381 returned by a receiver, this limits off-path data insertion. A 382 route-returnability check is verified when a message is sent with 383 a nonce, another message is returned with the same nonce, and the 384 destination of the original message appears as the source of the 385 returned message. 387 LISP site: LISP site is a set of routers in an edge network that are 388 under a single technical administration. LISP routers that reside 389 in the edge network are the demarcation points to separate the 390 edge network from the core network. 392 Client-side: Client-side is a term used in this document to indicate 393 a connection initiation attempt by an EID. The ITR(s) at the LISP 394 site are the first to get involved in obtaining database Map-Cache 395 entries by sending Map-Request messages. 397 Server-side: Server-side is a term used in this document to indicate 398 that a connection initiation attempt is being accepted for a 399 destination EID. The ETR(s) at the destination LISP site are the 400 first to send Map-Replies to the source site initiating the 401 connection. The ETR(s) at this destination site can obtain 402 mappings by gleaning information from Map-Requests, Data-Probes, 403 or encapsulated packets. 405 Locator-Status-Bits (LSBs): Locator-Status-Bits are present in the 406 LISP header. They are used by ITRs to inform ETRs about the up/ 407 down status of all ETRs at the local site. These bits are used as 408 a hint to convey up/down router status and not path reachability 409 status. The LSBs can be verified by use of one of the Locator 410 reachability algorithms described in Section 10. 412 Anycast Address: Anycast Address is a term used in this document to 413 refer to the same IPv4 or IPv6 address configured and used on 414 multiple systems at the same time. An EID or RLOC can be an 415 anycast address in each of their own address spaces. 417 4. Basic Overview 419 One key concept of LISP is that end-systems operate the same way they 420 do today. The IP addresses that hosts use for tracking sockets and 421 connections, and for sending and receiving packets, do not change. 422 In LISP terminology, these IP addresses are called Endpoint 423 Identifiers (EIDs). 425 Routers continue to forward packets based on IP destination 426 addresses. When a packet is LISP encapsulated, these addresses are 427 referred to as Routing Locators (RLOCs). Most routers along a path 428 between two hosts will not change; they continue to perform routing/ 429 forwarding lookups on the destination addresses. For routers between 430 the source host and the ITR as well as routers from the ETR to the 431 destination host, the destination address is an EID. For the routers 432 between the ITR and the ETR, the destination address is an RLOC. 434 Another key LISP concept is the "Tunnel Router". A Tunnel Router 435 prepends LISP headers on host-originated packets and strips them 436 prior to final delivery to their destination. The IP addresses in 437 this "outer header" are RLOCs. During end-to-end packet exchange 438 between two Internet hosts, an ITR prepends a new LISP header to each 439 packet, and an ETR strips the new header. The ITR performs EID-to- 440 RLOC lookups to determine the routing path to the ETR, which has the 441 RLOC as one of its IP addresses. 443 Some basic rules governing LISP are: 445 o End-systems only send to addresses that are EIDs. They don't know 446 that addresses are EIDs versus RLOCs but assume that packets get 447 to their intended destinations. In a system where LISP is 448 deployed, LISP routers intercept EID-addressed packets and assist 449 in delivering them across the network core where EIDs cannot be 450 routed. The procedure a host uses to send IP packets does not 451 change. 453 o EIDs are typically IP addresses assigned to hosts. 455 o Other types of EID are supported by LISP, see [RFC8060] for 456 further information. 458 o LISP routers mostly deal with Routing Locator addresses. See 459 details in Section 4.1 to clarify what is meant by "mostly". 461 o RLOCs are always IP addresses assigned to routers, preferably 462 topologically oriented addresses from provider CIDR (Classless 463 Inter-Domain Routing) blocks. 465 o When a router originates packets, it may use as a source address 466 either an EID or RLOC. When acting as a host (e.g., when 467 terminating a transport session such as Secure SHell (SSH), 468 TELNET, or the Simple Network Management Protocol (SNMP)), it may 469 use an EID that is explicitly assigned for that purpose. An EID 470 that identifies the router as a host MUST NOT be used as an RLOC; 471 an EID is only routable within the scope of a site. A typical BGP 472 configuration might demonstrate this "hybrid" EID/RLOC usage where 473 a router could use its "host-like" EID to terminate iBGP sessions 474 to other routers in a site while at the same time using RLOCs to 475 terminate eBGP sessions to routers outside the site. 477 o Packets with EIDs in them are not expected to be delivered end-to- 478 end in the absence of an EID-to-RLOC mapping operation. They are 479 expected to be used locally for intra-site communication or to be 480 encapsulated for inter-site communication. 482 o EID-Prefixes are likely to be hierarchically assigned in a manner 483 that is optimized for administrative convenience and to facilitate 484 scaling of the EID-to-RLOC mapping database. The hierarchy is 485 based on an address allocation hierarchy that is independent of 486 the network topology. 488 o EIDs may also be structured (subnetted) in a manner suitable for 489 local routing within an Autonomous System (AS). 491 An additional LISP header MAY be prepended to packets by a TE-ITR 492 when re-routing of the path for a packet is desired. A potential 493 use-case for this would be an ISP router that needs to perform 494 Traffic Engineering for packets flowing through its network. In such 495 a situation, termed "Recursive Tunneling", an ISP transit acts as an 496 additional ITR, and the RLOC it uses for the new prepended header 497 would be either a TE-ETR within the ISP (along an intra-ISP traffic 498 engineered path) or a TE-ETR within another ISP (an inter-ISP traffic 499 engineered path, where an agreement to build such a path exists). 501 In order to avoid excessive packet overhead as well as possible 502 encapsulation loops, this document mandates that a maximum of two 503 LISP headers can be prepended to a packet. For initial LISP 504 deployments, it is assumed that two headers is sufficient, where the 505 first prepended header is used at a site for Location/Identity 506 separation and the second prepended header is used inside a service 507 provider for Traffic Engineering purposes. 509 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 510 For example, the ITR for a particular end-to-end packet exchange 511 might be the first-hop or default router within a site for the source 512 host. Similarly, the ETR might be the last-hop router directly 513 connected to the destination host. Another example, perhaps for a 514 VPN service outsourced to an ISP by a site, the ITR could be the 515 site's border router at the service provider attachment point. 516 Mixing and matching of site-operated, ISP-operated, and other Tunnel 517 Routers is allowed for maximum flexibility. 519 4.1. Packet Flow Sequence 521 This section provides an example of the unicast packet flow, 522 including also control-plane information as specified in 523 [I-D.ietf-lisp-rfc6833bis]. The example also assumes the following 524 conditions: 526 o Source host "host1.abc.example.com" is sending a packet to 527 "host2.xyz.example.com", exactly what host1 would do if the site 528 was not using LISP. 530 o Each site is multihomed, so each Tunnel Router has an address 531 (RLOC) assigned from the service provider address block for each 532 provider to which that particular Tunnel Router is attached. 534 o The ITR(s) and ETR(s) are directly connected to the source and 535 destination, respectively, but the source and destination can be 536 located anywhere in the LISP site. 538 o Map-Requests are sent to the mapping database system by using the 539 LISP control-plane protocol documented in 540 [I-D.ietf-lisp-rfc6833bis]. A Map-Request is sent for an external 541 destination when the destination is not found in the forwarding 542 table or matches a default route. 544 o Map-Replies are sent on the underlying routing system topology 545 using the [I-D.ietf-lisp-rfc6833bis] control-plane protocol. 547 Client host1.abc.example.com wants to communicate with server 548 host2.xyz.example.com: 550 1. host1.abc.example.com wants to open a TCP connection to 551 host2.xyz.example.com. It does a DNS lookup on 552 host2.xyz.example.com. An A/AAAA record is returned. This 553 address is the destination EID. The locally assigned address of 554 host1.abc.example.com is used as the source EID. An IPv4 or IPv6 555 packet is built and forwarded through the LISP site as a normal 556 IP packet until it reaches a LISP ITR. 558 2. The LISP ITR must be able to map the destination EID to an RLOC 559 of one of the ETRs at the destination site. The specific method 560 used to do this is not described in this example. See 561 [I-D.ietf-lisp-rfc6833bis] for further information. 563 3. The ITR sends a LISP Map-Request as specified in 564 [I-D.ietf-lisp-rfc6833bis]. Map-Requests SHOULD be rate-limited. 566 4. The mapping system helps forwarding the Map-Request to the 567 corresponding ETR. When the Map-Request arrives at one of the 568 ETRs at the destination site, it will process the packet as a 569 control message. 571 5. The ETR looks at the destination EID of the Map-Request and 572 matches it against the prefixes in the ETR's configured EID-to- 573 RLOC mapping database. This is the list of EID-Prefixes the ETR 574 is supporting for the site it resides in. If there is no match, 575 the Map-Request is dropped. Otherwise, a LISP Map-Reply is 576 returned to the ITR. 578 6. The ITR receives the Map-Reply message, parses the message (to 579 check for format validity), and stores the mapping information 580 from the packet. This information is stored in the ITR's EID-to- 581 RLOC map-cache. Note that the map-cache is an on-demand cache. 582 An ITR will manage its map-cache in such a way that optimizes for 583 its resource constraints. 585 7. Subsequent packets from host1.abc.example.com to 586 host2.xyz.example.com will have a LISP header prepended by the 587 ITR using the appropriate RLOC as the LISP header destination 588 address learned from the ETR. Note that the packet MAY be sent 589 to a different ETR than the one that returned the Map-Reply due 590 to the source site's hashing policy or the destination site's 591 Locator-Set policy. 593 8. The ETR receives these packets directly (since the destination 594 address is one of its assigned IP addresses), checks the validity 595 of the addresses, strips the LISP header, and forwards packets to 596 the attached destination host. 598 9. In order to defer the need for a mapping lookup in the reverse 599 direction, an ETR can OPTIONALLY create a cache entry that maps 600 the source EID (inner-header source IP address) to the source 601 RLOC (outer-header source IP address) in a received LISP packet. 602 Such a cache entry is termed a "gleaned" mapping and only 603 contains a single RLOC for the EID in question. More complete 604 information about additional RLOCs SHOULD be verified by sending 605 a LISP Map-Request for that EID. Both the ITR and the ETR may 606 also influence the decision the other makes in selecting an RLOC. 608 5. LISP Encapsulation Details 610 Since additional tunnel headers are prepended, the packet becomes 611 larger and can exceed the MTU of any link traversed from the ITR to 612 the ETR. It is RECOMMENDED in IPv4 that packets do not get 613 fragmented as they are encapsulated by the ITR. Instead, the packet 614 is dropped and an ICMP Unreachable/Fragmentation-Needed message is 615 returned to the source. 617 This specification RECOMMENDS that implementations provide support 618 for one of the proposed fragmentation and reassembly schemes. Two 619 existing schemes are detailed in Section 7. 