idnits 2.17.1 draft-farinacci-lisp-10.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 18. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 2174. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 2185. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 2192. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 2198. 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 document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) == There are 5 instances of lines with non-RFC2606-compliant FQDNs in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'MUST not' in this paragraph: UDP Checksum: this field field MUST be transmitted as 0 and ignored on receipt by the ETR. Note, even when the UDP checksum is transmitted as 0 an intervening NAT device can recalculate the checksum and rewrite the UDP checksum field to non-zero. For performance reasons, the ETR MUST ignore the checksum and MUST not do a checksum computation. -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (November 26, 2008) is 5627 days in the past. Is this intentional? Checking references for intended status: Experimental ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 2434 (Obsoleted by RFC 5226) ** Obsolete normative reference: RFC 3775 (Obsoleted by RFC 6275) ** Obsolete normative reference: RFC 4423 (Obsoleted by RFC 9063) == Outdated reference: A later version (-05) exists of draft-fuller-lisp-alt-03 == Outdated reference: A later version (-01) exists of draft-jen-apt-00 == Outdated reference: A later version (-04) exists of draft-meyer-lisp-cons-03 == Outdated reference: A later version (-02) exists of draft-lewis-lisp-interworking-01 -- No information found for draft-mathy-lisp-dht - is the name correct? == Outdated reference: A later version (-09) exists of draft-lear-lisp-nerd-02 == Outdated reference: A later version (-05) exists of draft-narten-radir-problem-statement-00 == Outdated reference: A later version (-10) exists of draft-ietf-mip4-rfc3344bis-05 == Outdated reference: A later version (-08) exists of draft-ietf-pim-rpf-vector-03 -- No information found for draft-handley-p2ppush-unpublished-2007726 - is the name correct? == Outdated reference: A later version (-12) exists of draft-ietf-shim6-proto-06 Summary: 5 errors (**), 0 flaws (~~), 12 warnings (==), 9 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group D. Farinacci 3 Internet-Draft V. Fuller 4 Intended status: Experimental D. Oran 5 Expires: May 30, 2009 D. Meyer 6 S. Brim 7 cisco Systems 8 November 26, 2008 10 Locator/ID Separation Protocol (LISP) 11 draft-farinacci-lisp-10.txt 13 Status of this Memo 15 By submitting this Internet-Draft, each author represents that any 16 applicable patent or other IPR claims of which he or she is aware 17 have been or will be disclosed, and any of which he or she becomes 18 aware will be disclosed, in accordance with Section 6 of BCP 79. 20 Internet-Drafts are working documents of the Internet Engineering 21 Task Force (IETF), its areas, and its working groups. Note that 22 other groups may also distribute working documents as Internet- 23 Drafts. 25 Internet-Drafts are draft documents valid for a maximum of six months 26 and may be updated, replaced, or obsoleted by other documents at any 27 time. It is inappropriate to use Internet-Drafts as reference 28 material or to cite them other than as "work in progress." 30 The list of current Internet-Drafts can be accessed at 31 http://www.ietf.org/ietf/1id-abstracts.txt. 33 The list of Internet-Draft Shadow Directories can be accessed at 34 http://www.ietf.org/shadow.html. 36 This Internet-Draft will expire on May 30, 2009. 38 Copyright Notice 40 Copyright (C) The IETF Trust (2008). 42 Abstract 44 This draft describes a simple, incremental, network-based protocol to 45 implement separation of Internet addresses into Endpoint Identifiers 46 (EIDs) and Routing Locators (RLOCs). This mechanism requires no 47 changes to host stacks and no major changes to existing database 48 infrastructures. The proposed protocol can be implemented in a 49 relatively small number of routers. 51 This proposal was stimulated by the problem statement effort at the 52 Amsterdam IAB Routing and Addressing Workshop (RAWS), which took 53 place in October 2006. 55 Table of Contents 57 1. Requirements Notation . . . . . . . . . . . . . . . . . . . . 4 58 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 59 3. Definition of Terms . . . . . . . . . . . . . . . . . . . . . 8 60 4. Basic Overview . . . . . . . . . . . . . . . . . . . . . . . . 12 61 4.1. Packet Flow Sequence . . . . . . . . . . . . . . . . . . . 13 62 5. Tunneling Details . . . . . . . . . . . . . . . . . . . . . . 16 63 5.1. LISP IPv4-in-IPv4 Header Format . . . . . . . . . . . . . 17 64 5.2. LISP IPv6-in-IPv6 Header Format . . . . . . . . . . . . . 18 65 5.3. Tunnel Header Field Descriptions . . . . . . . . . . . . . 19 66 5.4. Dealing with Large Encapsulated Packets . . . . . . . . . 20 67 6. EID-to-RLOC Mapping . . . . . . . . . . . . . . . . . . . . . 22 68 6.1. Control Plane Packet Format . . . . . . . . . . . . . . . 22 69 6.1.1. LISP Packet Type Allocations . . . . . . . . . . . . . 24 70 6.1.2. Map-Request Message Format . . . . . . . . . . . . . . 24 71 6.1.3. EID-to-RLOC UDP Map-Request Message . . . . . . . . . 26 72 6.1.4. Map-Reply Message Format . . . . . . . . . . . . . . . 26 73 6.1.5. EID-to-RLOC UDP Map-Reply Message . . . . . . . . . . 29 74 6.2. Routing Locator Selection . . . . . . . . . . . . . . . . 29 75 6.3. Routing Locator Reachability . . . . . . . . . . . . . . . 30 76 6.4. Routing Locator Hashing . . . . . . . . . . . . . . . . . 32 77 6.5. Changing the Contents of EID-to-RLOC Mappings . . . . . . 33 78 6.5.1. Clock Sweep . . . . . . . . . . . . . . . . . . . . . 34 79 6.5.2. Solicit-Map-Request (SMR) . . . . . . . . . . . . . . 34 80 7. Router Performance Considerations . . . . . . . . . . . . . . 36 81 8. Deployment Scenarios . . . . . . . . . . . . . . . . . . . . . 37 82 8.1. First-hop/Last-hop Tunnel Routers . . . . . . . . . . . . 38 83 8.2. Border/Edge Tunnel Routers . . . . . . . . . . . . . . . . 38 84 8.3. ISP Provider-Edge (PE) Tunnel Routers . . . . . . . . . . 39 85 9. Traceroute Considerations . . . . . . . . . . . . . . . . . . 40 86 9.1. IPv6 Traceroute . . . . . . . . . . . . . . . . . . . . . 41 87 9.2. IPv4 Traceroute . . . . . . . . . . . . . . . . . . . . . 41 88 9.3. Traceroute using Mixed Locators . . . . . . . . . . . . . 41 90 10. Mobility Considerations . . . . . . . . . . . . . . . . . . . 43 91 10.1. Site Mobility . . . . . . . . . . . . . . . . . . . . . . 43 92 10.2. Slow Endpoint Mobility . . . . . . . . . . . . . . . . . . 43 93 10.3. Fast Endpoint Mobility . . . . . . . . . . . . . . . . . . 43 94 10.4. Fast Network Mobility . . . . . . . . . . . . . . . . . . 45 95 11. Multicast Considerations . . . . . . . . . . . . . . . . . . . 46 96 12. Security Considerations . . . . . . . . . . . . . . . . . . . 47 97 13. Prototype Plans and Status . . . . . . . . . . . . . . . . . . 48 98 14. References . . . . . . . . . . . . . . . . . . . . . . . . . . 50 99 14.1. Normative References . . . . . . . . . . . . . . . . . . . 50 100 14.2. Informative References . . . . . . . . . . . . . . . . . . 50 101 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 53 102 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 54 103 Intellectual Property and Copyright Statements . . . . . . . . . . 55 105 1. Requirements Notation 107 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 108 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 109 document are to be interpreted as described in [RFC2119]. 111 2. Introduction 113 Many years of discussion about the current IP routing and addressing 114 architecture have noted that its use of a single numbering space (the 115 "IP address") for both host transport session identification and 116 network routing creates scaling issues (see [CHIAPPA] and [RFC1498]). 117 A number of scaling benefits would be realized by separating the 118 current IP address into separate spaces for Endpoint Identifiers 119 (EIDs) and Routing Locators (RLOCs); among them are: 121 1. Reduction of routing table size in the "default-free zone" (DFZ). 122 Use of a separate numbering space for RLOCs will allow them to be 123 assigned topologically (in today's Internet, RLOCs would be 124 assigned by providers at client network attachment points), 125 greatly improving aggregation and reducing the number of 126 globally-visible, routable prefixes. 128 2. Easing of renumbering burden when clients change providers. 129 Because host EIDs are numbered from a separate, non-provider- 130 assigned and non-topologically-bound space, they do not need to 131 be renumbered when a client site changes its attachment points to 132 the network. 134 3. Traffic engineering capabilities that can be performed by network 135 elements and do not depend on injecting additional state into the 136 routing system. This will fall out of the mechanism that is used 137 to implement the EID/RLOC split (see Section 4). 139 4. Mobility without address changing. Existing mobility mechanisms 140 will be able to work in a locator/ID separation scenario. It 141 will be possible for a host (or a collection of hosts) to move to 142 a different point in the network topology either retaining its 143 home-based address or acquiring a new address based on the new 144 network location. A new network location could be a physically 145 different point in the network topology or the same physical 146 point of the topology with a different provider. 148 This draft describes protocol mechanisms to achieve the desired 149 functional separation. For flexibility, the document decouples the 150 mechanism used for forwarding packets from that used to determine EID 151 to RLOC mappings. This work is in response to and intended to 152 address the problem statement that came out of the RAWS effort 153 [RFC4984]. 155 The Routing and Addressing problem statement can be found in [RADIR]. 157 This draft focuses on a router-based solution. Building the solution 158 into the network should facilitate incremental deployment of the 159 technology on the Internet. Note that while the detailed protocol 160 specification and examples in this document assume IP version 4 161 (IPv4), there is nothing in the design that precludes use of the same 162 techniques and mechanisms for IPv6. It should be possible for IPv4 163 packets to use IPv6 RLOCs and for IPv6 EIDs to be mapped to IPv4 164 RLOCs. 166 Related work on host-based solutions is described in Shim6 [SHIM6] 167 and HIP [RFC4423]. Related work on a router-based solution is 168 described in [GSE]. This draft attempts to not compete or overlap 169 with such solutions and the proposed protocol changes are expected to 170 complement a host-based mechanism when Traffic Engineering 171 functionality is desired. 173 Some of the design goals of this proposal include: 175 1. Minimize required changes to Internet infrastructure. 177 2. Require no hardware or software changes to end-systems (hosts). 179 3. Be incrementally deployable. 181 4. Require no router hardware changes. 183 5. Minimize router software changes. 185 6. Avoid or minimize packet loss when EID-to-RLOC mappings need to 186 be performed. 188 There are 4 variants of LISP, which differ along a spectrum of strong 189 to weak dependence on the topological nature and possible need for 190 routability of EIDs. The variants are: 192 LISP 1: uses EIDs that are routable through the RLOC topology for 193 bootstrapping EID-to-RLOC mappings. [LISP1] This was intended as 194 a prototyping mechanism for early protocol implementation. It is 195 now deprecated and should not be deployed. 197 LISP 1.5: uses EIDs that are routable for bootstrapping EID-to-RLOC 198 mappings; such routing is via a separate topology. 200 LISP 2: uses EIDS that are not routable and EID-to-RLOC mappings are 201 implemented within the DNS. [LISP2] 203 LISP 3: uses non-routable EIDs that are used as lookup keys for a 204 new EID-to-RLOC mapping database. Use of Distributed Hash Tables 205 [DHTs] [LISPDHT] to implement such a database would be an area to 206 explore. Other examples of new mapping database services are 208 [CONS], [ALT], [RPMD], [NERD], and [APT]. 210 This document on LISP 1.5, and LISP 3 variants, both of which rely on 211 a router-based distributed cache and database for EID-to-RLOC 212 mappings. The LISP 1.0 mechanism works but does not allow reduction 213 of routing information in the default-free-zone of the Internet. The 214 LISP 2 mechanisms are put on hold and may never come to fruition 215 since it is not architecturally pure to have routing depend on 216 directory and directory depend on routing. The LISP 3 mechanisms 217 will be documented elsewhere but may use the control-plane options 218 specified in this specification. 220 3. Definition of Terms 222 Provider Independent (PI) Addresses: an address block assigned from 223 a pool that is not associated with any service provider and is 224 therefore not topologically-aggregatable in the routing system. 