621 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP 622 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner 623 header is in IPv4 packet format and the outer header is in IPv6 624 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header 625 is in IPv6 packet format and the outer header is in IPv4 packet 626 format). The next sub-sections illustrate packet formats for the 627 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6), but all 4 628 combinations MUST be supported. Additional types of EIDs are defined 629 in [RFC8060]. 631 5.1. LISP IPv4-in-IPv4 Header Format 633 0 1 2 3 634 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 1 635 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 636 / |Version| IHL |Type of Service| Total Length | 637 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 638 | | Identification |Flags| Fragment Offset | 639 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 640 OH | Time to Live | Protocol = 17 | Header Checksum | 641 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 642 | | Source Routing Locator | 643 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 644 \ | Destination Routing Locator | 645 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 646 / | Source Port = xxxx | Dest Port = 4341 | 647 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 648 \ | UDP Length | UDP Checksum | 649 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 650 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 651 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 652 S / | Instance ID/Locator-Status-Bits | 653 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 654 / |Version| IHL |Type of Service| Total Length | 655 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 656 | | Identification |Flags| Fragment Offset | 657 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 658 IH | Time to Live | Protocol | Header Checksum | 659 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 660 | | Source EID | 661 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 662 \ | Destination EID | 663 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 665 IHL = IP-Header-Length 667 5.2. LISP IPv6-in-IPv6 Header Format 669 0 1 2 3 670 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 1 671 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 672 / |Version| Traffic Class | Flow Label | 673 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 674 | | Payload Length | Next Header=17| Hop Limit | 675 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 676 | | 677 O + + 678 u | | 679 t + Source Routing Locator + 680 e | | 681 r + + 682 | | 683 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 684 d | | 685 r + + 686 | | 687 ^ + Destination Routing Locator + 688 | | | 689 \ + + 690 \ | | 691 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 692 / | Source Port = xxxx | Dest Port = 4341 | 693 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 694 \ | UDP Length | UDP Checksum | 695 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 696 L |N|L|E|V|I|R|K|K| Nonce/Map-Version | 697 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 698 S / | Instance ID/Locator-Status-Bits | 699 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 700 / |Version| Traffic Class | Flow Label | 701 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 702 / | Payload Length | Next Header | Hop Limit | 703 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 704 | | 705 I + + 706 n | | 707 n + Source EID + 708 e | | 709 r + + 710 | | 711 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 712 d | | 713 r + + 714 | | 716 ^ + Destination EID + 717 \ | | 718 \ + + 719 \ | | 720 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 722 5.3. Tunnel Header Field Descriptions 724 Inner Header (IH): The inner header is the header on the datagram 725 received from the originating host. The source and destination IP 726 addresses are EIDs [RFC0791] [RFC2460]. 728 Outer Header: (OH) The outer header is a new header prepended by an 729 ITR. The address fields contain RLOCs obtained from the ingress 730 router's EID-to-RLOC Cache. The IP protocol number is "UDP (17)" 731 from [RFC0768]. The setting of the Don't Fragment (DF) bit 732 'Flags' field is according to rules listed in Sections 7.1 and 733 7.2. 735 UDP Header: The UDP header contains an ITR selected source port when 736 encapsulating a packet. See Section 12 for details on the hash 737 algorithm used to select a source port based on the 5-tuple of the 738 inner header. The destination port MUST be set to the well-known 739 IANA-assigned port value 4341. 741 UDP Checksum: The 'UDP Checksum' field SHOULD be transmitted as zero 742 by an ITR for either IPv4 [RFC0768] or IPv6 encapsulation 743 [RFC6935] [RFC6936]. When a packet with a zero UDP checksum is 744 received by an ETR, the ETR MUST accept the packet for 745 decapsulation. When an ITR transmits a non-zero value for the UDP 746 checksum, it MUST send a correctly computed value in this field. 747 When an ETR receives a packet with a non-zero UDP checksum, it MAY 748 choose to verify the checksum value. If it chooses to perform 749 such verification, and the verification fails, the packet MUST be 750 silently dropped. If the ETR chooses not to perform the 751 verification, or performs the verification successfully, the 752 packet MUST be accepted for decapsulation. The handling of UDP 753 checksums for all tunneling protocols, including LISP, is under 754 active discussion within the IETF. When that discussion 755 concludes, any necessary changes will be made to align LISP with 756 the outcome of the broader discussion. 758 UDP Length: The 'UDP Length' field is set for an IPv4-encapsulated 759 packet to be the sum of the inner-header IPv4 Total Length plus 760 the UDP and LISP header lengths. For an IPv6-encapsulated packet, 761 the 'UDP Length' field is the sum of the inner-header IPv6 Payload 762 Length, the size of the IPv6 header (40 octets), and the size of 763 the UDP and LISP headers. 765 N: The N-bit is the nonce-present bit. When this bit is set to 1, 766 the low-order 24 bits of the first 32 bits of the LISP header 767 contain a Nonce. See Section 10.1 for details. Both N- and 768 V-bits MUST NOT be set in the same packet. If they are, a 769 decapsulating ETR MUST treat the 'Nonce/Map-Version' field as 770 having a Nonce value present. 772 L: The L-bit is the 'Locator-Status-Bits' field enabled bit. When 773 this bit is set to 1, the Locator-Status-Bits in the second 774 32 bits of the LISP header are in use. 776 x 1 x x 0 x x x 777 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 778 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 779 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 780 | Locator-Status-Bits | 781 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 783 E: The E-bit is the echo-nonce-request bit. This bit MUST be ignored 784 and has no meaning when the N-bit is set to 0. When the N-bit is 785 set to 1 and this bit is set to 1, an ITR is requesting that the 786 nonce value in the 'Nonce' field be echoed back in LISP- 787 encapsulated packets when the ITR is also an ETR. See 788 Section 10.1 for details. 790 V: The V-bit is the Map-Version present bit. When this bit is set to 791 1, the N-bit MUST be 0. Refer to Section 13.3 for more details. 792 This bit indicates that the LISP header is encoded in this 793 case as: 795 0 x 0 1 x x x x 796 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 797 |N|L|E|V|I|R|K|K| Source Map-Version | Dest Map-Version | 798 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 799 | Instance ID/Locator-Status-Bits | 800 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 802 I: The I-bit is the Instance ID bit. See Section 8 for more details. 803 When this bit is set to 1, the 'Locator-Status-Bits' field is 804 reduced to 8 bits and the high-order 24 bits are used as an 805 Instance ID. If the L-bit is set to 0, then the low-order 8 bits 806 are transmitted as zero and ignored on receipt. The format of the 807 LISP header would look like this: 809 x x x x 1 x x x 810 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 811 |N|L|E|V|I|R|K|K| Nonce/Map-Version | 812 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 813 | Instance ID | LSBs | 814 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 816 R: The R-bit is a Reserved bit for future use. It MUST be set to 0 817 on transmit and MUST be ignored on receipt. 819 KK: The KK-bits are a 2-bit field used when encapsualted packets are 820 encrypted. The field is set to 00 when the packet is not 821 encrypted. See [RFC8061] for further information. 823 LISP Nonce: The LISP 'Nonce' field is a 24-bit value that is 824 randomly generated by an ITR when the N-bit is set to 1. Nonce 825 generation algorithms are an implementation matter but are 826 required to generate different nonces when sending to different 827 destinations. However, the same nonce can be used for a period of 828 time to the same destination. The nonce is also used when the 829 E-bit is set to request the nonce value to be echoed by the other 830 side when packets are returned. When the E-bit is clear but the 831 N-bit is set, a remote ITR is either echoing a previously 832 requested echo-nonce or providing a random nonce. See 833 Section 10.1 for more details. 835 LISP Locator-Status-Bits (LSBs): When the L-bit is also set, the 836 'Locator-Status-Bits' field in the LISP header is set by an ITR to 837 indicate to an ETR the up/down status of the Locators in the 838 source site. Each RLOC in a Map-Reply is assigned an ordinal 839 value from 0 to n-1 (when there are n RLOCs in a mapping entry). 840 The Locator-Status-Bits are numbered from 0 to n-1 from the least 841 significant bit of the field. The field is 32 bits when the I-bit 842 is set to 0 and is 8 bits when the I-bit is set to 1. When a 843 Locator-Status-Bit is set to 1, the ITR is indicating to the ETR 844 that the RLOC associated with the bit ordinal has up status. See 845 Section 10 for details on how an ITR can determine the status of 846 the ETRs at the same site. When a site has multiple EID-Prefixes 847 that result in multiple mappings (where each could have a 848 different Locator-Set), the Locator-Status-Bits setting in an 849 encapsulated packet MUST reflect the mapping for the EID-Prefix 850 that the inner-header source EID address matches. If the LSB for 851 an anycast Locator is set to 1, then there is at least one RLOC 852 with that address, and the ETR is considered 'up'. 854 When doing ITR/PITR encapsulation: 856 o The outer-header 'Time to Live' field (or 'Hop Limit' field, in 857 the case of IPv6) SHOULD be copied from the inner-header 'Time to 858 Live' field. 860 o The outer-header 'Type of Service' field (or the 'Traffic Class' 861 field, in the case of IPv6) SHOULD be copied from the inner-header 862 'Type of Service' field (with one exception; see below). 864 When doing ETR/PETR decapsulation: 866 o The inner-header 'Time to Live' field (or 'Hop Limit' field, in 867 the case of IPv6) SHOULD be copied from the outer-header 'Time to 868 Live' field, when the Time to Live value of the outer header is 869 less than the Time to Live value of the inner header. Failing to 870 perform this check can cause the Time to Live of the inner header 871 to increment across encapsulation/decapsulation cycles. This 872 check is also performed when doing initial encapsulation, when a 873 packet comes to an ITR or PITR destined for a LISP site. 875 o The inner-header 'Type of Service' field (or the 'Traffic Class' 876 field, in the case of IPv6) SHOULD be copied from the outer-header 877 'Type of Service' field (with one exception; see below). 879 Note that if an ETR/PETR is also an ITR/PITR and chooses to re- 880 encapsulate after decapsulating, the net effect of this is that the 881 new outer header will carry the same Time to Live as the old outer 882 header minus 1. 884 Copying the Time to Live (TTL) serves two purposes: first, it 885 preserves the distance the host intended the packet to travel; 886 second, and more importantly, it provides for suppression of looping 887 packets in the event there is a loop of concatenated tunnels due to 888 misconfiguration. See Section 18.3 for TTL exception handling for 889 traceroute packets. 891 The Explicit Congestion Notification ('ECN') field occupies bits 6 892 and 7 of both the IPv4 'Type of Service' field and the IPv6 'Traffic 893 Class' field [RFC3168]. The 'ECN' field requires special treatment 894 in order to avoid discarding indications of congestion [RFC3168]. 895 ITR encapsulation MUST copy the 2-bit 'ECN' field from the inner 896 header to the outer header. Re-encapsulation MUST copy the 2-bit 897 'ECN' field from the stripped outer header to the new outer header. 898 If the 'ECN' field contains a congestion indication codepoint (the 899 value is '11', the Congestion Experienced (CE) codepoint), then ETR 900 decapsulation MUST copy the 2-bit 'ECN' field from the stripped outer 901 header to the surviving inner header that is used to forward the 902 packet beyond the ETR. These requirements preserve CE indications 903 when a packet that uses ECN traverses a LISP tunnel and becomes 904 marked with a CE indication due to congestion between the tunnel 905 endpoints. 907 6. LISP EID-to-RLOC Map-Cache 909 ITRs and PITRs maintain an on-demand cache, referred as LISP EID-to- 910 RLOC Map-Cache, that contains mappings from EID-prefixes to locator 911 sets. The cache is used to encapsulate packets from the EID space to 912 the corresponding RLOC network attachment point. 914 When an ITR/PITR receives a packet from inside of the LISP site to 915 destinations outside of the site a longest-prefix match lookup of the 916 EID is done to the map-cache. 918 When the lookup succeeds, the locator-set retrieved from the map- 919 cache is used to send the packet to the EID's topological location. 921 If the lookup fails, the ITR/PITR needs to retrieve the mapping using 922 the LISP control-plane protocol [I-D.ietf-lisp-rfc6833bis]. The 923 mapping is then stored in the local map-cache to forward subsequent 924 packets addressed to the same EID-prefix. 926 The map-cache is a local cache of mappings, entries are expired based 927 on the associated Time to live. In addition, entries can be updated 928 with more current information, see Section 13 for further information 929 on this. Finally, the map-cache also contains reachability 930 information about EIDs and RLOCs, and uses LISP reachability 931 information mechanisms to determine the reachability of RLOCs, see 932 Section 10 for the specific mechanisms. 934 7. Dealing with Large Encapsulated Packets 936 This section proposes two mechanisms to deal with packets that exceed 937 the path MTU between the ITR and ETR. 939 It is left to the implementor to decide if the stateless or stateful 940 mechanism should be implemented. Both or neither can be used, since 941 it is a local decision in the ITR regarding how to deal with MTU 942 issues, and sites can interoperate with differing mechanisms. 944 Both stateless and stateful mechanisms also apply to Re-encapsulating 945 and Recursive Tunneling, so any actions below referring to an ITR 946 also apply to a TE-ITR. 948 7.1. A Stateless Solution to MTU Handling 950 An ITR stateless solution to handle MTU issues is described as 951 follows: 953 1. Define H to be the size, in octets, of the outer header an ITR 954 prepends to a packet. This includes the UDP and LISP header 955 lengths. 957 2. Define L to be the size, in octets, of the maximum-sized packet 958 an ITR can send to an ETR without the need for the ITR or any 959 intermediate routers to fragment the packet. 961 3. Define an architectural constant S for the maximum size of a 962 packet, in octets, an ITR must receive from the source so the 963 effective MTU can be met. That is, L = S + H. 965 When an ITR receives a packet from a site-facing interface and adds H 966 octets worth of encapsulation to yield a packet size greater than L 967 octets (meaning the received packet size was greater than S octets 968 from the source), it resolves the MTU issue by first splitting the 969 original packet into 2 equal-sized fragments. A LISP header is then 970 prepended to each fragment. The size of the encapsulated fragments 971 is then (S/2 + H), which is less than the ITR's estimate of the path 972 MTU between the ITR and its correspondent ETR. 974 When an ETR receives encapsulated fragments, it treats them as two 975 individually encapsulated packets. It strips the LISP headers and 976 then forwards each fragment to the destination host of the 977 destination site. The two fragments are reassembled at the 978 destination host into the single IP datagram that was originated by 979 the source host. Note that reassembly can happen at the ETR if the 980 encapsulated packet was fragmented at or after the ITR. 982 This behavior is performed by the ITR when the source host originates 983 a packet with the 'DF' field of the IP header set to 0. When the 984 'DF' field of the IP header is set to 1, or the packet is an IPv6 985 packet originated by the source host, the ITR will drop the packet 986 when the size is greater than L and send an ICMP Unreachable/ 987 Fragmentation-Needed message to the source with a value of S, where S 988 is (L - H). 990 When the outer-header encapsulation uses an IPv4 header, an 991 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly 992 can be avoided. An implementation MAY set the DF bit in such headers 993 to 0 if it has good reason to believe there are unresolvable path MTU 994 issues between the sending ITR and the receiving ETR. 996 This specification RECOMMENDS that L be defined as 1500. 998 7.2. A Stateful Solution to MTU Handling 1000 An ITR stateful solution to handle MTU issues is described as follows 1001 and was first introduced in [OPENLISP]: 1003 1. The ITR will keep state of the effective MTU for each Locator per 1004 Map-Cache entry. The effective MTU is what the core network can 1005 deliver along the path between the ITR and ETR. 1007 2. When an IPv6-encapsulated packet, or an IPv4-encapsulated packet 1008 with the DF bit set to 1, exceeds what the core network can 1009 deliver, one of the intermediate routers on the path will send an 1010 ICMP Unreachable/Fragmentation-Needed message to the ITR. The 1011 ITR will parse the ICMP message to determine which Locator is 1012 affected by the effective MTU change and then record the new 1013 effective MTU value in the Map-Cache entry. 1015 3. When a packet is received by the ITR from a source inside of the 1016 site and the size of the packet is greater than the effective MTU 1017 stored with the Map-Cache entry associated with the destination 1018 EID the packet is for, the ITR will send an ICMP Unreachable/ 1019 Fragmentation-Needed message back to the source. The packet size 1020 advertised by the ITR in the ICMP Unreachable/Fragmentation- 1021 Needed message is the effective MTU minus the LISP encapsulation 1022 length. 1024 Even though this mechanism is stateful, it has advantages over the 1025 stateless IP fragmentation mechanism, by not involving the 1026 destination host with reassembly of ITR fragmented packets. 1028 8. Using Virtualization and Segmentation with LISP 1030 When multiple organizations inside of a LISP site are using private 1031 addresses [RFC1918] as EID-Prefixes, their address spaces MUST remain 1032 segregated due to possible address duplication. An Instance ID in 1033 the address encoding can aid in making the entire AFI-based address 1034 unique. See IANA Considerations of [I-D.ietf-lisp-rfc6833bis] for 1035 details on possible address encodings. 1037 An Instance ID can be carried in a LISP-encapsulated packet. An ITR 1038 that prepends a LISP header will copy a 24-bit value used by the LISP 1039 router to uniquely identify the address space. The value is copied 1040 to the 'Instance ID' field of the LISP header, and the I-bit is set 1041 to 1. 1043 When an ETR decapsulates a packet, the Instance ID from the LISP 1044 header is used as a table identifier to locate the forwarding table 1045 to use for the inner destination EID lookup. 1047 For example, an 802.1Q VLAN tag or VPN identifier could be used as a 1048 24-bit Instance ID. 1050 The Instance ID that is stored in the mapping database when LISP-DDT 1051 [I-D.ietf-lisp-ddt] is used is 32 bits in length. That means the 1052 control-plane can store more instances than a given data-plane can 1053 use. Multiple data-planes can use the same 32-bit space as long as 1054 the low-order 24 bits don't overlap among xTRs. 1056 9. Routing Locator Selection 1058 Both the client-side and server-side may need control over the 1059 selection of RLOCs for conversations between them. This control is 1060 achieved by manipulating the 'Priority' and 'Weight' fields in EID- 1061 to-RLOC Map-Reply messages. Alternatively, RLOC information MAY be 1062 gleaned from received tunneled packets or EID-to-RLOC Map-Request 1063 messages. 1065 The following are different scenarios for choosing RLOCs and the 1066 controls that are available: 1068 o The server-side returns one RLOC. The client-side can only use 1069 one RLOC. The server-side has complete control of the selection. 1071 o The server-side returns a list of RLOCs where a subset of the list 1072 has the same best Priority. The client can only use the subset 1073 list according to the weighting assigned by the server-side. In 1074 this case, the server-side controls both the subset list and load- 1075 splitting across its members. The client-side can use RLOCs 1076 outside of the subset list if it determines that the subset list 1077 is unreachable (unless RLOCs are set to a Priority of 255). Some 1078 sharing of control exists: the server-side determines the 1079 destination RLOC list and load distribution while the client-side 1080 has the option of using alternatives to this list if RLOCs in the 1081 list are unreachable. 1083 o The server-side sets a Weight of 0 for the RLOC subset list. In 1084 this case, the client-side can choose how the traffic load is 1085 spread across the subset list. Control is shared by the server- 1086 side determining the list and the client determining load 1087 distribution. Again, the client can use alternative RLOCs if the 1088 server-provided list of RLOCs is unreachable. 1090 o Either side (more likely the server-side ETR) decides not to send 1091 a Map-Request. For example, if the server-side ETR does not send 1092 Map-Requests, it gleans RLOCs from the client-side ITR, giving the 1093 client-side ITR responsibility for bidirectional RLOC reachability 1094 and preferability. Server-side ETR gleaning of the client-side 1095 ITR RLOC is done by caching the inner-header source EID and the 1096 outer-header source RLOC of received packets. The client-side ITR 1097 controls how traffic is returned and can alternate using an outer- 1098 header source RLOC, which then can be added to the list the 1099 server-side ETR uses to return traffic. Since no Priority or 1100 Weights are provided using this method, the server-side ETR MUST 1101 assume that each client-side ITR RLOC uses the same best Priority 1102 with a Weight of zero. In addition, since EID-Prefix encoding 1103 cannot be conveyed in data packets, the EID-to-RLOC Cache on 1104 Tunnel Routers can grow to be very large. 1106 o A "gleaned" Map-Cache entry, one learned from the source RLOC of a 1107 received encapsulated packet, is only stored and used for a few 1108 seconds, pending verification. Verification is performed by 1109 sending a Map-Request to the source EID (the inner-header IP 1110 source address) of the received encapsulated packet. A reply to 1111 this "verifying Map-Request" is used to fully populate the Map- 1112 Cache entry for the "gleaned" EID and is stored and used for the 1113 time indicated from the 'TTL' field of a received Map-Reply. When 1114 a verified Map-Cache entry is stored, data gleaning no longer 1115 occurs for subsequent packets that have a source EID that matches 1116 the EID-Prefix of the verified entry. This "gleaning" mechanism 1117 is OPTIONAL. 1119 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be 1120 reachable when the R-bit for the Locator record is set to 1. When 1121 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the 1122 RLOC. Neither the information contained in a Map-Reply nor that 1123 stored in the mapping database system provides reachability 1124 information for RLOCs. Note that reachability is not part of the 1125 mapping system and is determined using one or more of the Routing 1126 Locator reachability algorithms described in the next section. 