226 Provider Assigned (PA) Addresses: a block of IP addresses that are 227 assigned to a site by each service provider to which a site 228 connects. Typically, each block is sub-block of a service 229 provider CIDR block and is aggregated into the larger block before 230 being advertised into the global Internet. Traditionally, IP 231 multihoming has been implemented by each multi-homed site 232 acquiring its own, globally-visible prefix. LISP uses only 233 topologically-assigned and aggregatable address blocks for RLOCs, 234 eliminating this demonstrably non-scalable practice. 236 Routing Locator (RLOC): the IPv4 or IPv6 address of an egress 237 tunnel router (ETR). It is the output of a EID-to-RLOC mapping 238 lookup. An EID maps to one or more RLOCs. Typically, RLOCs are 239 numbered from topologically-aggregatable blocks that are assigned 240 to a site at each point to which it attaches to the global 241 Internet; where the topology is defined by the connectivity of 242 provider networks, RLOCs can be thought of as PA addresses. 243 Multiple RLOCs can be assigned to the same ETR device or to 244 multiple ETR devices at a site. 246 Endpoint ID (EID): a 32-bit (for IPv4) or 128-bit (for IPv6) value 247 used in the source and destination address fields of the first 248 (most inner) LISP header of a packet. The host obtains a 249 destination EID the same way it obtains an destination address 250 today, for example through a DNS lookup or SIP exchange. The 251 source EID is obtained via existing mechanisms used to set a 252 host's "local" IP address. An EID is allocated to a host from an 253 EID-prefix block associated with the site where the host is 254 located. An EID can be used by a host to refer to other hosts. 255 EIDs MUST NOT be used as LISP RLOCs. Note that EID blocks may be 256 assigned in a hierarchical manner, independent of the network 257 topology, to facilitate scaling of the mapping database. In 258 addition, an EID block assigned to a site may have site-local 259 structure (subnetting) for routing within the site; this structure 260 is not visible to the global routing system. 262 EID-prefix: A power-of-2 block of EIDs which are allocated to a 263 site by an address allocation authority. EID-prefixes are 264 associated with a set of RLOC addresses which make up a "database 265 mapping". EID-prefix allocations can be broken up into smaller 266 blocks when an RLOC set is to be associated with the smaller EID- 267 prefix. A globally routed address block (whether PI or PA) is not 268 an EID-prefix. However, a globally routed address block may be 269 removed from global routing and reused as an EID-prefix. A site 270 that receives an explicitly allocated EID-prefix may not use that 271 EID-prefix as a globally routed prefix assigned to RLOCs. 273 End-system: is an IPv4 or IPv6 device that originates packets with 274 a single IPv4 or IPv6 header. The end-system supplies an EID 275 value for the destination address field of the IP header when 276 communicating globally (i.e. outside of it's routing domain). An 277 end-system can be a host computer, a switch or router device, or 278 any network appliance. 280 Ingress Tunnel Router (ITR): a router which accepts an IP packet 281 with a single IP header (more precisely, an IP packet that does 282 not contain a LISP header). The router treats this "inner" IP 283 destination address as an EID and performs an EID-to-RLOC mapping 284 lookup. The router then prepends an "outer" IP header with one of 285 its globally-routable RLOCs in the source address field and the 286 result of the mapping lookup in the destination address field. 287 Note that this destination RLOC may be an intermediate, proxy 288 device that has better knowledge of the EID-to-RLOC mapping closer 289 to the destination EID. In general, an ITR receives IP packets 290 from site end-systems on one side and sends LISP-encapsulated IP 291 packets toward the Internet on the other side. 293 Specifically, when a service provider prepends a LISP header for 294 Traffic Engineering purposes, the router that does this is also 295 regarded as an ITR. The outer RLOC the ISP ITR uses can be based 296 on the outer destination address (the originating ITR's supplied 297 RLOC) or the inner destination address (the originating hosts 298 supplied EID). 300 TE-ITR: is an ITR that is deployed in a service provider network 301 that prepends an additional LISP header for Traffic Engineering 302 purposes. 304 Egress Tunnel Router (ETR): a router that accepts an IP packet 305 where the destination address in the "outer" IP header is one of 306 its own RLOCs. The router strips the "outer" header and forwards 307 the packet based on the next IP header found. In general, an ETR 308 receives LISP-encapsulated IP packets from the Internet on one 309 side and sends decapsulated IP packets to site end-systems on the 310 other side. ETR functionality does not have to be limited to a 311 router device. A server host can be the endpoint of a LISP tunnel 312 as well. 314 TE-ETR: is an ETR that is deployed in a service provider network 315 that strips an outer LISP header for Traffic Engineering purposes. 317 xTR: is a reference to an ITR or ETR when direction of data flow is 318 not part of the context description. xTR refers to the router that 319 is the tunnel endpoint. Used synonymously with the term "Tunnel 320 Router". For example, "An xTR can be located at the Customer Edge 321 (CE) router", meaning both ITR and ETR functionality is at the CE 322 router. 324 EID-to-RLOC Cache: a short-lived, on-demand database in an ITR that 325 stores, tracks, and is responsible for timing-out and otherwise 326 validating EID-to-RLOC mappings. This cache is distinct from the 327 "database", the cache is dynamic, local, and relatively small 328 while the database is distributed, relatively static, and much 329 more global in scope. 331 EID-to-RLOC Database: a global distributed database that contains 332 all known EID-prefix to RLOC mappings. Each potential ETR 333 typically contains a small piece of the database: the EID-to-RLOC 334 mappings for the EID prefixes "behind" the router. These map to 335 one of the router's own, globally-visible, IP addresses. 337 Recursive Tunneling: when a packet has more than one LISP IP 338 header. Additional layers of tunneling may be employed to 339 implement traffic engineering or other re-routing as needed. When 340 this is done, an additional "outer" LISP header is added and the 341 original RLOCs are preserved in the "inner" header. Any 342 references to tunnels in this specification refers to dynamic 343 encapsulating tunnels and never are they staticly configured. 345 Reencapsulating Tunnels: when a packet has no more than one LISP IP 346 header (two IP headers total) and when it needs to be diverted to 347 new RLOC, an ETR can decapsulate the packet (remove the LISP 348 header) and prepend a new tunnel header, with new RLOC, on to the 349 packet. Doing this allows a packet to be re-routed by the re- 350 encapsulating router without adding the overhead of additional 351 tunnel headers. Any references to tunnels in this specification 352 refers to dynamic encapsulating tunnels and never are they 353 staticly configured. 355 LISP Header: a term used in this document to refer to the outer 356 IPv4 or IPv6 header, a UDP header, and a LISP header, an ITR 357 prepends or an ETR strips. 359 Address Family Indicator (AFI): a term used to describe an address 360 encoding in a packet. An address family currently pertains to an 361 IPv4 or IPv6 address. See [AFI] for details. 363 Negative Mapping Entry: also known as a negative cache entry, is an 364 EID-to-RLOC entry where an EID-prefix is advertised or stored with 365 no RLOCs. That is, the locator-set for the EID-to-RLOC entry is 366 empty or has an encoded locator count of 0. This type of entry 367 could be used to describe a prefix from a non-LISP site, which is 368 explicitly not in the mapping database. 370 Data Probe: a LISP-encapsulated data packet where the inner header 371 destination address equals the outer header destination address 372 used to trigger a Map-Reply by a decapsulating ETR. In addition, 373 the original packet is decapsulated and delivered to the 374 destination host. A Data Probe is used in some of the mapping 375 database designs to "probe" or request a Map-Reply from an ETR; in 376 other cases, Map-Requests are used. See each mapping database 377 design for details. 379 4. Basic Overview 381 One key concept of LISP is that end-systems (hosts) operate the same 382 way they do today. The IP addresses that hosts use for tracking 383 sockets, connections, and for sending and receiving packets do not 384 change. In LISP terminology, these IP addresses are called Endpoint 385 Identifiers (EIDs). 387 Routers continue to forward packets based on IP destination 388 addresses. These addresses are referred to as Routing Locators 389 (RLOCs). Most routers along a path between two hosts will not 390 change; they continue to perform routing/forwarding lookups on 391 addresses (RLOCs) in the IP header. 393 This design introduces "Tunnel Routers", which prepend LISP headers 394 on host-originated packets and strip them prior to final delivery to 395 their destination. The IP addresses in this "outer header" are 396 RLOCs. During end-to-end packet exchange between two Internet hosts, 397 an ITR prepends a new LISP header to each packet and an egress tunnel 398 router strips the new header. The ITR performs EID-to-RLOC lookups 399 to determine the routing path to the the ETR, which has the RLOC as 400 one of its IP addresses. 402 Some basic rules governing LISP are: 404 o End-systems (hosts) only send to addresses which are EIDs. They 405 don't know addresses are EIDs versus RLOCs but assume packets get 406 to LISP routers, which in turn, deliver packets to the destination 407 the end-system has specified. 409 o EIDs are always IP addresses assigned to hosts. 411 o LISP routers mostly deal with Routing Locator addresses. See 412 details later in Section 4.1 to clarify what is meant by "mostly". 414 o RLOCs are always IP addresses assigned to routers; preferably, 415 topologically-oriented addresses from provider CIDR blocks. 417 o When a router originates packets it may use as a source address 418 either an EID or RLOC. When acting as a host (e.g. when 419 terminating a transport session such as SSH, TELNET, or SNMP), it 420 may use an EID that is explicitly assigned for that purpose. An 421 EID that identifies the router as a host MUST NOT be used as an 422 RLOC. Keep in mind that an EID is only routable within the scope 423 of a site. A typical BGP configuration might demonstrate this 424 "hybrid" EID/RLOC usage where a router could use its "host-like" 425 EID to terminate iBGP sessions to other routers in a site while at 426 the same time using RLOCs to terminate eBGP sessions to routers 427 outside the site. 429 o EIDs are not expected to be usable for global end-to-end 430 communication in the absence of an EID-to-RLOC mapping operation. 431 They are expected to be used locally for intra-site communication. 433 o EID prefixes are likely to be hierarchically assigned in a manner 434 which is optimized for administrative convenience and to 435 facilitate scaling of the EID-to-RLOC mapping database. The 436 hierarchy is based on a address allocation hierarchy which is not 437 dependent on the network topology. 439 o EIDs may also be structured (subnetted) in a manner suitable for 440 local routing within an autonomous system. 442 An additional LISP header may be pre-pended to packets by a transit 443 router (i.e. TE-ITR) when re-routing of the end-to-end path for a 444 packet is desired. An obvious instance of this would be an ISP 445 router that needs to perform traffic engineering for packets in flow 446 through its network. In such a situation, termed Recursive 447 Tunneling, an ISP transit acts as an additional ingress tunnel router 448 and the RLOC it uses for the new prepended header would be either an 449 TE-ETR within the ISP (along intra-ISP traffic engineered path) or in 450 an TE-ETR within another ISP (an inter-ISP traffic engineered path, 451 where an agreement to build such a path exists). 453 This specification mandates that no more than two LISP headers get 454 prepended to a packet. This avoids excessive packet overhead as well 455 as possible encapsulation loops. It is believed two headers is 456 sufficient, where the first prepended header is used at a site for 457 Locator/ID separation and second prepended header is used inside a 458 service provider for Traffic Engineering purposes. 460 Tunnel Routers can be placed fairly flexibly in a multi-AS topology. 461 For example, the ITR for a particular end-to-end packet exchange 462 might be the first-hop or default router within a site for the source 463 host. Similarly, the egress tunnel router might be the last-hop 464 router directly-connected to the destination host. Another example, 465 perhaps for a VPN service out-sourced to an ISP by a site, the ITR 466 could be the site's border router at the service provider attachment 467 point. Mixing and matching of site-operated, ISP-operated, and other 468 tunnel routers is allowed for maximum flexibility. See Section 8 for 469 more details. 471 4.1. Packet Flow Sequence 473 This section provides an example of the unicast packet flow with the 474 following parameters: 476 o Source host "host1.abc.com" is sending a packet to 477 "host2.xyz.com", exactly what host1 would do if the site was not 478 using LISP. 480 o Each site is multi-homed, so each tunnel router has an address 481 (RLOC) assigned from each of the site's attached service provider 482 address blocks. 