1128 10. Routing Locator Reachability 1130 Several mechanisms for determining RLOC reachability are currently 1131 defined: 1133 1. An ETR may examine the Locator-Status-Bits in the LISP header of 1134 an encapsulated data packet received from an ITR. If the ETR is 1135 also acting as an ITR and has traffic to return to the original 1136 ITR site, it can use this status information to help select an 1137 RLOC. 1139 2. An ITR may receive an ICMP Network Unreachable or Host 1140 Unreachable message for an RLOC it is using. This indicates that 1141 the RLOC is likely down. Note that trusting ICMP messages may 1142 not be desirable, but neither is ignoring them completely. 1143 Implementations are encouraged to follow current best practices 1144 in treating these conditions. 1146 3. An ITR that participates in the global routing system can 1147 determine that an RLOC is down if no BGP Routing Information Base 1148 (RIB) route exists that matches the RLOC IP address. 1150 4. An ITR may receive an ICMP Port Unreachable message from a 1151 destination host. This occurs if an ITR attempts to use 1152 interworking [RFC6832] and LISP-encapsulated data is sent to a 1153 non-LISP-capable site. 1155 5. An ITR may receive a Map-Reply from an ETR in response to a 1156 previously sent Map-Request. The RLOC source of the Map-Reply is 1157 likely up, since the ETR was able to send the Map-Reply to the 1158 ITR. 1160 6. When an ETR receives an encapsulated packet from an ITR, the 1161 source RLOC from the outer header of the packet is likely up. 1163 7. An ITR/ETR pair can use the Locator reachability algorithms 1164 described in this section, namely Echo-Noncing or RLOC-Probing. 1166 When determining Locator up/down reachability by examining the 1167 Locator-Status-Bits from the LISP-encapsulated data packet, an ETR 1168 will receive up-to-date status from an encapsulating ITR about 1169 reachability for all ETRs at the site. CE-based ITRs at the source 1170 site can determine reachability relative to each other using the site 1171 IGP as follows: 1173 o Under normal circumstances, each ITR will advertise a default 1174 route into the site IGP. 1176 o If an ITR fails or if the upstream link to its PE fails, its 1177 default route will either time out or be withdrawn. 1179 Each ITR can thus observe the presence or lack of a default route 1180 originated by the others to determine the Locator-Status-Bits it sets 1181 for them. 1183 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The 1184 Locator-Status-Bits in a LISP-encapsulated packet are numbered from 0 1185 to n-1 starting with the least significant bit. For example, if an 1186 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal 1187 value 2), then all ITRs at the site will clear the 3rd least 1188 significant bit (xxxx x0xx) of the 'Locator-Status-Bits' field for 1189 the packets they encapsulate. 1191 When an ETR decapsulates a packet, it will check for any change in 1192 the 'Locator-Status-Bits' field. When a bit goes from 1 to 0, the 1193 ETR, if acting also as an ITR, will refrain from encapsulating 1194 packets to an RLOC that is indicated as down. It will only resume 1195 using that RLOC if the corresponding Locator-Status-Bit returns to a 1196 value of 1. Locator-Status-Bits are associated with a Locator-Set 1197 per EID-Prefix. Therefore, when a Locator becomes unreachable, the 1198 Locator-Status-Bit that corresponds to that Locator's position in the 1199 list returned by the last Map-Reply will be set to zero for that 1200 particular EID-Prefix. 1202 When ITRs at the site are not deployed in CE routers, the IGP can 1203 still be used to determine the reachability of Locators, provided 1204 they are injected into the IGP. This is typically done when a /32 1205 address is configured on a loopback interface. 1207 When ITRs receive ICMP Network Unreachable or Host Unreachable 1208 messages as a method to determine unreachability, they will refrain 1209 from using Locators that are described in Locator lists of Map- 1210 Replies. However, using this approach is unreliable because many 1211 network operators turn off generation of ICMP Destination Unreachable 1212 messages. 1214 If an ITR does receive an ICMP Network Unreachable or Host 1215 Unreachable message, it MAY originate its own ICMP Destination 1216 Unreachable message destined for the host that originated the data 1217 packet the ITR encapsulated. 1219 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a 1220 locator address from a Locator-Set in a mapping entry matches a 1221 prefix. If it does not find one and BGP is running in the Default- 1222 Free Zone (DFZ), it can decide to not use the Locator even though the 1223 Locator-Status-Bits indicate that the Locator is up. In this case, 1224 the path from the ITR to the ETR that is assigned the Locator is not 1225 available. More details are in [I-D.meyer-loc-id-implications]. 1227 Optionally, an ITR can send a Map-Request to a Locator, and if a Map- 1228 Reply is returned, reachability of the Locator has been determined. 1229 Obviously, sending such probes increases the number of control 1230 messages originated by Tunnel Routers for active flows, so Locators 1231 are assumed to be reachable when they are advertised. 1233 This assumption does create a dependency: Locator unreachability is 1234 detected by the receipt of ICMP Host Unreachable messages. When a 1235 Locator has been determined to be unreachable, it is not used for 1236 active traffic; this is the same as if it were listed in a Map-Reply 1237 with Priority 255. 1239 The ITR can test the reachability of the unreachable Locator by 1240 sending periodic Requests. Both Requests and Replies MUST be rate- 1241 limited. Locator reachability testing is never done with data 1242 packets, since that increases the risk of packet loss for end-to-end 1243 sessions. 1245 When an ETR decapsulates a packet, it knows that it is reachable from 1246 the encapsulating ITR because that is how the packet arrived. In 1247 most cases, the ETR can also reach the ITR but cannot assume this to 1248 be true, due to the possibility of path asymmetry. In the presence 1249 of unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD 1250 NOT use the lack of return traffic as an indication that the ETR is 1251 unreachable. Instead, it MUST use an alternate mechanism to 1252 determine reachability. 1254 10.1. Echo Nonce Algorithm 1256 When data flows bidirectionally between Locators from different 1257 sites, a data-plane mechanism called "nonce echoing" can be used to 1258 determine reachability between an ITR and ETR. When an ITR wants to 1259 solicit a nonce echo, it sets the N- and E-bits and places a 24-bit 1260 nonce [RFC4086] in the LISP header of the next encapsulated data 1261 packet. 1263 When this packet is received by the ETR, the encapsulated packet is 1264 forwarded as normal. When the ETR next sends a data packet to the 1265 ITR, it includes the nonce received earlier with the N-bit set and 1266 E-bit cleared. The ITR sees this "echoed nonce" and knows that the 1267 path to and from the ETR is up. 1269 The ITR will set the E-bit and N-bit for every packet it sends while 1270 in the echo-nonce-request state. The time the ITR waits to process 1271 the echoed nonce before it determines the path is unreachable is 1272 variable and is a choice left for the implementation. 1274 If the ITR is receiving packets from the ETR but does not see the 1275 nonce echoed while being in the echo-nonce-request state, then the 1276 path to the ETR is unreachable. This decision may be overridden by 1277 other Locator reachability algorithms. Once the ITR determines that 1278 the path to the ETR is down, it can switch to another Locator for 1279 that EID-Prefix. 1281 Note that "ITR" and "ETR" are relative terms here. Both devices MUST 1282 be implementing both ITR and ETR functionality for the echo nonce 1283 mechanism to operate. 1285 The ITR and ETR may both go into the echo-nonce-request state at the 1286 same time. The number of packets sent or the time during which echo 1287 nonce requests are sent is an implementation-specific setting. 1288 However, when an ITR is in the echo-nonce-request state, it can echo 1289 the ETR's nonce in the next set of packets that it encapsulates and 1290 subsequently continue sending echo-nonce-request packets. 1292 This mechanism does not completely solve the forward path 1293 reachability problem, as traffic may be unidirectional. That is, the 1294 ETR receiving traffic at a site may not be the same device as an ITR 1295 that transmits traffic from that site, or the site-to-site traffic is 1296 unidirectional so there is no ITR returning traffic. 1298 The echo-nonce algorithm is bilateral. That is, if one side sets the 1299 E-bit and the other side is not enabled for echo-noncing, then the 1300 echoing of the nonce does not occur and the requesting side may 1301 erroneously consider the Locator unreachable. An ITR SHOULD only set 1302 the E-bit in an encapsulated data packet when it knows the ETR is 1303 enabled for echo-noncing. This is conveyed by the E-bit in the Map- 1304 Reply message. 1306 Note that other Locator reachability mechanisms are being researched 1307 and can be used to compliment or even override the echo nonce 1308 algorithm. See the next section for an example of control-plane 1309 probing. 1311 10.2. RLOC-Probing Algorithm 1313 RLOC-Probing is a method that an ITR or PITR can use to determine the 1314 reachability status of one or more Locators that it has cached in a 1315 Map-Cache entry. The probe-bit of the Map-Request and Map-Reply 1316 messages is used for RLOC-Probing. 1318 RLOC-Probing is done in the control plane on a timer basis, where an 1319 ITR or PITR will originate a Map-Request destined to a locator 1320 address from one of its own locator addresses. A Map-Request used as 1321 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or to 1322 the mapping database system as one would when soliciting mapping 1323 data. The EID record encoded in the Map-Request is the EID-Prefix of 1324 the Map-Cache entry cached by the ITR or PITR. The ITR may include a 1325 mapping data record for its own database mapping information that 1326 contains the local EID-Prefixes and RLOCs for its site. RLOC-probes 1327 are sent periodically using a jittered timer interval. 1329 When an ETR receives a Map-Request message with the probe-bit set, it 1330 returns a Map-Reply with the probe-bit set. The source address of 1331 the Map-Reply is set according to the procedure described in 1332 [I-D.ietf-lisp-rfc6833bis]. The Map-Reply SHOULD contain mapping 1333 data for the EID-Prefix contained in the Map-Request. This provides 1334 the opportunity for the ITR or PITR that sent the RLOC-probe to get 1335 mapping updates if there were changes to the ETR's database mapping 1336 entries. 1338 There are advantages and disadvantages of RLOC-Probing. The greatest 1339 benefit of RLOC-Probing is that it can handle many failure scenarios 1340 allowing the ITR to determine when the path to a specific Locator is 1341 reachable or has become unreachable, thus providing a robust 1342 mechanism for switching to using another Locator from the cached 1343 Locator. RLOC-Probing can also provide rough Round-Trip Time (RTT) 1344 estimates between a pair of Locators, which can be useful for network 1345 management purposes as well as for selecting low delay paths. The 1346 major disadvantage of RLOC-Probing is in the number of control 1347 messages required and the amount of bandwidth used to obtain those 1348 benefits, especially if the requirement for failure detection times 1349 is very small. 1351 Continued research and testing will attempt to characterize the 1352 tradeoffs of failure detection times versus message overhead. 1354 11. EID Reachability within a LISP Site 1356 A site may be multihomed using two or more ETRs. The hosts and 1357 infrastructure within a site will be addressed using one or more EID- 1358 Prefixes that are mapped to the RLOCs of the relevant ETRs in the 1359 mapping system. One possible failure mode is for an ETR to lose 1360 reachability to one or more of the EID-Prefixes within its own site. 1361 When this occurs when the ETR sends Map-Replies, it can clear the 1362 R-bit associated with its own Locator. And when the ETR is also an 1363 ITR, it can clear its Locator-Status-Bit in the encapsulation data 1364 header. 1366 It is recognized that there are no simple solutions to the site 1367 partitioning problem because it is hard to know which part of the 1368 EID-Prefix range is partitioned and which Locators can reach any sub- 1369 ranges of the EID-Prefixes. This problem is under investigation with 1370 the expectation that experiments will tell us more. Note that this 1371 is not a new problem introduced by the LISP architecture. The 1372 problem exists today when a multihomed site uses BGP to advertise its 1373 reachability upstream. 1375 12. Routing Locator Hashing 1377 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 1378 a requesting ITR, the Locator-Set for the EID-Prefix may contain 1379 different Priority values for each locator address. When more than 1380 one best Priority Locator exists, the ITR can decide how to load- 1381 share traffic against the corresponding Locators. 1383 The following hash algorithm may be used by an ITR to select a 1384 Locator for a packet destined to an EID for the EID-to-RLOC mapping: 1386 1. Either a source and destination address hash or the traditional 1387 5-tuple hash can be used. The traditional 5-tuple hash includes 1388 the source and destination addresses; source and destination TCP, 1389 UDP, or Stream Control Transmission Protocol (SCTP) port numbers; 1390 and the IP protocol number field or IPv6 next-protocol fields of 1391 a packet that a host originates from within a LISP site. When a 1392 packet is not a TCP, UDP, or SCTP packet, the source and 1393 destination addresses only from the header are used to compute 1394 the hash. 1396 2. Take the hash value and divide it by the number of Locators 1397 stored in the Locator-Set for the EID-to-RLOC mapping. 1399 3. The remainder will yield a value of 0 to "number of Locators 1400 minus 1". Use the remainder to select the Locator in the 1401 Locator-Set. 1403 Note that when a packet is LISP encapsulated, the source port number 1404 in the outer UDP header needs to be set. Selecting a hashed value 1405 allows core routers that are attached to Link Aggregation Groups 1406 (LAGs) to load-split the encapsulated packets across member links of 1407 such LAGs. Otherwise, core routers would see a single flow, since 1408 packets have a source address of the ITR, for packets that are 1409 originated by different EIDs at the source site. A suggested setting 1410 for the source port number computed by an ITR is a 5-tuple hash 1411 function on the inner header, as described above. 1413 Many core router implementations use a 5-tuple hash to decide how to 1414 balance packet load across members of a LAG. The 5-tuple hash 1415 includes the source and destination addresses of the packet and the 1416 source and destination ports when the protocol number in the packet 1417 is TCP or UDP. For this reason, UDP encoding is used for LISP 1418 encapsulation. 1420 13. Changing the Contents of EID-to-RLOC Mappings 1422 Since the LISP architecture uses a caching scheme to retrieve and 1423 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 1424 date mapping is to re-request the mapping. However, the ITRs do not 1425 know when the mappings change, and the ETRs do not keep track of 1426 which ITRs requested its mappings. For scalability reasons, we want 1427 to maintain this approach but need to provide a way for ETRs to 1428 change their mappings and inform the sites that are currently 1429 communicating with the ETR site using such mappings. 1431 When adding a new Locator record in lexicographic order to the end of 1432 a Locator-Set, it is easy to update mappings. We assume that new 1433 mappings will maintain the same Locator ordering as the old mapping 1434 but will just have new Locators appended to the end of the list. So, 1435 some ITRs can have a new mapping while other ITRs have only an old 1436 mapping that is used until they time out. When an ITR has only an 1437 old mapping but detects bits set in the Locator-Status-Bits that 1438 correspond to Locators beyond the list it has cached, it simply 1439 ignores them. However, this can only happen for locator addresses 1440 that are lexicographically greater than the locator addresses in the 1441 existing Locator-Set. 1443 When a Locator record is inserted in the middle of a Locator-Set, to 1444 maintain lexicographic order, the SMR procedure in Section 13.2 is 1445 used to inform ITRs and PITRs of the new Locator-Status-Bit mappings. 1447 When a Locator record is removed from a Locator-Set, ITRs that have 1448 the mapping cached will not use the removed Locator because the xTRs 1449 will set the Locator-Status-Bit to 0. So, even if the Locator is in 1450 the list, it will not be used. For new mapping requests, the xTRs 1451 can set the Locator AFI to 0 (indicating an unspecified address), as 1452 well as setting the corresponding Locator-Status-Bit to 0. This 1453 forces ITRs with old or new mappings to avoid using the removed 1454 Locator. 1456 If many changes occur to a mapping over a long period of time, one 1457 will find empty record slots in the middle of the Locator-Set and new 1458 records appended to the Locator-Set. At some point, it would be 1459 useful to compact the Locator-Set so the Locator-Status-Bit settings 1460 can be efficiently packed. 1462 We propose here three approaches for Locator-Set compaction: one 1463 operational mechanism and two protocol mechanisms. The operational 1464 approach uses a clock sweep method. The protocol approaches use the 1465 concept of Solicit-Map-Requests and Map-Versioning. 1467 13.1. Clock Sweep 1469 The clock sweep approach uses planning in advance and the use of 1470 count-down TTLs to time out mappings that have already been cached. 1471 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So, 1472 there is a 24-hour window to time out old mappings. The following 1473 clock sweep procedure is used: 1475 1. 24 hours before a mapping change is to take effect, a network 1476 administrator configures the ETRs at a site to start the clock 1477 sweep window. 1479 2. During the clock sweep window, ETRs continue to send Map-Reply 1480 messages with the current (unchanged) mapping records. The TTL 1481 for these mappings is set to 1 hour. 1483 3. 24 hours later, all previous cache entries will have timed out, 1484 and any active cache entries will time out within 1 hour. During 1485 this 1-hour window, the ETRs continue to send Map-Reply messages 1486 with the current (unchanged) mapping records with the TTL set to 1487 1 minute. 1489 4. At the end of the 1-hour window, the ETRs will send Map-Reply 1490 messages with the new (changed) mapping records. So, any active 1491 caches can get the new mapping contents right away if not cached, 1492 or in 1 minute if they had the mapping cached. The new mappings 1493 are cached with a TTL equal to the TTL in the Map-Reply. 1495 13.2. Solicit-Map-Request (SMR) 1497 Soliciting a Map-Request is a selective way for ETRs, at the site 1498 where mappings change, to control the rate they receive requests for 1499 Map-Reply messages. SMRs are also used to tell remote ITRs to update 1500 the mappings they have cached. 1502 Since the ETRs don't keep track of remote ITRs that have cached their 1503 mappings, they do not know which ITRs need to have their mappings 1504 updated. As a result, an ETR will solicit Map-Requests (called an 1505 SMR message) from those sites to which it has been sending 1506 encapsulated data for the last minute. In particular, an ETR will 1507 send an SMR to an ITR to which it has recently sent encapsulated 1508 data. 1510 An SMR message is simply a bit set in a Map-Request message. An ITR 1511 or PITR will send a Map-Request when they receive an SMR message. 1512 Both the SMR sender and the Map-Request responder MUST rate-limit 1513 these messages. Rate-limiting can be implemented as a global rate- 1514 limiter or one rate-limiter per SMR destination. 1516 The following procedure shows how an SMR exchange occurs when a site 1517 is doing Locator-Set compaction for an EID-to-RLOC mapping: 1519 1. When the database mappings in an ETR change, the ETRs at the site 1520 begin to send Map-Requests with the SMR bit set for each Locator 1521 in each Map-Cache entry the ETR caches. 1523 2. A remote ITR that receives the SMR message will schedule sending 1524 a Map-Request message to the source locator address of the SMR 1525 message or to the mapping database system. A newly allocated 1526 random nonce is selected, and the EID-Prefix used is the one 1527 copied from the SMR message. If the source Locator is the only 1528 Locator in the cached Locator-Set, the remote ITR SHOULD send a 1529 Map-Request to the database mapping system just in case the 1530 single Locator has changed and may no longer be reachable to 1531 accept the Map-Request. 1533 3. The remote ITR MUST rate-limit the Map-Request until it gets a 1534 Map-Reply while continuing to use the cached mapping. When 1535 Map-Versioning as described in Section 13.3 is used, an SMR 1536 sender can detect if an ITR is using the most up-to-date database 1537 mapping. 1539 4. The ETRs at the site with the changed mapping will reply to the 1540 Map-Request with a Map-Reply message that has a nonce from the 1541 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate- 1542 limited. This is important to avoid Map-Reply implosion. 1544 5. The ETRs at the site with the changed mapping record the fact 1545 that the site that sent the Map-Request has received the new 1546 mapping data in the Map-Cache entry for the remote site so the 1547 Locator-Status-Bits are reflective of the new mapping for packets 1548 going to the remote site. The ETR then stops sending SMR 1549 messages. 1551 For security reasons, an ITR MUST NOT process unsolicited Map- 1552 Replies. To avoid Map-Cache entry corruption by a third party, a 1553 sender of an SMR-based Map-Request MUST be verified. If an ITR 1554 receives an SMR-based Map-Request and the source is not in the 1555 Locator-Set for the stored Map-Cache entry, then the responding Map- 1556 Request MUST be sent with an EID destination to the mapping database 1557 system. Since the mapping database system is a more secure way to 1558 reach an authoritative ETR, it will deliver the Map-Request to the 1559 authoritative source of the mapping data. 1561 When an ITR receives an SMR-based Map-Request for which it does not 1562 have a cached mapping for the EID in the SMR message, it MAY not send 1563 an SMR-invoked Map-Request. This scenario can occur when an ETR 1564 sends SMR messages to all Locators in the Locator-Set it has stored 1565 in its map-cache but the remote ITRs that receive the SMR may not be 1566 sending packets to the site. There is no point in updating the ITRs 1567 until they need to send, in which case they will send Map-Requests to 1568 obtain a Map-Cache entry. 1570 13.3. Database Map-Versioning 1572 When there is unidirectional packet flow between an ITR and ETR, and 1573 the EID-to-RLOC mappings change on the ETR, it needs to inform the 1574 ITR so encapsulation to a removed Locator can stop and can instead be 1575 started to a new Locator in the Locator-Set. 1577 An ETR, when it sends Map-Reply messages, conveys its own Map-Version 1578 Number. This is known as the Destination Map-Version Number. ITRs 1579 include the Destination Map-Version Number in packets they 1580 encapsulate to the site. When an ETR decapsulates a packet and 1581 detects that the Destination Map-Version Number is less than the 1582 current version for its mapping, the SMR procedure described in 1583 Section 13.2 occurs. 1585 An ITR, when it encapsulates packets to ETRs, can convey its own Map- 1586 Version Number. This is known as the Source Map-Version Number. 1587 When an ETR decapsulates a packet and detects that the Source Map- 1588 Version Number is greater than the last Map-Version Number sent in a 1589 Map-Reply from the ITR's site, the ETR will send a Map-Request to one 1590 of the ETRs for the source site. 1592 A Map-Version Number is used as a sequence number per EID-Prefix, so 1593 values that are greater are considered to be more recent. A value of 1594 0 for the Source Map-Version Number or the Destination Map-Version 1595 Number conveys no versioning information, and an ITR does no 1596 comparison with previously received Map-Version Numbers. 1598 A Map-Version Number can be included in Map-Register messages as 1599 well. This is a good way for the Map-Server to assure that all ETRs 1600 for a site registering to it will be synchronized according to Map- 1601 Version Number. 1603 See [RFC6834] for a more detailed analysis and description of 1604 Database Map-Versioning. 