484 o The ITR and ETR are directly connected to the source and 485 destination, respectively. 487 Client host1.abc.com wants to communicate with server host2.xyz.com: 489 1. host1.abc.com wants to open a TCP connection to host2.xyz.com. 490 It does a DNS lookup on host2.xyz.com. An A/AAAA record is 491 returned. This address is used as the destination EID and the 492 locally-assigned address of host1.abc.com is used as the source 493 EID. An IP/IPv6 packet is built using the EIDs in the IP/IPv6 494 header and sent to the default router. 496 2. The default router is configured as an ITR. The ITR must be able 497 to map the EID destination to an RLOC of the ETR at the 498 destination site. The ITR prepends a LISP header to the packet, 499 with one of its RLOCs as the source IP/IPv6 address. The 500 destination EID from the original packet header is used as the 501 destination IP/IPv6 in the prepended LISP header. Subsequent 502 packets will be sent using the same LISP header until EID-to-RLOC 503 mapping is learned. 505 3. In LISP 1, the packet is routed through the Internet as it is 506 today. In LISP 1.5, the packet is routed on a different topology 507 which may have EID prefixes distributed and advertised in an 508 aggregatable fashion. In either case, the packet arrives at the 509 ETR. The router is configured to "punt" the packet to the 510 router's processor. See Section 7 for more details. 512 4. The LISP header is stripped so that the packet can be forwarded 513 by the router control plane. The router looks up the destination 514 EID in the router's EID-to-RLOC database (not the cache, but the 515 configured data structure of RLOCs). An EID-to-RLOC Map-Reply 516 message is originated by the egress router and is addressed to 517 the source RLOC from the LISP header of the original packet (this 518 is the ITR). The source RLOC in the IP header of the UDP message 519 is one of the ETR's RLOCs (one of the RLOCs that is embedded in 520 the UDP payload). 522 5. The ITR receives the UDP Map-Reply message, parses the message 523 (to check for format validity) and stores the EID-to-RLOC 524 information from the packet. This information is put in the 525 ITR's EID-to-RLOC mapping cache (this is the on-demand cache, the 526 cache where entries time out due to inactivity). 528 6. Subsequent packets from host1.abc.com to host2.xyz.com will have 529 a LISP header prepended by the ITR using the appropriate RLOC as 530 the LISP header destination address learned from the ETR. Note, 531 the packet may be sent to a different ETR than the one which 532 returned the UDP Map-Reply. 534 7. The ETR receives these packets directly (since the destination 535 address is one of its assigned IP addresses), strips the LISP 536 header and forwards the packets to the attached destination host. 538 In order to eliminate the need for a mapping lookup in the reverse 539 direction, an ETR MAY create a cache entry that maps the source EID 540 (inner header source IP address) to the source RLOC (outer header 541 source IP address) in a received LISP packet. Such a cache entry is 542 termed a "gleaned" mapping and only contains a single RLOC for the 543 EID in question. More complete information about additional RLOCs 544 SHOULD be verified by sending a LISP Map-Request for that EID. Both 545 ITR and the ETR may also influence the decision the other makes in 546 selecting an RLOC. See Section 6 for more details. 548 5. Tunneling Details 550 This section describes the LISP Data Message which defines the 551 tunneling header used to encapsulate IPv4 and IPv6 packets which 552 contain EID addresses. Even though the following formats illustrate 553 IPv4-in-IPv4 and IPv6-in-IPv6 encapsulations, the other 2 554 combinations are supported as well. 556 Since additional tunnel headers are prepended, the packet becomes 557 larger and in theory can exceed the MTU of any link traversed from 558 the ITR to the ETR. It is recommended, in IPv4 that packets do not 559 get fragmented as they are encapsulated by the ITR. Instead, the 560 packet is dropped and an ICMP Too Big message is returned to the 561 source. 563 Based on informal surveys of large ISP traffic patterns, it appears 564 that most transit paths can accommodate a path MTU of at least 4470 565 bytes. The exceptions, in terms of data rate, number of hosts 566 affected, or any other metric are expected to be vanishingly small. 568 To address MTU concerns, mainly raised on the RRG mailing list, the 569 LISP deployment process will include collecting data during its pilot 570 phase to either verify or refute the assumption about minimum 571 available MTU. If the assumption proves true and transit networks 572 with links limited to 1500 byte MTUs are corner cases, it would seem 573 more cost-effective to either upgrade or modify the equipment in 574 those transit networks to support larger MTUs or to use existing 575 mechanisms for accommodating packets that are too large. 577 For this reason, there is currently no plan for LISP to add an 578 additional, complex mechanism for implementing fragmentation and 579 reassembly in the face of limited-MTU transit links. If analysis 580 during LISP pilot deployment reveals that the assumption of 581 essentially ubiquitous, 4470+ byte transit path MTUs, is incorrect, 582 then LISP can be modified prior to protocol standardization to add 583 support for one of the proposed fragmentation and reassembly schemes. 584 Note that one simple scheme is detailed in Section 5.4. 586 5.1. LISP IPv4-in-IPv4 Header Format 588 0 1 2 3 589 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 1 590 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 591 / |Version| IHL |Type of Service| Total Length | 592 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 593 / | Identification |Flags| Fragment Offset | 594 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 595 OH | Time to Live | Protocol = 17 | Header Checksum | 596 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 597 \ | Source Routing Locator | 598 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 599 \ | Destination Routing Locator | 600 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 601 / | Source Port = xxxx | Dest Port = 4341 | 602 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 603 \ | UDP Length | UDP Checksum | 604 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 605 / |S|M| Locator Reach Bits | 606 LISP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 607 \ | Nonce | 608 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 609 / |Version| IHL |Type of Service| Total Length | 610 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 611 / | Identification |Flags| Fragment Offset | 612 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 613 IH | Time to Live | Protocol | Header Checksum | 614 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 615 \ | Source EID | 616 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 617 \ | Destination EID | 618 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 620 5.2. LISP IPv6-in-IPv6 Header Format 622 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 623 / |Version| Traffic Class | Flow Label | 624 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 625 / | Payload Length | Next Header=17| Hop Limit | 626 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 627 | | 628 O + + 629 u | | 630 t + Source Routing Locator + 631 e | | 632 r + + 633 | | 634 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 635 d | | 636 r + + 637 | | 638 \ + Destination Routing Locator + 639 \ | | 640 \ + + 641 \ | | 642 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 643 / | Source Port = xxxx | Dest Port = 4341 | 644 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 645 \ | UDP Length | UDP Checksum | 646 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 647 / |S|M| Locator Reach Bits | 648 LISP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 649 \ | Nonce | 650 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 651 / |Version| Traffic Class | Flow Label | 652 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 653 / | Payload Length | Next Header | Hop Limit | 654 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 655 | | 656 I + + 657 n | | 658 n + Source EID + 659 e | | 660 r + + 661 | | 662 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 663 d | | 664 r + + 665 | | 667 \ + Destination EID + 668 \ | | 669 \ + + 670 \ | | 671 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 673 5.3. Tunnel Header Field Descriptions 675 IH Header: is the inner header, preserved from the datagram received 676 from the originating host. The source and destination IP 677 addresses are EIDs. 679 OH Header: is the outer header prepended by an ITR. The address 680 fields contain RLOCs obtained from the ingress router's EID-to- 681 RLOC cache. The IP protocol number is "UDP (17)" from [RFC0768]. 683 UDP Header: contains a ITR selected source port when encapsulating a 684 packet. See Section 6.4 for details on the hash algorithm used 685 select a source port based on the 5-tuple of the inner header. 686 The destination port MUST be set to the well-known IANA assigned 687 port value 4341. 689 UDP Checksum: this field field MUST be transmitted as 0 and ignored 690 on receipt by the ETR. Note, even when the UDP checksum is 691 transmitted as 0 an intervening NAT device can recalculate the 692 checksum and rewrite the UDP checksum field to non-zero. For 693 performance reasons, the ETR MUST ignore the checksum and MUST not 694 do a checksum computation. 696 UDP Length: for an IPv4 encapsulated packet, the inner header Total 697 Length plus the UDP and LISP header lengths are used. For an IPv6 698 encapsulated packet, the inner header Payload Length plus the size 699 of the IPv6 header (40 bytes) plus the size of the UDP and LISP 700 headers are used. The UDP header length is 8 bytes. The LISP 701 header length is 8 bytes when no loc-reach-bit header extensions 702 are used. 704 S: this is the Solicit-Map-Request (SMR) bit. See section 705 Section 6.5.2 for details. 707 LISP Locator Reach Bits: in the LISP header are set by an ITR to 708 indicate to an ETR the reachability of the Locators in the source 709 site. Each RLOC in a Map-Reply is assigned an ordinal value from 710 0 to n-1 (when there are n RLOCs in a mapping entry). The Locator 711 Reach Bits are numbered from 0 to n-1 from the right significant 712 bit of the 30-bit field. When a bit is set to 1, the ITR is 713 indicating to the ETR the RLOC associated with the bit ordinal is 714 reachable. See Section 6.3 for details on how an ITR can 715 determine other ITRs at the site are reachable. When a site has 716 multiple EID-prefixes which result in multiple mappings (where 717 each could have a different locator-set), the Locator Reach Bits 718 setting in an encapsulated packet MUST reflect the mapping for the 719 EID-prefix that the inner-header source EID address matches. When 720 the M bit is set, an additional 32-bit locator reachability field 721 follows, which contains an M-bit field and 31 locator reachability 722 bits. The M-bit may be set for further extension (and so on). 723 This extension mechanism allows an EID to be mapped to an 724 arbitrary number of RLOCs, subject only to the maximum number of 725 32-bit fields that can fit into the response packet. For 726 practical purposes, a future version of this specification will 727 likely set a limit on the number of these fields. 729 LISP Nonce: is a 32-bit value that is randomly generated by an ITR. 730 It is used to test route-returnability when xTRs exchange 731 encapsulated data packets with the SMR bit set, Data-Probe, Map- 732 Request, or Map-Reply messages. 734 When doing Recursive Tunneling: 736 o The OH header Time to Live field (or Hop Limit field, in case of 737 IPv6) MUST be copied from the IH header Time to Live field. 739 o The OH header Type of Service field (or the Traffic Class field, 740 in the case of IPv6) SHOULD be copied from the IH header Type of 741 Service field. 743 When doing Re-encapsulated Tunneling: 745 o The new OH header Time to Live field SHOULD be copied from the 746 stripped OH header Time to Live field. 748 o The new OH header Type of Service field SHOULD be copied from the 749 stripped OH header Type of Service field. 751 Copying the TTL serves two purposes: first, it preserves the distance 752 the host intended the packet to travel; second, and more importantly, 753 it provides for suppression of looping packets in the event there is 754 a loop of concatenated tunnels due to misconfiguration. 