1606 14. Multicast Considerations 1608 A multicast group address, as defined in the original Internet 1609 architecture, is an identifier of a grouping of topologically 1610 independent receiver host locations. The address encoding itself 1611 does not determine the location of the receiver(s). The multicast 1612 routing protocol, and the network-based state the protocol creates, 1613 determine where the receivers are located. 1615 In the context of LISP, a multicast group address is both an EID and 1616 a Routing Locator. Therefore, no specific semantic or action needs 1617 to be taken for a destination address, as it would appear in an IP 1618 header. Therefore, a group address that appears in an inner IP 1619 header built by a source host will be used as the destination EID. 1620 The outer IP header (the destination Routing Locator address), 1621 prepended by a LISP router, can use the same group address as the 1622 destination Routing Locator, use a multicast or unicast Routing 1623 Locator obtained from a Mapping System lookup, or use other means to 1624 determine the group address mapping. 1626 With respect to the source Routing Locator address, the ITR prepends 1627 its own IP address as the source address of the outer IP header. 1628 Just like it would if the destination EID was a unicast address. 1629 This source Routing Locator address, like any other Routing Locator 1630 address, MUST be globally routable. 1632 There are two approaches for LISP-Multicast, one that uses native 1633 multicast routing in the underlay with no support from the Mapping 1634 System and the other that uses only unicast routing in the underlay 1635 with support from the Mapping System. See [RFC6831] and 1636 [I-D.ietf-lisp-signal-free-multicast], respectively, for details. 1637 Details for LISP-Multicast and interworking with non-LISP sites are 1638 described in [RFC6831] and [RFC6832]. 1640 15. Router Performance Considerations 1642 LISP is designed to be very "hardware-based forwarding friendly". A 1643 few implementation techniques can be used to incrementally implement 1644 LISP: 1646 o When a tunnel-encapsulated packet is received by an ETR, the outer 1647 destination address may not be the address of the router. This 1648 makes it challenging for the control plane to get packets from the 1649 hardware. This may be mitigated by creating special Forwarding 1650 Information Base (FIB) entries for the EID-Prefixes of EIDs served 1651 by the ETR (those for which the router provides an RLOC 1652 translation). These FIB entries are marked with a flag indicating 1653 that control-plane processing should be performed. The forwarding 1654 logic of testing for particular IP protocol number values is not 1655 necessary. There are a few proven cases where no changes to 1656 existing deployed hardware were needed to support the LISP data- 1657 plane. 1659 o On an ITR, prepending a new IP header consists of adding more 1660 octets to a MAC rewrite string and prepending the string as part 1661 of the outgoing encapsulation procedure. Routers that support 1662 Generic Routing Encapsulation (GRE) tunneling [RFC2784] or 6to4 1663 tunneling [RFC3056] may already support this action. 1665 o A packet's source address or interface the packet was received on 1666 can be used to select VRF (Virtual Routing/Forwarding). The VRF's 1667 routing table can be used to find EID-to-RLOC mappings. 1669 For performance issues related to map-cache management, see 1670 Section 19. 1672 16. Mobility Considerations 1674 There are several kinds of mobility, of which only some might be of 1675 concern to LISP. Essentially, they are as follows. 1677 16.1. Slow Mobility 1679 A site wishes to change its attachment points to the Internet, and 1680 its LISP Tunnel Routers will have new RLOCs when it changes upstream 1681 providers. Changes in EID-to-RLOC mappings for sites are expected to 1682 be handled by configuration, outside of LISP. 1684 An individual endpoint wishes to move but is not concerned about 1685 maintaining session continuity. Renumbering is involved. LISP can 1686 help with the issues surrounding renumbering [RFC4192] [LISA96] by 1687 decoupling the address space used by a site from the address spaces 1688 used by its ISPs [RFC4984]. 1690 16.2. Fast Mobility 1692 Fast endpoint mobility occurs when an endpoint moves relatively 1693 rapidly, changing its IP-layer network attachment point. Maintenance 1694 of session continuity is a goal. This is where the Mobile IPv4 1695 [RFC5944] and Mobile IPv6 [RFC6275] [RFC4866] mechanisms are used and 1696 primarily where interactions with LISP need to be explored, such as 1697 the mechanisms in [I-D.portoles-lisp-eid-mobility] when the EID moves 1698 but the RLOC is in the network infrastructure. 1700 In LISP, one possibility is to "glean" information. When a packet 1701 arrives, the ETR could examine the EID-to-RLOC mapping and use that 1702 mapping for all outgoing traffic to that EID. It can do this after 1703 performing a route-returnability check, to ensure that the new 1704 network location does have an internal route to that endpoint. 1705 However, this does not cover the case where an ITR (the node assigned 1706 the RLOC) at the mobile-node location has been compromised. 1708 Mobile IP packet exchange is designed for an environment in which all 1709 routing information is disseminated before packets can be forwarded. 1710 In order to allow the Internet to grow to support expected future 1711 use, we are moving to an environment where some information may have 1712 to be obtained after packets are in flight. Modifications to IP 1713 mobility should be considered in order to optimize the behavior of 1714 the overall system. Anything that decreases the number of new EID- 1715 to-RLOC mappings needed when a node moves, or maintains the validity 1716 of an EID-to-RLOC mapping for a longer time, is useful. 1718 In addition to endpoints, a network can be mobile, possibly changing 1719 xTRs. A "network" can be as small as a single router and as large as 1720 a whole site. This is different from site mobility in that it is 1721 fast and possibly short-lived, but different from endpoint mobility 1722 in that a whole prefix is changing RLOCs. However, the mechanisms 1723 are the same, and there is no new overhead in LISP. A map request 1724 for any endpoint will return a binding for the entire mobile prefix. 1726 If mobile networks become a more common occurrence, it may be useful 1727 to revisit the design of the mapping service and allow for dynamic 1728 updates of the database. 1730 The issue of interactions between mobility and LISP needs to be 1731 explored further. Specific improvements to the entire system will 1732 depend on the details of mapping mechanisms. Mapping mechanisms 1733 should be evaluated on how well they support session continuity for 1734 mobile nodes. See [I-D.farinacci-lisp-predictive-rlocs] for more 1735 recent mechanisms which can provide near-zero packet loss during 1736 handoffs. 1738 16.3. LISP Mobile Node Mobility 1740 A mobile device can use the LISP infrastructure to achieve mobility 1741 by implementing the LISP encapsulation and decapsulation functions 1742 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile 1743 node" can use topologically independent EID IP addresses that are not 1744 advertised into and do not impose a cost on the global routing 1745 system. These EIDs are maintained at the edges of the mapping system 1746 in LISP Map-Servers and Map-Resolvers) and are provided on demand to 1747 only the correspondents of the LISP mobile node. 1749 Refer to [I-D.ietf-lisp-mn] for more details for when the EID and 1750 RLOC are co-located in the roaming node. 1752 17. LISP xTR Placement and Encapsulation Methods 1754 This section will explore how and where ITRs and ETRs can be placed 1755 in the network and will discuss the pros and cons of each scenario. 1756 For a more detailed networkd design deployment recommendation, refer 1757 to [RFC7215]. 1759 There are two basic deployment tradeoffs to consider: centralized 1760 versus distributed caches; and flat, Recursive, or Re-encapsulating 1761 Tunneling. When deciding on centralized versus distributed caching, 1762 the following issues should be considered: 1764 o Are the xTRs spread out so that the caches are spread across all 1765 the memories of each router? A centralized cache is when an ITR 1766 keeps a cache for all the EIDs it is encapsulating to. The packet 1767 takes a direct path to the destination Locator. A distributed 1768 cache is when an ITR needs help from other Re-Encapsulating Tunnel 1769 Routers (RTRs) because it does not store all the cache entries for 1770 the EIDs it is encapsulating to. So, the packet takes a path 1771 through RTRs that have a different set of cache entries. 1773 o Should management "touch points" be minimized by only choosing a 1774 few xTRs, just enough for redundancy? 1776 o In general, using more ITRs doesn't increase management load, 1777 since caches are built and stored dynamically. On the other hand, 1778 using more ETRs does require more management, since EID-Prefix-to- 1779 RLOC mappings need to be explicitly configured. 1781 When deciding on flat, Recursive, or Re-Encapsulating Tunneling, the 1782 following issues should be considered: 1784 o Flat tunneling implements a single encapsulation path between the 1785 source site and destination site. This generally offers better 1786 paths between sources and destinations with a single encapsulation 1787 path. 1789 o Recursive Tunneling is when encapsulated traffic is again further 1790 encapsulated in another tunnel, either to implement VPNs or to 1791 perform Traffic Engineering. When doing VPN-based tunneling, the 1792 site has some control, since the site is prepending a new 1793 encapsulation header. In the case of TE-based tunneling, the site 1794 may have control if it is prepending a new tunnel header, but if 1795 the site's ISP is doing the TE, then the site has no control. 1796 Recursive Tunneling generally will result in suboptimal paths but 1797 with the benefit of steering traffic to parts of the network that 1798 have more resources available. 1800 o The technique of Re-Encapsulation ensures that packets only 1801 require one encapsulation header. So, if a packet needs to be re- 1802 routed, it is first decapsulated by the RTR and then Re- 1803 Encapsulated with a new encapsulation header using a new RLOC. 1805 The next sub-sections will examine where xTRs and RTRs can reside in 1806 the network. 1808 17.1. First-Hop/Last-Hop xTRs 1810 By locating xTRs close to hosts, the EID-Prefix set is at the 1811 granularity of an IP subnet. So, at the expense of more EID-Prefix- 1812 to-RLOC sets for the site, the caches in each xTR can remain 1813 relatively small. But caches always depend on the number of non- 1814 aggregated EID destination flows active through these xTRs. 1816 With more xTRs doing encapsulation, the increase in control traffic 1817 grows as well: since the EID granularity is greater, more Map- 1818 Requests and Map-Replies are traveling between more routers. 1820 The advantage of placing the caches and databases at these stub 1821 routers is that the products deployed in this part of the network 1822 have better price-memory ratios than their core router counterparts. 1823 Memory is typically less expensive in these devices, and fewer routes 1824 are stored (only IGP routes). These devices tend to have excess 1825 capacity, both for forwarding and routing states. 1827 LISP functionality can also be deployed in edge switches. These 1828 devices generally have layer-2 ports facing hosts and layer-3 ports 1829 facing the Internet. Spare capacity is also often available in these 1830 devices. 1832 17.2. Border/Edge xTRs 1834 Using Customer Edge (CE) routers for xTR placement allows the EID 1835 space associated with a site to be reachable via a small set of RLOCs 1836 assigned to the CE-based xTRs for that site. 1838 This offers the opposite benefit of the first-hop/last-hop xTR 1839 scenario: the number of mapping entries and network management touch 1840 points is reduced, allowing better scaling. 