756 5.4. Dealing with Large Encapsulated Packets 758 In the event that the MTU issues mentioned above prove to be more 759 serious than expected, this section proposes a simple and stateless 760 mechanism to deal with large packets. The mechanism is described as 761 follows: 763 1. Define an architectural constant S for the maximum size of a 764 packet, in bytes, an ITR would receive from a source inside of 765 its site. 767 2. Define L to be the maximum size, in bytes, a packet of size S 768 would be after the ITR prepends the LISP header, UDP header, and 769 outer network layer header of size H. 771 3. Calculate: S + H = L. 773 When an ITR receives a packet from a site-facing interface and adds H 774 bytes worth of encapsulation to yield a packet size of L bytes, it 775 resolves the MTU issue by first splitting the original packet into 2 776 equal-sized fragments. A LISP header is then pre-pended to each 777 fragment. This will ensure that the new, encapsulated packets are of 778 size (S/2 + H), which is always below the effective tunnel MTU. 780 When an ETR receives encapsulated fragments, it treats them as two 781 individually encapsulated packets. It strips the LISP headers then 782 forwards each fragment to the destination host of the destination 783 site. The two fragments are reassembled at the destination host into 784 the single IP datagram that was originated by the source host. 786 This behavior is performed by the ITR when the source host originates 787 a packet with the DF field of the IP header is set to 0. When the DF 788 field of the IP header is set to 1, or the packet is an IPv6 packet 789 originated by the source host, the ITR will drop the packet when the 790 size is greater than L, and sends an ICMP Too Big message to the 791 source with a value of S, where S is (L - H). 793 This specification recommends that L be defined as 1500. 795 6. EID-to-RLOC Mapping 797 6.1. Control Plane Packet Format 799 When LISP 1 or LISP 1.5 is used, new UDP packet types encode the EID- 800 to-RLOC mappings: 802 0 1 2 3 803 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 1 804 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 805 |Version| IHL |Type of Service| Total Length | 806 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 807 | Identification |Flags| Fragment Offset | 808 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 809 | Time to Live | Protocol = 17 | Header Checksum | 810 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 811 | Source Routing Locator | 812 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 813 | Destination Routing Locator | 814 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 815 / | Source Port | Dest Port | 816 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 817 \ | UDP Length | UDP Checksum | 818 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 819 | | 820 | LISP Message | 821 | | 822 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 824 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 825 |Version| Traffic Class | Flow Label | 826 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 827 | Payload Length | Next Header=17| Hop Limit | 828 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 829 | | 830 + + 831 | | 832 + Source Routing Locator + 833 | | 834 + + 835 | | 836 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 837 | | 838 + + 839 | | 840 + Destination Routing Locator + 841 | | 842 + + 843 | | 844 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 845 / | Source Port | Dest Port | 846 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 847 \ | UDP Length | UDP Checksum | 848 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 849 | | 850 | LISP Message | 851 | | 852 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 854 The LISP UDP-based messages are the Map-Request and Map-Reply 855 messages. When a UDP Map-Request is sent, the UDP source port is 856 chosen by the sender and the destination UDP port number is set to 857 4342. When a UDP Map-Reply is sent, the source UDP port number is 858 set to 4342 and the destination UDP port number is copied from the 859 source port of either the Map-Request or the invoking data packet. 861 The UDP Length field will reflect the length of the UDP header and 862 the LISP Message payload. 864 The UDP Checksum is computed and set to non-zero for Map-Request and 865 Map-Reply messages. It MUST be checked on receipt and if the 866 checksum fails, the packet MUST be dropped. 868 LISP-CONS [CONS] use TCP to send LISP control messages. The format 869 of control messages includes the UDP header so the checksum and 870 length fields can be used to protect and delimit message boundaries. 872 This main LISP specification is the authoritative source for message 873 format definitions for the Map-Request and Map-Reply messages. 875 6.1.1. LISP Packet Type Allocations 877 This section will be the authoritative source for allocating LISP 878 Type values. Current allocations are: 880 Reserved: 0 b'0000' 881 LISP Map-Request: 1 b'0001' 882 LISP Map-Reply: 2 b'0010' 883 LISP-CONS Open Message: 8 b'1000' 884 LISP-CONS Push-Add Message: 9 b'1001' 885 LISP-CONS Push-Delete Message: 10 b'1010' 886 LISP-CONS Unreachable Message 11 b'1011' 888 6.1.2. Map-Request Message Format 890 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 891 |S|M| Locator Reach Bits | 892 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 893 | Nonce | 894 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 895 |Type=1 |A|R| Reserved | Record Count | 896 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 897 | Source-EID-AFI | ITR-AFI | 898 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 899 | Source EID Address ... | 900 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 901 | Originating ITR RLOC Address ... | 902 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 903 / | Reserved | EID mask-len | EID-prefix-AFI | 904 Rec < +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 905 \ | EID-prefix ... | 906 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 907 | Map-Reply Record ... | 908 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 909 | Mapping Protocol Data | 910 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 912 Packet field descriptions: 914 S: This is the SMR bit. See Section 6.5.2 for details. 916 Locator Reach Bits: Refer to Section 5.3. 918 Nonce: A 4-byte random value created by the sender of the Map- 919 Request. 921 Type: 1 (Map-Request) 923 A: This is an authoritative bit, which is set to 0 for UDP-based Map- 924 Requests sent by an ITR. See other control-specific documents 925 [CONS] for TCP-based Map-Requests. 927 R: When set, it indicates a Map-Reply Record segment is included in 928 the Map-Request. 930 Reserved: Set to 0 on transmission and ignored on receipt. 932 Record Count: The number of records in this request message. A 933 record is comprised of the portion of the packet is labeled 'Rec" 934 above and occurs the number of times equal to Record count. 936 Source-EID-AFI: Address family of the "Source EID Address" field. 938 ITR-AFI: Address family of the "Originating ITR RLOC Address" field. 940 Source EID Address: This is the EID of the source host which 941 originated the packet which is invoking this Map-Request. 943 Originating ITR RLOC Address: Used to give the ETR the option to 944 return a Map-Reply in the address-family of this locator. 946 EID mask-len: Mask length for EID prefix. 948 EID-AFI: Address family of EID-prefix according to [RFC2434] 950 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 951 address-family. 953 Map-Reply Record: When the R bit is set, this field is the size of 954 the "Record" field in the Map-Reply format. This Map-Reply record 955 contains the EID-to-RLOC mapping entry associated with the Source 956 EID. This allows the ETR which will receive this Map-Request to 957 cache the data if it chooses to do so. 959 Mapping Protocol Data: See [CONS] or [ALT] for details. 961 6.1.3. EID-to-RLOC UDP Map-Request Message 963 A Map-Request is sent from an ITR when it needs a mapping for an EID, 964 wants to test an RLOC for reachability, or wants to refresh a mapping 965 before TTL expiration. This is performed by using the RLOC as the 966 destination address for Map-Request message with a randomly allocated 967 source UDP port number and the well-known destination port number 968 4342. A successful Map-Reply updates the cached set of RLOCs 969 associated with the EID prefix range. 971 Map-Requests MUST be rate-limited. It is recommended that a Map- 972 Request for the same EID-prefix be sent no more than once per second. 974 6.1.4. Map-Reply Message Format 976 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 977 |x|M| Locator Reach Bits | 978 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 979 | Nonce | 980 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 981 |Type=2 | Reserved | Record Count | 982 +----> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 983 | | Record TTL | 984 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 985 R | Locator Count | EID mask-len |A| Reserved | 986 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 987 c | Reserved | EID-AFI | 988 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 989 r | EID-prefix | 990 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 991 | /| Priority | Weight | M Priority | M Weight | 992 | / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 993 | Loc | Unused Flags |R| Loc-AFI | 994 | \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 995 | \| Locator | 996 +---> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 997 | Mapping Protocol Data | 998 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1000 Packet field descriptions: 1002 x: Set to 0 on transmission and ignored on receipt. 1004 Locator Reach Bits: Refer to Section 5.3. When there are multiple 1005 records in the Map-Reply message, this field is set to 0 and the 1006 R-bit for each Locator record within each mapping record is used 1007 to determine the locator reachability. 1009 Nonce: A 4-byte value set in a Data-Probe packet or a Map-Request 1010 that is echoed here in the Map-Reply. 1012 Type: 2 (Map-Reply) 1014 Reserved: Set to 0 on transmission and ignored on receipt. 1016 Record Count: The number of records in this reply message. A record 1017 is comprised of that portion of the packet labeled 'Record' above 1018 and occurs the number of times equal to Record count. If the 1019 record count is not 1 then the Locator Reach Bits field MUST be 0. 1021 Record TTL: The time in minutes the recipient of the Map-Reply will 1022 store the mapping. If the TTL is 0, the entry should be removed 1023 from the cache immediately. If the value is 0xffffffff, the 1024 recipient can decide locally how long to store the mapping. 1026 Locator Count: The number of Locator entries. A locator entry 1027 comprises what is labeled above as 'Loc'. The locator count can 1028 be 0 indicating there are no locators for the EID-prefix. 1030 EID mask-len: Mask length for EID prefix. 1032 A: The Authoritative bit, when sent by a UDP-based message is always 1033 set by the ETR. See [CONS] for TCP-based Map-Replies. 1035 EID-AFI: Address family of EID-prefix according to [RFC2434]. 1037 EID-prefix: 4 bytes if an IPv4 address-family, 16 bytes if an IPv6 1038 address-family. 1040 Priority: each RLOC is assigned a unicast priority. Lower values 1041 are more preferable. When multiple RLOCs have the same priority, 1042 they may be used in a load-split fashion. A value of 255 means 1043 the RLOC MUST NOT be used for unicast forwarding. 1045 Weight: when priorities are the same for multiple RLOCs, the weight 1046 indicates how to balance unicast traffic between them. Weight is 1047 encoded as a percentage of total unicast packets that match the 1048 mapping entry. If a non-zero weight value is used for any RLOC, 1049 then all RLOCs must use a non-zero weight value and then the sum 1050 of all weight values MUST equal 100. If a zero value is used for 1051 any RLOC weight, then all weights MUST be zero and the receiver of 1052 the Map-Reply will decide how to load-split traffic. See 1053 Section 6.4 for a suggested hash algorithm to distribute load 1054 across locators with same priority and equal weight values. When 1055 a single RLOC exists in a mapping entry, the weight value MUST be 1056 set to 100 and ignored on receipt. 1058 M Priority: each RLOC is assigned a multicast priority used by an 1059 ETR in a receiver multicast site to select an ITR in a source 1060 multicast site for building multicast distribution trees. A value 1061 of 255 means the RLOC MUST NOT be used for joining a multicast 1062 distribution tree. 1064 M Weight: when priorities are the same for multiple RLOCs, the 1065 weight indicates how to balance building multicast distribution 1066 trees across multiple ITRs. The weight is encoded as a percentage 1067 of total number of trees build to the source site identified by 1068 the EID-prefix. If a non-zero weight value is used for any RLOC, 1069 then all RLOCs must use a non-zero weight value and then the sum 1070 of all weight values MUST equal 100. If a zero value is used for 1071 any RLOC weight, then all weights MUST be zero and the receiver of 1072 the Map-Reply will decide how to distribute multicast state across 1073 ITRs. 1075 Unused Flags: set to 0 when sending and ignored on receipt. 1077 R: when this bit is set, the locator is known to be reachable from 1078 the Map-Reply sender's perspective. When there is a single 1079 mapping record in the message, the R-bit for each locator must 1080 have a consistent setting with the bitfield setting of the 'Loc 1081 Reach Bits' field in the early part of the header. When there are 1082 multiple mapping records in the message, the 'Loc Reach Bits' 1083 field is set to 0. 1085 Locator: an IPv4 or IPv6 address (as encoded by the 'Loc-AFI' field) 1086 assigned to an ETR or router acting as a proxy replier for the 1087 EID-prefix. Note that the destination RLOC address MAY be an 1088 anycast address. A source RLOC can be an anycast address as well. 1089 The source or destination RLOC MUST NOT be the broadcast address 1090 (255.255.255.255 or any subnet broadcast address known to the 1091 router), and MUST NOT be a link-local multicast address. The 1092 source RLOC MUST NOT be a multicast address. The destination RLOC 1093 SHOULD be a multicast address if it is being mapped from a 1094 multicast destination EID. 1096 Mapping Protocol Data: See [CONS] or [ALT] for details. 1098 6.1.5. EID-to-RLOC UDP Map-Reply Message 1100 When a Data Probe packet or a Map-Request triggers a Map-Reply to be 1101 sent, the RLOCs associated with the EID-prefix matched by the EID in 1102 the original packet destination IP address field will be returned. 1103 The RLOCs in the Map-Reply are the globally-routable IP addresses of 1104 the ETR but are not necessarily reachable; separate testing of 1105 reachability is required. 1107 Note that a Map-Reply may contain different EID-prefix granularity 1108 (prefix + length) than the Map-Request which triggers it. This might 1109 occur if a Map-Request were for a prefix that had been returned by an 1110 earlier Map-Reply. In such a case, the requester updates its cache 1111 with the new prefix information and granularity. For example, a 1112 requester with two cached EID-prefixes that are covered by a Map- 1113 Reply containing one, less-specific prefix, replaces the entry with 1114 the less-specific EID-prefix. Note that the reverse, replacement of 1115 one less-specific prefix with multiple more-specific prefixes, can 1116 also occur but not by removing the less-specific prefix rather by 1117 adding the more-specific prefixes which during a lookup will override 1118 the less-specific prefix. 1120 Replies SHOULD be sent for an EID-prefix no more often than once per 1121 second to the same requesting router. For scalability, it is 1122 expected that aggregation of EID addresses into EID-prefixes will 1123 allow one Map-Reply to satisfy a mapping for the EID addresses in the 1124 prefix range thereby reducing the number of Map-Request messages. 1126 The addresses for a encapsulated data packets or Map-Request message 1127 are swapped and used for sending the Map-Reply. The UDP source and 1128 destination ports are swapped as well. That is, the source port in 1129 the UDP header for the Map-Reply is set to the well-known UDP port 1130 number 4342. 1132 6.2. Routing Locator Selection 1134 Both client-side and server-side may need control over the selection 1135 of RLOCs for conversations between them. This control is achieved by 1136 manipulating the Priority and Weight fields in EID-to-RLOC Map-Reply 1137 messages. Alternatively, RLOC information may be gleaned from 1138 received tunneled packets or EID-to-RLOC Map-Request messages. 1140 The following enumerates different scenarios for choosing RLOCs and 1141 the controls that are available: 1143 o Server-side returns one RLOC. Client-side can only use one RLOC. 1144 Server-side has complete control of the selection. 1146 o Server-side returns a list of RLOC where a subset of the list has 1147 the same best priority. Client can only use the subset list 1148 according to the weighting assigned by the server-side. In this 1149 case, the server-side controls both the subset list and load- 1150 splitting across its members. The client-side can use RLOCs 1151 outside of the subset list if it determines that the subset list 1152 is unreachable (unless RLOCs are set to a Priority of 255). Some 1153 sharing of control exists: the server-side determines the 1154 destination RLOC list and load distribution while the client-side 1155 has the option of using alternatives to this list if RLOCs in the 1156 list are unreachable. 1158 o Server-side sets weight of 0 for the RLOC subset list. In this 1159 case, the client-side can choose how the traffic load is spread 1160 across the subset list. Control is shared by the server-side 1161 determining the list and the client determining load distribution. 1162 Again, the client can use alternative RLOCs if the server-provided 1163 list of RLOCs are unreachable. 1165 o Either side (more likely on the server-side ETR) decides not to 1166 send a Map-Request. For example, if the server-side ETR does not 1167 send Map-Requests, it gleans RLOCs from the client-side ITR, 1168 giving the client-side ITR responsibility for bidirectional RLOC 1169 reachability and preferability. Server-side ETR gleaning of the 1170 client-side ITR RLOC is done by caching the inner header source 1171 EID and the outer header source RLOC of received packets. The 1172 client-side ITR controls how traffic is returned and can alternate 1173 using an outer header source RLOC, which then can be added to the 1174 list the server-side ETR uses to return traffic. Since no 1175 Priority or Weights are provided using this method, the server- 1176 side ETR must assume each client-side ITR RLOC uses the same best 1177 Priority with a Weight of zero. In addition, since EID-prefix 1178 encoding cannot be conveyed in data packets, the EID-to-RLOC cache 1179 on tunnel routers can grow to be very large. 1181 RLOCs that appear in EID-to-RLOC Map-Reply messages are considered 1182 reachable. The Map-Reply and the database mapping service does not 1183 provide any reachability status for Locators. This is done outside 1184 of the mapping service. See next section for details. 1186 6.3. Routing Locator Reachability 1188 There are 4 methods for determining when a Locator is either 1189 reachable or has become unreachable: 1191 1. Locator reachability is determined by an ETR by examining the 1192 Loc-Reach-Bits from a LISP header of a encapsulated data packet 1193 which is provided by an ITR when an ITR encapsulates data. 1195 2. Locator unreachability is determined by an ITR by receiving ICMP 1196 Network or Host Unreachable messages. 1198 3. ETR unreachability is determined when a host sends an ICMP Port 1199 Unreachable message. This occurs when an ITR may not use any 1200 methods of interworking. one which is describe in [INTERWORK] and 1201 the encapsulated data packet is received by a host at the 1202 destination non-LISP site. 1204 4. Locator reachability is determined by receiving a Map-Reply 1205 message from a ETR's Locator address in response to a previously 1206 sent Map-Request. 1208 When determining Locator reachability by examining the Loc-Reach-Bits 1209 from the LISP encapsulate data packet, an ETR will receive up to date 1210 status from the ITR closest to the Locators at the source site. The 1211 ITRs at the source site can determine reachability when running their 1212 IGP at the site. When the ITRs are deployed on CE routers, typically 1213 a default route is injected into the site's IGP from each of the 1214 ITRs. If an ITR goes down, the CE-PE link goes down, or the PE 1215 router goes down, the CE router withdraws the default route. This 1216 allows the other ITRs at the site to determine one of the Locators 1217 has gone unreachable. 1219 The Locators listed in a Map-Reply are numbered with ordinals 0 to 1220 n-1. The Loc-Reach-Bits in a LISP Data Message are numbered from 0 1221 to n-1 starting with the least significant bit numbered as 0. So, 1222 for example, if the ITR with locator listed as the 3rd Locator 1223 position in the Map-Reply goes down, all other ITRs at the site will 1224 have the 3rd bit from the right cleared (the bit that corresponds to 1225 ordinal 2). 1227 When an ETR decapsulates a packet, it will look for a change in the 1228 Loc-Reach-Bits value. When a bit goes from 1 to 0, the ETR will 1229 refrain from encapsulating packets to the Locator that has just gone 1230 unreachable. It can start using the Locator again when the bit that 1231 corresponds to the Locator goes from 0 to 1. Loc-Reach-Bits are 1232 associated with a locator-set per EID-prefix. Therefore, when a 1233 locator becomes unreachable, the loc-reach-bit that corresponds to 1234 that locator's position in the list returned by the last Map-Reply 1235 will be set to zero for that particular EID-prefix. 1237 When ITRs at the site are not deployed in CE routers, the IGP can 1238 still be used to determine the reachability of Locators provided they 1239 are injected a stub links into the IGP. This is typically done when 1240 a /32 address is configured on a loopback interface. 1242 When ITRs receive ICMP Network or Host Unreachable messages as a 1243 method to determine unreachability, they will refrain from using 1244 Locators which are described in Locator lists of Map-Replies. 1245 However, using this approach is unreliable because many network 1246 operators turn off generation of ICMP Unreachable messages. 1248 If an ITR does receive an ICMP Network or Host Unreachable message, 1249 it MAY originate its own ICMP Unreachable message destined for the 1250 host that originated the data packet the ITR encapsulated. 1252 Optionally, an ITR can send a Map-Request to a Locator and if a Map- 1253 Reply is returned, reachability of the Locator has been determined. 1254 Obviously, sending such probes increases the number of control 1255 messages originated by tunnel routers for active flows, so Locators 1256 are assumed to be reachable when they are advertised. 1258 This assumption does create a dependency: Locator unreachability is 1259 detected by the receipt of ICMP Host Unreachable messages. When an 1260 Locator has been determined to be unreachable, it is not used for 1261 active traffic; this is the same as if it were listed in a Map-Reply 1262 with priority 255. 1264 The ITR can test the reachability of the unreachable Locator by 1265 sending periodic Requests. Both Requests and Replies MUST be rate- 1266 limited. Locator reachability testing is never done with data 1267 packets since that increases the risk of packet loss for end-to-end 1268 sessions. 1270 6.4. Routing Locator Hashing 1272 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to 1273 a requesting ITR, the locator-set for the EID-prefix may contain 1274 different priority values for each locator address. When more than 1275 one best priority locator exists, the ITR can decide how to load 1276 share traffic against the corresponding locators. 1278 The following hash algorithm may be used by an ITR to select a 1279 locator for a packet destined to an EID for the EID-to-RLOC mapping: 1281 1. Either a source and destination address hash can be used or the 1282 traditional 5-tuple hash which includes the source and 1283 destination addresses, source and destination TCP, UDP, or SCTP 1284 port numbers and the IP protocol number field or IPv6 next- 1285 protocol fields of a packet a host originates from within a LISP 1286 site. When a packet is not a TCP, UDP, or SCTP packet, the 1287 source and destination addresses only from the header are used to 1288 compute the hash. 1290 2. Take the hash value and divide it by the number of locators 1291 stored in the locator-set for the EID-to-RLOC mapping. 1293 3. The remainder will be yield a value of 0 to "number of locators 1294 minus 1". Use the remainder to select the locator in the 1295 locator-set. 1297 Note that when a packet is LISP encapsulated, the source port number 1298 in the outer UDP header needs to be set. Selecting a random value 1299 allows core routers which are attached to Link Aggregation Groups 1300 (LAGs) to load-split the encapsulated packets across member links of 1301 such LAGs. Otherwise, core routers would see a single flow, since 1302 packets have a source address of the ITR, for packets which are 1303 originated by different EIDs at the source site. A suggested setting 1304 for the source port number computed by an ITR is a 5-tuple hash 1305 function on the inner header, as described above. 1307 6.5. Changing the Contents of EID-to-RLOC Mappings 1309 Since the LISP architecture uses a caching scheme to retrieve and 1310 store EID-to-RLOC mappings, the only way an ITR can get a more up-to- 1311 date mapping is to re-request the mapping. However, the ITRs do not 1312 know when the mappings change and the ETRs do not keep track of who 1313 requested its mappings. For scalability reasons, we want to maintain 1314 this approach but need to provide a way for ETRs change their 1315 mappings and inform the sites that are currently communicating with 1316 the ETR site using such mappings. 1318 When a locator record is added to the end of a locator-set, it is 1319 easy to update mappings. We assume new mappings will maintain the 1320 same locator ordering as the old mapping but just have new locators 1321 appended to the end of the list. So some ITRs can have a new mapping 1322 while other ITRs have only an old mapping that is used until they 1323 time out. When an ITR has only an old mapping but detects bits set 1324 in the loc-reach-bits that correspond to locators beyond the list it 1325 has cached, it simply ignores them. 1327 When a locator record is removed from a locator-set, ITRs that have 1328 the mapping cached will not use the removed locator because the xTRs 1329 will set the loc-reach-bit to 0. So even if the locator is in the 1330 list, it will not be used. For new mapping requests, the xTRs can 1331 set the locator address to 0 as well as setting the corresponding 1332 loc-reach-bit to 0. This forces ITRs with old or new mappings to 1333 avoid using the removed locator. 