1842 One disadvantage is that fewer network resources are used to reach 1843 host endpoints, thereby centralizing the point-of-failure domain and 1844 creating network choke points at the CE xTR. 1846 Note that more than one CE xTR at a site can be configured with the 1847 same IP address. In this case, an RLOC is an anycast address. This 1848 allows resilience between the CE xTRs. That is, if a CE xTR fails, 1849 traffic is automatically routed to the other xTRs using the same 1850 anycast address. However, this comes with the disadvantage where the 1851 site cannot control the entrance point when the anycast route is 1852 advertised out from all border routers. Another disadvantage of 1853 using anycast Locators is the limited advertisement scope of /32 (or 1854 /128 for IPv6) routes. 1856 17.3. ISP Provider Edge (PE) xTRs 1858 The use of ISP PE routers as xTRs is not the typical deployment 1859 scenario envisioned in this specification. This section attempts to 1860 capture some of the reasoning behind this preference for implementing 1861 LISP on CE routers. 1863 The use of ISP PE routers for xTR placement gives an ISP, rather than 1864 a site, control over the location of the ETRs. That is, the ISP can 1865 decide whether the xTRs are in the destination site (in either CE 1866 xTRs or last-hop xTRs within a site) or at other PE edges. The 1867 advantage of this case is that two encapsulation headers can be 1868 avoided. By having the PE be the first router on the path to 1869 encapsulate, it can choose a TE path first, and the ETR can 1870 decapsulate and Re-Encapsulate for a new encapsuluation path to the 1871 destination end site. 1873 An obvious disadvantage is that the end site has no control over 1874 where its packets flow or over the RLOCs used. Other disadvantages 1875 include difficulty in synchronizing path liveness updates between CE 1876 and PE routers. 1878 As mentioned in earlier sections, a combination of these scenarios is 1879 possible at the expense of extra packet header overhead; if both site 1880 and provider want control, then Recursive or Re-Encapsulating Tunnels 1881 are used. 1883 17.4. LISP Functionality with Conventional NATs 1885 LISP routers can be deployed behind Network Address Translator (NAT) 1886 devices to provide the same set of packet services hosts have today 1887 when they are addressed out of private address space. 1889 It is important to note that a locator address in any LISP control 1890 message MUST be a globally routable address and therefore SHOULD NOT 1891 contain [RFC1918] addresses. If a LISP xTR is configured with 1892 private RLOC addresses, they MUST be used only in the outer IP header 1893 so the NAT device can translate properly. Otherwise, EID addresses 1894 MUST be translated before encapsulation is performed when LISP VPNs 1895 are not in use. Both NAT translation and LISP encapsulation 1896 functions could be co-located in the same device. 1898 17.5. Packets Egressing a LISP Site 1900 When a LISP site is using two ITRs for redundancy, the failure of one 1901 ITR will likely shift outbound traffic to the second. This second 1902 ITR's cache may not be populated with the same EID-to-RLOC mapping 1903 entries as the first. If this second ITR does not have these 1904 mappings, traffic will be dropped while the mappings are retrieved 1905 from the mapping system. The retrieval of these messages may 1906 increase the load of requests being sent into the mapping system. 1908 18. Traceroute Considerations 1910 When a source host in a LISP site initiates a traceroute to a 1911 destination host in another LISP site, it is highly desirable for it 1912 to see the entire path. Since packets are encapsulated from the ITR 1913 to the ETR, the hop across the tunnel could be viewed as a single 1914 hop. However, LISP traceroute will provide the entire path so the 1915 user can see 3 distinct segments of the path from a source LISP host 1916 to a destination LISP host: 1918 Segment 1 (in source LISP site based on EIDs): 1920 source host ---> first hop ... next hop ---> ITR 1922 Segment 2 (in the core network based on RLOCs): 1924 ITR ---> next hop ... next hop ---> ETR 1926 Segment 3 (in the destination LISP site based on EIDs): 1928 ETR ---> next hop ... last hop ---> destination host 1930 For segment 1 of the path, ICMP Time Exceeded messages are returned 1931 in the normal manner as they are today. The ITR performs a TTL 1932 decrement and tests for 0 before encapsulating. Therefore, the ITR's 1933 hop is seen by the traceroute source as having an EID address (the 1934 address of the site-facing interface). 1936 For segment 2 of the path, ICMP Time Exceeded messages are returned 1937 to the ITR because the TTL decrement to 0 is done on the outer 1938 header, so the destinations of the ICMP messages are the ITR RLOC 1939 address and the source RLOC address of the encapsulated traceroute 1940 packet. The ITR looks inside of the ICMP payload to inspect the 1941 traceroute source so it can return the ICMP message to the address of 1942 the traceroute client and also retain the core router IP address in 1943 the ICMP message. This is so the traceroute client can display the 1944 core router address (the RLOC address) in the traceroute output. The 1945 ETR returns its RLOC address and responds to the TTL decrement to 0, 1946 as the previous core routers did. 1948 For segment 3, the next-hop router downstream from the ETR will be 1949 decrementing the TTL for the packet that was encapsulated, sent into 1950 the core, decapsulated by the ETR, and forwarded because it isn't the 1951 final destination. If the TTL is decremented to 0, any router on the 1952 path to the destination of the traceroute, including the next-hop 1953 router or destination, will send an ICMP Time Exceeded message to the 1954 source EID of the traceroute client. The ICMP message will be 1955 encapsulated by the local ITR and sent back to the ETR in the 1956 originated traceroute source site, where the packet will be delivered 1957 to the host. 1959 18.1. IPv6 Traceroute 1961 IPv6 traceroute follows the procedure described above, since the 1962 entire traceroute data packet is included in the ICMP Time Exceeded 1963 message payload. Therefore, only the ITR needs to pay special 1964 attention to forwarding ICMP messages back to the traceroute source. 1966 18.2. IPv4 Traceroute 1968 For IPv4 traceroute, we cannot follow the above procedure, since IPv4 1969 ICMP Time Exceeded messages only include the invoking IP header and 1970 8 octets that follow the IP header. Therefore, when a core router 1971 sends an IPv4 Time Exceeded message to an ITR, all the ITR has in the 1972 ICMP payload is the encapsulated header it prepended, followed by a 1973 UDP header. The original invoking IP header, and therefore the 1974 identity of the traceroute source, is lost. 1976 The solution we propose to solve this problem is to cache traceroute 1977 IPv4 headers in the ITR and to match them up with corresponding IPv4 1978 Time Exceeded messages received from core routers and the ETR. The 1979 ITR will use a circular buffer for caching the IPv4 and UDP headers 1980 of traceroute packets. It will select a 16-bit number as a key to 1981 find them later when the IPv4 Time Exceeded messages are received. 1982 When an ITR encapsulates an IPv4 traceroute packet, it will use the 1983 16-bit number as the UDP source port in the encapsulating header. 1984 When the ICMP Time Exceeded message is returned to the ITR, the UDP 1985 header of the encapsulating header is present in the ICMP payload, 1986 thereby allowing the ITR to find the cached headers for the 1987 traceroute source. The ITR puts the cached headers in the payload 1988 and sends the ICMP Time Exceeded message to the traceroute source 1989 retaining the source address of the original ICMP Time Exceeded 1990 message (a core router or the ETR of the site of the traceroute 1991 destination). 1993 The signature of a traceroute packet comes in two forms. The first 1994 form is encoded as a UDP message where the destination port is 1995 inspected for a range of values. The second form is encoded as an 1996 ICMP message where the IP identification field is inspected for a 1997 well-known value. 1999 18.3. Traceroute Using Mixed Locators 2001 When either an IPv4 traceroute or IPv6 traceroute is originated and 2002 the ITR encapsulates it in the other address family header, one 2003 cannot get all 3 segments of the traceroute. Segment 2 of the 2004 traceroute cannot be conveyed to the traceroute source, since it is 2005 expecting addresses from intermediate hops in the same address format 2006 for the type of traceroute it originated. Therefore, in this case, 2007 segment 2 will make the tunnel look like one hop. All the ITR has to 2008 do to make this work is to not copy the inner TTL to the outer, 2009 encapsulating header's TTL when a traceroute packet is encapsulated 2010 using an RLOC from a different address family. This will cause no 2011 TTL decrement to 0 to occur in core routers between the ITR and ETR. 2013 19. Security Considerations 2015 Security considerations for LISP are discussed in [RFC7833], in 2016 addition [I-D.ietf-lisp-sec] provides authentication and integrity to 2017 LISP mappings. 2019 A complete LISP threat analysis can be found in [RFC7835], in what 2020 follows we provide a summary. 2022 The optional mechanisms of gleaning is offered to directly obtain a 2023 mapping from the LISP encapsulated packets. Specifically, an xTR can 2024 learn the EID-to-RLOC mapping by inspecting the source RLOC and 2025 source EID of an encapsulated packet, and insert this new mapping 2026 into its map-cache. An off-path attacker can spoof the source EID 2027 address to divert the traffic sent to the victim's spoofed EID. If 2028 the attacker spoofs the source RLOC, it can mount a DoS attack by 2029 redirecting traffic to the spoofed victim;s RLOC, potentially 2030 overloading it. 2032 The LISP Data-Plane defines several mechanisms to monitor RLOC data- 2033 plane reachability, in this context Locator-Status Bits, Nonce- 2034 Present and Echo-Nonce bits of the LISP encapsulation header can be 2035 manipulated by an attacker to mount a DoS attack. An off-path 2036 attacker able to spoof the RLOC of a victim's xTR can manipulate such 2037 mechanisms to declare a set of RLOCs unreachable. This can be used 2038 also, for instance, to declare only one RLOC reachable with the aim 2039 of overload it. 2041 Map-Versioning is a data-plane mechanism used to signal a peering xTR 2042 that a local EID-to-RLOC mapping has been updated, so that the 2043 peering xTR uses LISP Control-Plane signaling message to retrieve a 2044 fresh mapping. This can be used by an attacker to forge the map- 2045 versioning field of a LISP encapsulated header and force an excessive 2046 amount of signaling between xTRs that may overload them. 2048 Most of the attack vectors can be mitigated with careful deployment 2049 and configuration, information learned opportunistically (such as LSB 2050 or gleaning) should be verified with other reachability mechanisms. 2051 In addition, systematic rate-limitation and filtering is an effective 2052 technique to mitigate attacks that aim to overload the control-plane. 2054 20. Network Management Considerations 2056 Considerations for network management tools exist so the LISP 2057 protocol suite can be operationally managed. These mechanisms can be 2058 found in [RFC7052] and [RFC6835]. 2060 21. IANA Considerations 2062 This section provides guidance to the Internet Assigned Numbers 2063 Authority (IANA) regarding registration of values related to this 2064 data-plane LISP specification, in accordance with BCP 26 [RFC5226]. 2066 21.1. LISP UDP Port Numbers 2068 The IANA registry has allocated UDP port numbers 4341 and 4342 for 2069 lisp-data and lisp-control operation, respectively. IANA has updated 2070 the description for UDP ports 4341 and 4342 as follows: 2072 lisp-data 4341 udp LISP Data Packets 2073 lisp-control 4342 udp LISP Control Packets 2075 22. References 2077 22.1. Normative References 2079 [I-D.ietf-lisp-ddt] 2080 Fuller, V., Lewis, D., Ermagan, V., Jain, A., and A. 2081 Smirnov, "LISP Delegated Database Tree", draft-ietf-lisp- 2082 ddt-09 (work in progress), January 2017. 2084 [I-D.ietf-lisp-introduction] 2085 Cabellos-Aparicio, A. and D. Saucez, "An Architectural 2086 Introduction to the Locator/ID Separation Protocol 2087 (LISP)", draft-ietf-lisp-introduction-13 (work in 2088 progress), April 2015. 2090 [I-D.ietf-lisp-rfc6833bis] 2091 Fuller, V., Farinacci, D., and A. Cabellos-Aparicio, 2092 "Locator/ID Separation Protocol (LISP) Control-Plane", 2093 draft-ietf-lisp-rfc6833bis-03 (work in progress), April 2094 2017. 