1335 If many changes occur to a mapping over a long period of time, one 1336 will find empty record slots in the middle of the locator-set and new 1337 records appended to the locator-set. At some point, it would be 1338 useful to compact the locator-set so the loc-reach-bit settings can 1339 be efficiently packed. 1341 We propose here two approaches for locator-set compaction, one 1342 operational and the other a protocol mechanism. The operational 1343 approach uses a clock sweep method. The protocol approach uses the 1344 concept of Solicit-Map-Requests. 1346 6.5.1. Clock Sweep 1348 The clock sweep approach uses planning in advance and the use of 1349 count-down TTLs to time out mappings that have already been cached. 1350 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So 1351 there is a 24 hour window to time out old mappings. The following 1352 clock sweep procedure is used: 1354 1. 24 hours before a mapping change is to take effect, a network 1355 administrator configures the ETRs at a site to start the clock 1356 sweep window. 1358 2. During the clock sweep window, ETRs continue to send Map-Reply 1359 messages with the current (unchanged) mapping records. The TTL 1360 for these mappings is set to 1 hour. 1362 3. 24 hours later, all previous cache entries will have timed out, 1363 and any active cache entries will time out within 1 hour. During 1364 this 1 hour window the ETRs continue to send Map-Reply messages 1365 with the current (unchanged) mapping records with the TTL set to 1366 1 minute. 1368 4. At the end of the 1 hour window, the ETRs will send Map-Reply 1369 messages with the new (changed) mapping records. So any active 1370 caches can get the new mapping contents right away if not cached, 1371 or in 1 minute if they had the mapping cached. 1373 6.5.2. Solicit-Map-Request (SMR) 1375 Soliciting a Map-Request is a selective way for xTRs, at the site 1376 where mappings change, to control the rate they receive requests for 1377 Map-Reply messages. SMRs are also used to tell remote ITRs to update 1378 the mappings they have cached. 1380 Since the xTRs don't keep track of remote ITRs that have cached their 1381 mappings, they can not tell exactly who needs the new mapping 1382 entries. So an xTR will solicit Map-Requests from sites it is 1383 currently sending encapsulated data to, and only from those sites. 1384 The xTRs can locally decide the algorithm for how often and to how 1385 many sites it sends SMR messages. 1387 An SMR message is simply a bit set in an encapsulated data packet 1388 (and a Map-Request message). When an ETR at a remote site 1389 decapsulates a data packet that has the SMR bit set, it can tell that 1390 a new Map-Request message is being solicited. Both the xTR that 1391 sends the SMR message and the site that acts on the SMR message MUST 1392 be rate-limited. 1394 The following procedure shows how a SMR exchange occurs when a site 1395 is doing locator-set compaction for an EID-to-RLOC mapping: 1397 1. When the database mappings in an ETR change, the ITRs at the site 1398 begin to set the SMR bit in packets they encapsulate to the sites 1399 they communicate with. 1401 2. A remote xTR which decapsulates a packet with the SMR bit set 1402 will schedule sending a Map-Request message to the source locator 1403 address of the encapsulated packet. The nonce in the Map-Request 1404 is copied from the nonce in the encapsulated data packet that has 1405 the SMR bit set. 1407 3. The remote xTR retransmits the Map-Request slowly until it gets a 1408 Map-Reply while continuing to use the cached mapping. 1410 4. The ETRs at the site with the changed mapping will reply to the 1411 Map-Request with a Map-Reply message provided the Map-Request 1412 nonce matches the nonce from the SMR. The Map-Reply messages 1413 SHOULD be rate limited. This is important to avoid Map-Reply 1414 implosion. 1416 5. The ETRs, at the site with the changed mapping, records the fact 1417 that the site that sent the Map-Request has received the new 1418 mapping data in the mapping cache entry for the remote site so 1419 the loc-reach-bits are reflective of the new mapping for packets 1420 going to the remote site. The ETR then stops sending packets 1421 with the SMR-bit set. 1423 7. Router Performance Considerations 1425 LISP is designed to be very hardware-based forwarding friendly. By 1426 doing tunnel header prepending [RFC1955] and stripping instead of re- 1427 writing addresses, existing hardware can support the forwarding model 1428 with little or no modification. Where modifications are required, 1429 they should be limited to re-programming existing hardware rather 1430 than requiring expensive design changes to hard-coded algorithms in 1431 silicon. 1433 A few implementation techniques can be used to incrementally 1434 implement LISP: 1436 o When a tunnel encapsulated packet is received by an ETR, the outer 1437 destination address may not be the address of the router. This 1438 makes it challenging for the control plane to get packets from the 1439 hardware. This may be mitigated by creating special FIB entries 1440 for the EID-prefixes of EIDs served by the ETR (those for which 1441 the router provides an RLOC translation). These FIB entries are 1442 marked with a flag indicating that control plane processing should 1443 be performed. The forwarding logic of testing for particular IP 1444 protocol number value is not necessary. No changes to existing, 1445 deployed hardware should be needed to support this. 1447 o On an ITR, prepending a new IP header is as simple as adding more 1448 bytes to a MAC rewrite string and prepending the string as part of 1449 the outgoing encapsulation procedure. Many routers that support 1450 GRE tunneling [RFC2784] or 6to4 tunneling [RFC3056] can already 1451 support this action. 1453 o When a received packet's outer destination address contains an EID 1454 which is not intended to be forwarded on the routable topology 1455 (i.e. LISP 1.5), the source address of a data packet or the 1456 router interface with which the source is associated (the 1457 interface from which it was received) can be associated with a VRF 1458 (Virtual Routing/Forwarding), in which a different (i.e. non- 1459 congruent) topology can be used to find EID-to-RLOC mappings. 1461 8. Deployment Scenarios 1463 This section will explore how and where ITRs and ETRs can be deployed 1464 and will discuss the pros and cons of each deployment scenario. 1465 There are two basic deployment trade-offs to consider: centralized 1466 versus distributed caches and flat, recursive, or re-encapsulating 1467 tunneling. 1469 When deciding on centralized versus distributed caching, the 1470 following issues should be considered: 1472 o Are the tunnel routers spread out so that the caches are spread 1473 across all the memories of each router? 1475 o Should management "touch points" be minimized by choosing few 1476 tunnel routers, just enough for redundancy? 1478 o In general, using more ITRs doesn't increase management load, 1479 since caches are built and stored dynamically. On the other hand, 1480 more ETRs does require more management since EID-prefix-to-RLOC 1481 mappings need to be explicitly configured. 1483 When deciding on flat, recursive, or re-encapsulation tunneling, the 1484 following issues should be considered: 1486 o Flat tunneling implements a single tunnel between source site and 1487 destination site. This generally offers better paths between 1488 sources and destinations with a single tunnel path. 1490 o Recursive tunneling is when tunneled traffic is again further 1491 encapsulated in another tunnel, either to implement VPNs or to 1492 perform Traffic Engineering. When doing VPN-based tunneling, the 1493 site has some control since the site is prepending a new tunnel 1494 header. In the case of TE-based tunneling, the site may have 1495 control if it is prepending a new tunnel header, but if the site's 1496 ISP is doing the TE, then the site has no control. Recursive 1497 tunneling generally will result in suboptimal paths but at the 1498 benefit of steering traffic to resource available parts of the 1499 network. 1501 o The technique of re-encapsulation ensures that packets only 1502 require one tunnel header. So if a packet needs to be rerouted, 1503 it is first decapsulated by the ETR and then re-encapsulated with 1504 a new tunnel header using a new RLOC. 1506 The next sub-sections will describe where tunnel routers can reside 1507 in the network. 1509 8.1. First-hop/Last-hop Tunnel Routers 1511 By locating tunnel routers close to hosts, the EID-prefix set is at 1512 the granularity of an IP subnet. So at the expense of more EID- 1513 prefix-to-RLOC sets for the site, the caches in each tunnel router 1514 can remain relatively small. But caches always depend on the number 1515 of non-aggregated EID destination flows active through these tunnel 1516 routers. 1518 With more tunnel routers doing encapsulation, the increase in control 1519 traffic grows as well: since the EID-granularity is greater, more 1520 Map-Requests and Map-Replies are traveling between more routers. 1522 The advantage of placing the caches and databases at these stub 1523 routers is that the products deployed in this part of the network 1524 have better price-memory ratios then their core router counterparts. 1525 Memory is typically less expensive in these devices and fewer routes 1526 are stored (only IGP routes). These devices tend to have excess 1527 capacity, both for forwarding and routing state. 1529 LISP functionality can also be deployed in edge switches. These 1530 devices generally have layer-2 ports facing hosts and layer-3 ports 1531 facing the Internet. Spare capacity is also often available in these 1532 devices as well. 1534 8.2. Border/Edge Tunnel Routers 1536 Using customer-edge (CE) routers for tunnel endpoints allows the EID 1537 space associated with a site to be reachable via a small set of RLOCs 1538 assigned to the CE routers for that site. 1540 This offers the opposite benefit of the first-hop/last-hop tunnel 1541 router scenario: the number of mapping entries and network management 1542 touch points are reduced, allowing better scaling. 1544 One disadvantage is that less of the network's resources are used to 1545 reach host endpoints thereby centralizing the point-of-failure domain 1546 and creating network choke points at the CE router. 1548 Note that more than one CE router at a site can be configured with 1549 the same IP address. In this case an RLOC is an anycast address. 1550 This allows resilience between the CE routers. That is, if a CE 1551 router fails, traffic is automatically routed to the other routers 1552 using the same anycast address. However, this comes with the 1553 disadvantage where the site cannot control the entrance point when 1554 the anycast route is advertised out from all border routers. 1556 8.3. ISP Provider-Edge (PE) Tunnel Routers 1558 Use of ISP PE routers as tunnel endpoint routers gives an ISP control 1559 over the location of the egress tunnel endpoints. That is, the ISP 1560 can decide if the tunnel endpoints are in the destination site (in 1561 either CE routers or last-hop routers within a site) or at other PE 1562 edges. The advantage of this case is that two or more tunnel headers 1563 can be avoided. By having the PE be the first router on the path to 1564 encapsulate, it can choose a TE path first, and the ETR can 1565 decapsulate and re-encapsulate for a tunnel to the destination end 1566 site. 1568 An obvious disadvantage is that the end site has no control over 1569 where its packets flow or the RLOCs used. 1571 As mentioned in earlier sections a combination of these scenarios is 1572 possible at the expense of extra packet header overhead, if both site 1573 and provider want control, then recursive or re-encapsulating tunnels 1574 are used. 1576 9. Traceroute Considerations 1578 When a source host in a LISP site initiates a traceroute to a 1579 destination host in another LISP site, it is highly desirable for it 1580 to see the entire path. Since packets are encapsulated from ITR to 1581 ETR, the hop across the tunnel could be viewed as a single hop. 1582 However, LISP traceroute will provide the entire path so the user can 1583 see 3 distinct segments of the path from a source LISP host to a 1584 destination LISP host: 1586 Segment 1 (in source LISP site based on EIDs): 1588 source-host ---> first-hop ... next-hop ---> ITR 1590 Segment 2 (in the core network based on RLOCs): 1592 ITR ---> next-hop ... next-hop ---> ETR 1594 Segment 3 (in the destination LISP site based on EIDs): 1596 ETR ---> next-hop ... last-hop ---> destination-host 1598 For segment 1 of the path, ICMP Time Exceeded messages are returned 1599 in the normal matter as they are today. The ITR performs a TTL 1600 decrement and test for 0 before encapsulating. So the ITR hop is 1601 seen by the traceroute source has an EID address (the address of 1602 site-facing interface). 1604 For segment 2 of the path, ICMP Time Exceeded messages are returned 1605 to the ITR because the TTL decrement to 0 is done on the outer 1606 header, so the destination of the ICMP messages are to the ITR RLOC 1607 address, the source source RLOC address of the encapsulated 1608 traceroute packet. The ITR looks inside of the ICMP payload to 1609 inspect the traceroute source so it can return the ICMP message to 1610 the address of the traceroute client as well as retaining the core 1611 router IP address in the ICMP message. This is so the traceroute 1612 client can display the core router address (the RLOC address) in the 1613 traceroute output. The ETR returns its RLOC address and responds to 1614 the TTL decrement to 0 like the previous core routers did. 1616 For segment 3, the next-hop router downstream from the ETR will be 1617 decrementing the TTL for the packet that was encapsulated, sent into 1618 the core, decapsulated by the ETR, and forwarded because it isn't the 1619 final destination. If the TTL is decremented to 0, any router on the 1620 path to the destination of the traceroute, including the next-hop 1621 router or destination, will send an ICMP Time Exceeded message to the 1622 source EID of the traceroute client. The ICMP message will be 1623 encapsulated by the local ITR and sent back to the ETR in the 1624 originated traceroute source site, where the packet will be delivered 1625 to the host. 1627 9.1. IPv6 Traceroute 1629 IPv6 traceroute follows the procedure described above since the 1630 entire traceroute data packet is included in ICMP Time Exceeded 1631 message payload. Therefore, only the ITR needs to pay special 1632 attention for forwarding ICMP messages back to the traceroute source. 1634 9.2. IPv4 Traceroute 1636 For IPv4 traceroute, we cannot follow the above procedure since IPv4 1637 ICMP Time Exceeded messages only include the invoking IP header and 8 1638 bytes that follow the IP header. Therefore, when a core router sends 1639 an IPv4 Time Exceeded message to an ITR, all the ITR has in the ICMP 1640 payload is the encapsulated header it prepended followed by a UDP 1641 header. The original invoking IP header, and therefore the identity 1642 of the traceroute source is lost. 1644 The solution we propose to solve this problem is to cache traceroute 1645 IPv4 headers in the ITR and to match them up with corresponding IPv4 1646 Time Exceeded messages received from core routers and the ETR. The 1647 ITR will use a circular buffer for caching the IPv4 and UDP headers 1648 of traceroute packets. It will select a 16-bit number as a key to 1649 find them later when the IPv4 Time Exceeded messages are received. 1650 When an ITR encapsulates an IPv4 traceroute packet, it will use the 1651 16-bit number as the UDP source port in the encapsulating header. 1652 When the ICMP Time Exceeded message is returned to the ITR, the UDP 1653 header of the encapsulating header is present in the ICMP payload 1654 thereby allowing the ITR to find the cached headers for the 1655 traceroute source. The ITR puts the cached headers in the payload 1656 and sends the ICMP Time Exceeded message to the traceroute source 1657 retaining the source address of the original ICMP Time Exceeded 1658 message (a core router or the ETR of the site of the traceroute 1659 destination). 1661 9.3. Traceroute using Mixed Locators 1663 When either an IPv4 traceroute or IPv6 traceroute is originated and 1664 the ITR encapsulates it in the other address family header, you 1665 cannot get all 3 segments of the traceroute. Segment 2 of the 1666 traceroute can not be conveyed to the traceroute source since it is 1667 expecting addresses from intermediate hops in the same address format 1668 for the type of traceroute it originated. Therefore, in this case, 1669 segment 2 will make the tunnel look like one hop. All the ITR has to 1670 do to make this work is to not copy the inner TTL to the outer, 1671 encapsulating header's TTL when a traceroute packet is encapsulated 1672 using an RLOC from a different address family. This will cause no 1673 TTL decrement to 0 to occur in core routers between the ITR and ETR. 1675 10. Mobility Considerations 1677 There are several kinds of mobility of which only some might be of 1678 concern to LISP. Essentially they are as follows. 1680 10.1. Site Mobility 1682 A site wishes to change its attachment points to the Internet, and 1683 its LISP Tunnel Routers will have new RLOCs when it changes upstream 1684 providers. Changes in EID-RLOC mappings for sites are expected to be 1685 handled by configuration, outside of the LISP protocol. 1687 10.2. Slow Endpoint Mobility 1689 An individual endpoint wishes to move, but is not concerned about 1690 maintaining session continuity. Renumbering is involved. LISP can 1691 help with the issues surrounding renumbering [RFC4192] [LISA96] by 1692 decoupling the address space used by a site from the address spaces 1693 used by its ISPs. [RFC4984] 1695 10.3. Fast Endpoint Mobility 1697 Fast endpoint mobility occurs when an endpoint moves relatively 1698 rapidly, changing its IP layer network attachment point. Maintenance 1699 of session continuity is a goal. This is where the Mobile IPv4 1700 [RFC3344bis] and Mobile IPv6 [RFC3775] [RFC4866] mechanisms are used, 1701 and primarily where interactions with LISP need to be explored. 1703 The problem is that as an endpoint moves, it may require changes to 1704 the mapping between its EID and a set of RLOCs for its new network 1705 location. When this is added to the overhead of mobile IP binding 1706 updates, some packets might be delayed or dropped. 1708 In IPv4 mobility, when an endpoint is away from home, packets to it 1709 are encapsulated and forwarded via a home agent which resides in the 1710 home area the endpoint's address belongs to. The home agent will 1711 encapsulate and forward packets either directly to the endpoint or to 1712 a foreign agent which resides where the endpoint has moved to. 1713 Packets from the endpoint may be sent directly to the correspondent 1714 node, may be sent via the foreign agent, or may be reverse-tunneled 1715 back to the home agent for delivery to the mobile node. As the 1716 mobile node's EID or available RLOC changes, LISP EID-to-RLOC 1717 mappings are required for communication between the mobile node and 1718 the home agent, whether via foreign agent or not. As a mobile 1719 endpoint changes networks, up to three LISP mapping changes may be 1720 required: 1722 o The mobile node moves from an old location to a new visited 1723 network location and notifies its home agent that it has done so. 1724 The Mobile IPv4 control packets the mobile node sends pass through 1725 one of the new visited network's ITRs, which needs a EID-RLOC 1726 mapping for the home agent. 1728 o The home agent might not have the EID-RLOC mappings for the mobile 1729 node's "care-of" address or its foreign agent in the new visited 1730 network, in which case it will need to acquire them. 1732 o When packets are sent directly to the correspondent node, it may 1733 be that no traffic has been sent from the new visited network to 1734 the correspondent node's network, and the new visited network's 1735 ITR will need to obtain an EID-RLOC mapping for the correspondent 1736 node's site. 1738 In addition, if the IPv4 endpoint is sending packets from the new 1739 visited network using its original EID, then LISP will need to 1740 perform a route-returnability check on the new EID-RLOC mapping for 1741 that EID. 1743 In IPv6 mobility, packets can flow directly between the mobile node 1744 and the correspondent node in either direction. The mobile node uses 1745 its "care-of" address (EID). In this case, the route-returnability 1746 check would not be needed but one more LISP mapping lookup may be 1747 required instead: 1749 o As above, three mapping changes may be needed for the mobile node 1750 to communicate with its home agent and to send packets to the 1751 correspondent node. 1753 o In addition, another mapping will be needed in the correspondent 1754 node's ITR, in order for the correspondent node to send packets to 1755 the mobile node's "care-of" address (EID) at the new network 1756 location. 1758 When both endpoints are mobile the number of potential mapping 1759 lookups increases accordingly. 1761 As a mobile node moves there are not only mobility state changes in 1762 the mobile node, correspondent node, and home agent, but also state 1763 changes in the ITRs and ETRs for at least some EID-prefixes. 1765 The goal is to support rapid adaptation, with little delay or packet 1766 loss for the entire system. Heuristics can be added to LISP to 1767 reduce the number of mapping changes required and to reduce the delay 1768 per mapping change. Also IP mobility can be modified to require 1769 fewer mapping changes. In order to increase overall system 1770 performance, there may be a need to reduce the optimization of one 1771 area in order to place fewer demands on another. 1773 In LISP, one possibility is to "glean" information. When a packet 1774 arrives, the ETR could examine the EID-RLOC mapping and use that 1775 mapping for all outgoing traffic to that EID. It can do this after 1776 performing a route-returnability check, to ensure that the new 1777 network location does have a internal route to that endpoint. 1778 However, this does not cover the case where an ITR (the node assigned 1779 the RLOC) at the mobile-node location has been compromised. 1781 Mobile IP packet exchange is designed for an environment in which all 1782 routing information is disseminated before packets can be forwarded. 1783 In order to allow the Internet to grow to support expected future 1784 use, we are moving to an environment where some information may have 1785 to be obtained after packets are in flight. Modifications to IP 1786 mobility should be considered in order to optimize the behavior of 1787 the overall system. Anything which decreases the number of new EID- 1788 RLOC mappings needed when a node moves, or maintains the validity of 1789 an EID-RLOC mapping for a longer time, is useful. 1791 10.4. Fast Network Mobility 1793 In addition to endpoints, a network can be mobile, possibly changing 1794 xTRs. A "network" can be as small as a single router and as large as 1795 a whole site. This is different from site mobility in that it is 1796 fast and possibly short-lived, but different from endpoint mobility 1797 in that a whole prefix is changing RLOCs. However, the mechanisms 1798 are the same and there is no new overhead in LISP. A map request for 1799 any endpoint will return a binding for the entire mobile prefix. 1801 If mobile networks become a more common occurrence, it may be useful 1802 to revisit the design of the mapping service and allow for dynamic 1803 updates of the database. 1805 The issue of interactions between mobility and LISP needs to be 1806 explored further. Specific improvements to the entire system will 1807 depend on the details of mapping mechanisms. Mapping mechanisms 1808 should be evaluated on how well they support session continuity for 1809 mobile nodes. 1811 11. Multicast Considerations 1813 A multicast group address, as defined in the original Internet 1814 architecture is an identifier of a grouping of topologically 1815 independent receiver host locations. The address encoding itself 1816 does not determine the location of the receiver(s). The multicast 1817 routing protocol, and the network-based state the protocol creates, 1818 determines where the receivers are located. 1820 In the context of LISP, a multicast group address is both an EID and 1821 a Routing Locator. Therefore, no specific semantic or action needs 1822 to be taken for a destination address, as it would appear in an IP 1823 header. Therefore, a group address that appears in an inner IP 1824 header built by a source host will be used as the destination EID. 1825 The outer IP header (the destination Routing Locator address), 1826 prepended by a LISP router, will use the same group address as the 1827 destination Routing Locator. 1829 Having said that, only the source EID and source Routing Locator 1830 needs to be dealt with. Therefore, an ITR merely needs to put its 1831 own IP address in the source Routing Locator field when prepending 1832 the outer IP header. This source Routing Locator address, like any 1833 other Routing Locator address MUST be globally routable. 1835 Therefore, an EID-to-RLOC mapping does not need to be performed by an 1836 ITR when a received data packet is a multicast data packet or when 1837 processing a source-specific Join (either by IGMPv3 or PIM). But the 1838 source Routing Locator is decided by the multicast routing protocol 1839 in a receiver site. That is, an EID to Routing Locator translation 1840 is done at control-time. 1842 Another approach is to have the ITR not encapsulate a multicast 1843 packet and allow the the host built packet to flow into the core even 1844 if the source address is allocated out of the EID namespace. If the 1845 RPF-Vector TLV [RPFV] is used by PIM in the core, then core routers 1846 can RPF to the ITR (the Locator address which is injected into core 1847 routing) rather than the host source address (the EID address which 1848 is not injected into core routing). 1850 To avoid any EID-based multicast state in the network core, the first 1851 approach is chosen for LISP-Multicast. Details for LISP-Multicast 1852 and Interworking with non-LISP sites is described in specification 1853 [MLISP]. 