2096 [I-D.ietf-lisp-sec] 2097 Maino, F., Ermagan, V., Cabellos-Aparicio, A., and D. 2098 Saucez, "LISP-Security (LISP-SEC)", draft-ietf-lisp-sec-12 2099 (work in progress), November 2016. 2101 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 2102 DOI 10.17487/RFC0768, August 1980, 2103 . 2105 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 2106 DOI 10.17487/RFC0791, September 1981, 2107 . 2109 [RFC1918] Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G., 2110 and E. Lear, "Address Allocation for Private Internets", 2111 BCP 5, RFC 1918, DOI 10.17487/RFC1918, February 1996, 2112 . 2114 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2115 Requirement Levels", BCP 14, RFC 2119, 2116 DOI 10.17487/RFC2119, March 1997, 2117 . 2119 [RFC2404] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within 2120 ESP and AH", RFC 2404, DOI 10.17487/RFC2404, November 2121 1998, . 2123 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 2124 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 2125 December 1998, . 2127 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 2128 of Explicit Congestion Notification (ECN) to IP", 2129 RFC 3168, DOI 10.17487/RFC3168, September 2001, 2130 . 2132 [RFC3232] Reynolds, J., Ed., "Assigned Numbers: RFC 1700 is Replaced 2133 by an On-line Database", RFC 3232, DOI 10.17487/RFC3232, 2134 January 2002, . 2136 [RFC4086] Eastlake 3rd, D., Schiller, J., and S. Crocker, 2137 "Randomness Requirements for Security", BCP 106, RFC 4086, 2138 DOI 10.17487/RFC4086, June 2005, 2139 . 2141 [RFC4632] Fuller, V. and T. Li, "Classless Inter-domain Routing 2142 (CIDR): The Internet Address Assignment and Aggregation 2143 Plan", BCP 122, RFC 4632, DOI 10.17487/RFC4632, August 2144 2006, . 2146 [RFC4868] Kelly, S. and S. Frankel, "Using HMAC-SHA-256, HMAC-SHA- 2147 384, and HMAC-SHA-512 with IPsec", RFC 4868, 2148 DOI 10.17487/RFC4868, May 2007, 2149 . 2151 [RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an 2152 IANA Considerations Section in RFCs", BCP 26, RFC 5226, 2153 DOI 10.17487/RFC5226, May 2008, 2154 . 2156 [RFC5496] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path 2157 Forwarding (RPF) Vector TLV", RFC 5496, 2158 DOI 10.17487/RFC5496, March 2009, 2159 . 2161 [RFC5944] Perkins, C., Ed., "IP Mobility Support for IPv4, Revised", 2162 RFC 5944, DOI 10.17487/RFC5944, November 2010, 2163 . 2165 [RFC6115] Li, T., Ed., "Recommendation for a Routing Architecture", 2166 RFC 6115, DOI 10.17487/RFC6115, February 2011, 2167 . 2169 [RFC6275] Perkins, C., Ed., Johnson, D., and J. Arkko, "Mobility 2170 Support in IPv6", RFC 6275, DOI 10.17487/RFC6275, July 2171 2011, . 2173 [RFC6834] Iannone, L., Saucez, D., and O. Bonaventure, "Locator/ID 2174 Separation Protocol (LISP) Map-Versioning", RFC 6834, 2175 DOI 10.17487/RFC6834, January 2013, 2176 . 2178 [RFC6836] Fuller, V., Farinacci, D., Meyer, D., and D. Lewis, 2179 "Locator/ID Separation Protocol Alternative Logical 2180 Topology (LISP+ALT)", RFC 6836, DOI 10.17487/RFC6836, 2181 January 2013, . 2183 [RFC7052] Schudel, G., Jain, A., and V. Moreno, "Locator/ID 2184 Separation Protocol (LISP) MIB", RFC 7052, 2185 DOI 10.17487/RFC7052, October 2013, 2186 . 2188 [RFC7214] Andersson, L. and C. Pignataro, "Moving Generic Associated 2189 Channel (G-ACh) IANA Registries to a New Registry", 2190 RFC 7214, DOI 10.17487/RFC7214, May 2014, 2191 . 2193 [RFC7215] Jakab, L., Cabellos-Aparicio, A., Coras, F., Domingo- 2194 Pascual, J., and D. Lewis, "Locator/Identifier Separation 2195 Protocol (LISP) Network Element Deployment 2196 Considerations", RFC 7215, DOI 10.17487/RFC7215, April 2197 2014, . 2199 [RFC7833] Howlett, J., Hartman, S., and A. Perez-Mendez, Ed., "A 2200 RADIUS Attribute, Binding, Profiles, Name Identifier 2201 Format, and Confirmation Methods for the Security 2202 Assertion Markup Language (SAML)", RFC 7833, 2203 DOI 10.17487/RFC7833, May 2016, 2204 . 2206 [RFC7835] Saucez, D., Iannone, L., and O. Bonaventure, "Locator/ID 2207 Separation Protocol (LISP) Threat Analysis", RFC 7835, 2208 DOI 10.17487/RFC7835, April 2016, 2209 . 2211 [RFC8061] Farinacci, D. and B. Weis, "Locator/ID Separation Protocol 2212 (LISP) Data-Plane Confidentiality", RFC 8061, 2213 DOI 10.17487/RFC8061, February 2017, 2214 . 2216 22.2. Informative References 2218 [AFN] IANA, "Address Family Numbers", August 2016, 2219 . 2221 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed", 2222 1999, 2223 . 2225 [I-D.farinacci-lisp-predictive-rlocs] 2226 Farinacci, D. and P. Pillay-Esnault, "LISP Predictive 2227 RLOCs", draft-farinacci-lisp-predictive-rlocs-01 (work in 2228 progress), November 2016. 2230 [I-D.ietf-lisp-mn] 2231 Farinacci, D., Lewis, D., Meyer, D., and C. White, "LISP 2232 Mobile Node", draft-ietf-lisp-mn-00 (work in progress), 2233 April 2017. 2235 [I-D.ietf-lisp-signal-free-multicast] 2236 Moreno, V. and D. Farinacci, "Signal-Free LISP Multicast", 2237 draft-ietf-lisp-signal-free-multicast-03 (work in 2238 progress), April 2017. 2240 [I-D.meyer-loc-id-implications] 2241 Meyer, D. and D. Lewis, "Architectural Implications of 2242 Locator/ID Separation", draft-meyer-loc-id-implications-01 2243 (work in progress), January 2009. 2245 [I-D.portoles-lisp-eid-mobility] 2246 Portoles-Comeras, M., Ashtaputre, V., Moreno, V., Maino, 2247 F., and D. Farinacci, "LISP L2/L3 EID Mobility Using a 2248 Unified Control Plane", draft-portoles-lisp-eid- 2249 mobility-02 (work in progress), April 2017. 2251 [LISA96] Lear, E., Tharp, D., Katinsky, J., and J. Coffin, 2252 "Renumbering: Threat or Menace?", Usenix Tenth System 2253 Administration Conference (LISA 96), October 1996. 2255 [OPENLISP] 2256 Iannone, L., Saucez, D., and O. Bonaventure, "OpenLISP 2257 Implementation Report", Work in Progress, July 2008. 2259 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 2260 STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, 2261 . 2263 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 2264 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 2265 DOI 10.17487/RFC2784, March 2000, 2266 . 2268 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 2269 via IPv4 Clouds", RFC 3056, DOI 10.17487/RFC3056, February 2270 2001, . 2272 [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, 2273 A., Peterson, J., Sparks, R., Handley, M., and E. 2274 Schooler, "SIP: Session Initiation Protocol", RFC 3261, 2275 DOI 10.17487/RFC3261, June 2002, 2276 . 2278 [RFC4107] Bellovin, S. and R. Housley, "Guidelines for Cryptographic 2279 Key Management", BCP 107, RFC 4107, DOI 10.17487/RFC4107, 2280 June 2005, . 2282 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2283 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2284 DOI 10.17487/RFC4192, September 2005, 2285 . 2287 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2288 Optimization for Mobile IPv6", RFC 4866, 2289 DOI 10.17487/RFC4866, May 2007, 2290 . 2292 [RFC4984] Meyer, D., Ed., Zhang, L., Ed., and K. Fall, Ed., "Report 2293 from the IAB Workshop on Routing and Addressing", 2294 RFC 4984, DOI 10.17487/RFC4984, September 2007, 2295 . 2297 [RFC6480] Lepinski, M. and S. Kent, "An Infrastructure to Support 2298 Secure Internet Routing", RFC 6480, DOI 10.17487/RFC6480, 2299 February 2012, . 2301 [RFC6518] Lebovitz, G. and M. Bhatia, "Keying and Authentication for 2302 Routing Protocols (KARP) Design Guidelines", RFC 6518, 2303 DOI 10.17487/RFC6518, February 2012, 2304 . 2306 [RFC6831] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, "The 2307 Locator/ID Separation Protocol (LISP) for Multicast 2308 Environments", RFC 6831, DOI 10.17487/RFC6831, January 2309 2013, . 2311 [RFC6832] Lewis, D., Meyer, D., Farinacci, D., and V. Fuller, 2312 "Interworking between Locator/ID Separation Protocol 2313 (LISP) and Non-LISP Sites", RFC 6832, 2314 DOI 10.17487/RFC6832, January 2013, 2315 . 2317 [RFC6835] Farinacci, D. and D. Meyer, "The Locator/ID Separation 2318 Protocol Internet Groper (LIG)", RFC 6835, 2319 DOI 10.17487/RFC6835, January 2013, 2320 . 2322 [RFC6837] Lear, E., "NERD: A Not-so-novel Endpoint ID (EID) to 2323 Routing Locator (RLOC) Database", RFC 6837, 2324 DOI 10.17487/RFC6837, January 2013, 2325 . 2327 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 2328 UDP Checksums for Tunneled Packets", RFC 6935, 2329 DOI 10.17487/RFC6935, April 2013, 2330 . 2332 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 2333 for the Use of IPv6 UDP Datagrams with Zero Checksums", 2334 RFC 6936, DOI 10.17487/RFC6936, April 2013, 2335 . 2337 [RFC8060] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical 2338 Address Format (LCAF)", RFC 8060, DOI 10.17487/RFC8060, 2339 February 2017, . 2341 Appendix A. Acknowledgments 2343 An initial thank you goes to Dave Oran for planting the seeds for the 2344 initial ideas for LISP. His consultation continues to provide value 2345 to the LISP authors. 2347 A special and appreciative thank you goes to Noel Chiappa for 2348 providing architectural impetus over the past decades on separation 2349 of location and identity, as well as detailed reviews of the LISP 2350 architecture and documents, coupled with enthusiasm for making LISP a 2351 practical and incremental transition for the Internet. 2353 The authors would like to gratefully acknowledge many people who have 2354 contributed discussions and ideas to the making of this proposal. 2355 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller, 2356 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston, 2357 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, 2358 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, 2359 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi 2360 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry 2361 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van 2362 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien 2363 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David 2364 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin, 2365 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari 2366 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu, 2367 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri 2368 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina 2369 Ermagan, Fabio Maino, Victor Moreno, Chris White, Clarence Filsfils, 2370 Alia Atlas, Florin Coras and Alberto Rodriguez. 2372 This work originated in the Routing Research Group (RRG) of the IRTF. 2373 An individual submission was converted into the IETF LISP working 2374 group document that became this RFC. 2376 The LISP working group would like to give a special thanks to Jari 2377 Arkko, the Internet Area AD at the time that the set of LISP 2378 documents were being prepared for IESG last call, and for his 2379 meticulous reviews and detailed commentaries on the 7 working group 2380 last call documents progressing toward standards-track RFCs. 2382 Appendix B. Document Change Log 2384 [RFC Editor: Please delete this section on publication as RFC.] 2386 B.1. Changes to draft-ietf-lisp-rfc6830bis-03 2388 o Posted May 2017. 2390 o Move the control-plane related codepoints in the IANA 2391 Considerations section to RFC6833bis. 2393 B.2. Changes to draft-ietf-lisp-rfc6830bis-02 2395 o Posted April 2017. 2397 o Reflect some editorial comments from Damien Sausez. 2399 B.3. Changes to draft-ietf-lisp-rfc6830bis-01 2401 o Posted March 2017. 2403 o Include references to new RFCs published. 2405 o Change references from RFC6833 to RFC6833bis. 2407 o Clarified LCAF text in the IANA section. 2409 o Remove references to "experimental". 2411 B.4. Changes to draft-ietf-lisp-rfc6830bis-00 2413 o Posted December 2016. 2415 o Created working group document from draft-farinacci-lisp 2416 -rfc6830-00 individual submission. No other changes made. 2418 Authors' Addresses 2420 Dino Farinacci 2421 Cisco Systems 2422 Tasman Drive 2423 San Jose, CA 95134 2424 USA 2426 EMail: farinacci@gmail.com 2427 Vince Fuller 2428 Cisco Systems 2429 Tasman Drive 2430 San Jose, CA 95134 2431 USA 2433 EMail: vince.fuller@gmail.com 2435 Dave Meyer 2436 Cisco Systems 2437 170 Tasman Drive 2438 San Jose, CA 2439 USA 2441 EMail: dmm@1-4-5.net 2443 Darrel Lewis 2444 Cisco Systems 2445 170 Tasman Drive 2446 San Jose, CA 2447 USA 2449 EMail: darlewis@cisco.com 2451 Albert Cabellos 2452 UPC/BarcelonaTech 2453 Campus Nord, C. Jordi Girona 1-3 2454 Barcelona, Catalunya 2455 Spain 2457 EMail: acabello@ac.upc.edu