1855 12. Security Considerations 1857 It is believed that most of the security mechanisms will be part of 1858 the mapping database service when using control plane procedures for 1859 obtaining EID-to-RLOC mappings. For data plane triggered mappings, 1860 as described in this specification, protection is provided against 1861 ETR spoofing by using Return- Routability mechanisms evidenced by the 1862 use of a 4-byte Nonce field in the LISP encapsulation header. The 1863 nonce, coupled with the ITR accepting only solicited Map-Replies goes 1864 a long way toward providing decent authentication. 1866 LISP does not rely on a PKI infrastructure or a more heavy weight 1867 authentication system. These systems challenge the scalability of 1868 LISP which was a primary design goal. 1870 DoS attack prevention will depend on implementations rate-limiting 1871 Map-Requests and Map-Replies to the control plane as well as rate- 1872 limiting the number of data-triggered Map-Replies. 1874 13. Prototype Plans and Status 1876 The operator community has requested that the IETF take a practical 1877 approach to solving the scaling problems associated with global 1878 routing state growth. This document offers a simple solution which 1879 is intended for use in a pilot program to gain experience in working 1880 on this problem. 1882 The authors hope that publishing this specification will allow the 1883 rapid implementation of multiple vendor prototypes and deployment on 1884 a small scale. Doing this will help the community: 1886 o Decide whether a new EID-to-RLOC mapping database infrastructure 1887 is needed or if a simple, UDP-based, data-triggered approach is 1888 flexible and robust enough. 1890 o Experiment with provider-independent assignment of EIDs while at 1891 the same time decreasing the size of DFZ routing tables through 1892 the use of topologically-aligned, provider-based RLOCs. 1894 o Determine whether multiple levels of tunneling can be used by ISPs 1895 to achieve their Traffic Engineering goals while simultaneously 1896 removing the more specific routes currently injected into the 1897 global routing system for this purpose. 1899 o Experiment with mobility to determine if both acceptable 1900 convergence and session continuity properties can be scalably 1901 implemented to support both individual device roaming and site 1902 service provider changes. 1904 Here is a rough set of milestones: 1906 1. This draft will be the draft for interoperable implementations to 1907 code against. Interoperable implementations will be ready 1908 beginning of 2009. 1910 2. Continue pilot deployment using LISP-ALT as the database mapping 1911 mechanism. 1913 3. Continue prototyping and studying other database lookup schemes, 1914 be it DNS, DHTs, CONS, ALT, NERD, or other mechanisms. 1916 4. Implement the LISP Multicast draft [MLISP]. 1918 5. Research more on how policy affects what gets returned in a Map- 1919 Reply from an ETR. 1921 6. Continue to experiment with mixed locator-sets to understand how 1922 LISP can help the IPv4 to IPv6 transition. 1924 7. Add more robustness to locator reachability between LISP sites. 1926 As of this writing the following accomplishments have been achieved: 1928 1. A unit- and system-tested software switching implementation has 1929 been completed on cisco NX-OS for this draft for both IPv4 and 1930 IPv6 EIDs using a mixed locator-set of IPv4 and IPv6 locators. 1932 2. A unit- and system-tested software switching implementation on 1933 cisco NX-OS has been completed for draft [ALT]. 1935 3. A unit- and system-tested software switching implementation on 1936 cisco NX-OS has been completed for draft [INTERWORK]. Support 1937 for IPv4 translation is provided and PTR support for IPv4 and 1938 IPv6 is provided. 1940 4. The cisco NX-OS implementation supports an experimental mechanism 1941 for slow mobility. 1943 5. Dave Meyer, Vince Fuller, Darrel Lewis, Greg Shepherd, and Andrew 1944 Partan continue to test all the features described above on a 1945 dual-stack infrastructure. 1947 6. Darrel Lewis and Dave Meyer have deployed both LISP translation 1948 and LISP PTR support in the pilot network. Point your browser to 1949 http://www.lisp4.net to see translation happening in action so 1950 your non-LISP site can access a web server in a LISP site. 1952 7. Soon http://www.lisp6.net will work where your IPv6 LISP site can 1953 talk to a IPv6 web server in a LISP site by using mixed address- 1954 family based locators. 1956 8. An public domain implementation of LISP is underway. See 1957 [OPENLISP] for details. 1959 9. A cisco IOS implementation is underway which currently supports 1960 IPv4 encapsulation and decapsulation features. 1962 If interested in writing a LISP implementation, testing any of the 1963 LISP implementations, or want to be part of the LISP pilot program, 1964 please contact lisp@ietf.org. 1966 14. References 1968 14.1. Normative References 1970 [RFC0768] Postel, J., "User Datagram Protocol", STD 6, RFC 768, 1971 August 1980. 1973 [RFC1498] Saltzer, J., "On the Naming and Binding of Network 1974 Destinations", RFC 1498, August 1993. 1976 [RFC1955] Hinden, R., "New Scheme for Internet Routing and 1977 Addressing (ENCAPS) for IPNG", RFC 1955, June 1996. 1979 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1980 Requirement Levels", BCP 14, RFC 2119, March 1997. 1982 [RFC2434] Narten, T. and H. Alvestrand, "Guidelines for Writing an 1983 IANA Considerations Section in RFCs", BCP 26, RFC 2434, 1984 October 1998. 1986 [RFC2784] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. 1987 Traina, "Generic Routing Encapsulation (GRE)", RFC 2784, 1988 March 2000. 1990 [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains 1991 via IPv4 Clouds", RFC 3056, February 2001. 1993 [RFC3775] Johnson, D., Perkins, C., and J. Arkko, "Mobility Support 1994 in IPv6", RFC 3775, June 2004. 1996 [RFC4423] Moskowitz, R. and P. Nikander, "Host Identity Protocol 1997 (HIP) Architecture", RFC 4423, May 2006. 1999 [RFC4866] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route 2000 Optimization for Mobile IPv6", RFC 4866, May 2007. 2002 [RFC4984] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB 2003 Workshop on Routing and Addressing", RFC 4984, 2004 September 2007. 2006 14.2. Informative References 2008 [AFI] IANA, "Address Family Indicators (AFIs)", ADDRESS FAMILY 2009 NUMBERS http://www.iana.org/numbers.html, Febuary 2007. 2011 [ALT] Farinacci, D., Fuller, V., and D. Meyer, "LISP Alternative 2012 Topology (LISP-ALT)", draft-fuller-lisp-alt-03.txt (work 2013 in progress), October 2008. 2015 [APT] Jen, D., Meisel, M., Massey, D., Wang, L., Zhang, B., and 2016 L. Zhang, "APT: A Practical Transit Mapping Service", 2017 draft-jen-apt-00.txt (work in progress), July 2007. 2019 [CHIAPPA] Chiappa, J., "Endpoints and Endpoint names: A Proposed 2020 Enhancement to the Internet Architecture", Internet- 2021 Draft http://www.chiappa.net/~jnc/tech/endpoints.txt, 2022 1999. 2024 [CONS] Farinacci, D., Fuller, V., and D. Meyer, "LISP-CONS: A 2025 Content distribution Overlay Network Service for LISP", 2026 draft-meyer-lisp-cons-03.txt (work in progress), 2027 November 2007. 2029 [DHTs] Ratnasamy, S., Shenker, S., and I. Stoica, "Routing 2030 Algorithms for DHTs: Some Open Questions", PDF 2031 file http://www.cs.rice.edu/Conferences/IPTPS02/174.pdf. 2033 [GSE] "GSE - An Alternate Addressing Architecture for IPv6", 2034 draft-ietf-ipngwg-gseaddr-00.txt (work in progress), 1997. 2036 [INTERWORK] 2037 Lewis, D., Meyer, D., and D. Farinacci, "Interworking LISP 2038 with IPv4 and IPv6", draft-lewis-lisp-interworking-01.txt 2039 (work in progress), July 2008. 2041 [LISA96] Lear, E., Katinsky, J., Coffin, J., and D. Tharp, 2042 "Renumbering: Threat or Menace?", Usenix , September 1996. 2044 [LISP1] Farinacci, D., Oran, D., Fuller, V., and J. Schiller, 2045 "Locator/ID Separation Protocol (LISP1) [Routable ID 2046 Version]", 2047 Slide-set http://www.dinof.net/~dino/ietf/lisp1.ppt, 2048 October 2006. 2050 [LISP2] Farinacci, D., Oran, D., Fuller, V., and J. Schiller, 2051 "Locator/ID Separation Protocol (LISP2) [DNS-based 2052 Version]", 2053 Slide-set http://www.dinof.net/~dino/ietf/lisp2.ppt, 2054 November 2006. 2056 [LISPDHT] Mathy, L., Iannone, L., and O. Bonaventure, "LISP-DHT: 2057 Towards a DHT to map identifiers onto locators", 2058 draft-mathy-lisp-dht-00.txt (work in progress), 2059 February 2008. 2061 [MLISP] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, 2062 "LISP for Multicast Environments", 2063 draft-farinacci-lisp-multicast-01.txt (work in progress), 2064 November 2008. 2066 [NERD] Lear, E., "NERD: A Not-so-novel EID to RLOC Database", 2067 draft-lear-lisp-nerd-02.txt (work in progress), 2068 January 2008. 2070 [OPENLISP] 2071 Iannone, L. and O. Bonaventure, "OpenLISP Implementation 2072 Report", draft-iannone-openlisp-implementation-01.txt 2073 (work in progress), July 2008. 2075 [RADIR] Narten, T., "Routing and Addressing Problem Statement", 2076 draft-narten-radir-problem-statement-00.txt (work in 2077 progress), July 2007. 2079 [RFC3344bis] 2080 Perkins, C., "IP Mobility Support for IPv4, revised", 2081 draft-ietf-mip4-rfc3344bis-05 (work in progress), 2082 July 2007. 2084 [RFC4192] Baker, F., Lear, E., and R. Droms, "Procedures for 2085 Renumbering an IPv6 Network without a Flag Day", RFC 4192, 2086 September 2005. 2088 [RPFV] Wijnands, IJ., Boers, A., and E. Rosen, "The RPF Vector 2089 TLV", draft-ietf-pim-rpf-vector-03.txt (work in progress), 2090 October 2006. 2092 [RPMD] Handley, M., Huici, F., and A. Greenhalgh, "RPMD: Protocol 2093 for Routing Protocol Meta-data Dissemination", 2094 draft-handley-p2ppush-unpublished-2007726.txt (work in 2095 progress), July 2007. 2097 [SHIM6] Nordmark, E. and M. Bagnulo, "Level 3 multihoming shim 2098 protocol", draft-ietf-shim6-proto-06.txt (work in 2099 progress), October 2006. 2101 Appendix A. Acknowledgments 2103 The authors would like to gratefully acknowledge many people who have 2104 contributed discussion and ideas to the making of this proposal. 2105 They include Noel Chiappa, Jason Schiller, Lixia Zhang, Dorian Kim, 2106 Peter Schoenmaker, Darrel Lewis, Vijay Gill, Geoff Huston, David 2107 Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley, Chris 2108 Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler, Eliot 2109 Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi Iannone, 2110 Robin Whittle, Brian Carpenter, Joel Halpern, Roger Jorgensen, John 2111 Zwiebel, Ran Atkinson, Stig Venaas, Iljitsch van Beijnum, Roland 2112 Bless, Andrew Partan, Dana Blair, Bill Lynch, Marc Woolward, Damien 2113 Saucez, and Damian Lezama. 2115 In particular, we would like to thank Dave Meyer for his clever 2116 suggestion for the name "LISP". ;-) 2118 Authors' Addresses 2120 Dino Farinacci 2121 cisco Systems 2122 Tasman Drive 2123 San Jose, CA 95134 2124 USA 2126 Email: dino@cisco.com 2128 Vince Fuller 2129 cisco Systems 2130 Tasman Drive 2131 San Jose, CA 95134 2132 USA 2134 Email: vaf@cisco.com 2136 Dave Oran 2137 cisco Systems 2138 7 Ladyslipper Lane 2139 Acton, MA 2140 USA 2142 Email: oran@cisco.com 2144 Dave Meyer 2145 cisco Systems 2146 170 Tasman Drive 2147 San Jose, CA 2148 USA 2150 Email: dmm@cisco.com 2152 Scott Brim 2153 cisco Systems 2154 170 Tasman Drive 2155 San Jose, CA 2156 USA 2158 Email: sbrim@cisco.com 2160 Full Copyright Statement 2162 Copyright (C) The IETF Trust (2008). 2164 This document is subject to the rights, licenses and restrictions 2165 contained in BCP 78, and except as set forth therein, the authors 2166 retain all their rights. 2168 This document and the information contained herein are provided on an 2169 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 2170 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 2171 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 2172 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 2173 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 2174 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 2176 Intellectual Property 2178 The IETF takes no position regarding the validity or scope of any 2179 Intellectual Property Rights or other rights that might be claimed to 2180 pertain to the implementation or use of the technology described in 2181 this document or the extent to which any license under such rights 2182 might or might not be available; nor does it represent that it has 2183 made any independent effort to identify any such rights. Information 2184 on the procedures with respect to rights in RFC documents can be 2185 found in BCP 78 and BCP 79. 2187 Copies of IPR disclosures made to the IETF Secretariat and any 2188 assurances of licenses to be made available, or the result of an 2189 attempt made to obtain a general license or permission for the use of 2190 such proprietary rights by implementers or users of this 2191 specification can be obtained from the IETF on-line IPR repository at 2192 http://www.ietf.org/ipr. 2194 The IETF invites any interested party to bring to its attention any 2195 copyrights, patents or patent applications, or other proprietary 2196 rights that may cover technology that may be required to implement 2197 this standard. Please address the information to the IETF at 2198 ietf-ipr@ietf.org. 2200 Acknowledgment 2202 Funding for the RFC Editor function is provided by the IETF 2203 Administrative Support Activity (IASA).