1 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
2 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"><head profile="http://dublincore.org/documents/2008/08/04/dc-html/">
3 <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
4 <meta name="robots" content="index,follow">
5 <meta name="creator" content="rfcmarkup version 1.104">
6 <link rel="schema.DC" href="http://purl.org/dc/elements/1.1/">
7 <meta name="DC.Relation.Replaces" content="draft-farinacci-lisp">
8 <meta name="DC.Identifier" content="urn:ietf:rfc:6830">
9 <meta name="DC.Date.Issued" content="January, 2013">
10 <meta name="DC.Creator" content="Farinacci, Dino">
11 <meta name="DC.Creator" content="Lewis, Darrel">
12 <meta name="DC.Creator" content="Meyer, David">
13 <meta name="DC.Creator" content="Fuller, Vince">
14 <meta name="DC.Description.Abstract" content="This document describes a network-layer-based protocol that enables\nseparation of IP addresses into two new numbering spaces: Endpoint\nIdentifiers (EIDs) and Routing Locators (RLOCs). No changes are\nrequired to either host protocol stacks or to the "core" of the\nInternet infrastructure. The Locator/ID Separation Protocol (LISP) can\nbe incrementally deployed, without a "flag day", and offers Traffic\nEngineering, multihoming, and mobility benefits to early adopters,\neven when there are relatively few LISP-capable sites. Design and\ndevelopment of LISP was largely motivated by the problem statement\nproduced by the October 2006 IAB Routing and Addressing Workshop. This\ndocument defines an Experimental Protocol for the Internet community.">
15 <meta name="DC.Title" content="The Locator/ID Separation Protocol (LISP)">
17 <link rel="icon" href="http://tools.ietf.org/images/rfc.png" type="image/png">
18 <link rel="shortcut icon" href="http://tools.ietf.org/images/rfc.png" type="image/png">
19 <title>RFC 6830 - The Locator/ID Separation Protocol (LISP)</title>
21 <link rel="stylesheet" type="text/css" href="marking.css" />
22 <style type="text/css">
27 h1, h2, h3, h4, h5, h6, .h1, .h2, .h3, .h4, .h5, .h6 {
32 font-family: monospace;
43 font-family: monospace;
49 page-break-before: always;
52 text-decoration: none;
57 text-decoration: none;
61 font-family: monospace;
64 h1, h2, h3, h4, h5, h6 {
70 text-decoration: none;
77 .grey, .grey a:link, .grey a:visited {
81 background-color: #EEE;
84 border-top: 7px solid #EEE;
86 .bgwhite { background-color: white; }
87 .bgred { background-color: #F44; }
88 .bggrey { background-color: #666; }
89 .bgbrown { background-color: #840; }
90 .bgorange { background-color: #FA0; }
91 .bgyellow { background-color: #EE0; }
92 .bgmagenta{ background-color: #F4F; }
93 .bgblue { background-color: #66F; }
94 .bgcyan { background-color: #4DD; }
95 .bggreen { background-color: #4F4; }
97 .legend { font-size: 90%; }
98 .cplate { font-size: 70%; border: solid grey 1px; }
110 <script type="text/javascript"><!--
111 function addHeaderTags() {
112 var spans = document.getElementsByTagName("span");
113 for (var i=0; i < spans.length; i++) {
116 var level = elem.getAttribute("class");
117 if (level == "h1" || level == "h2" || level == "h3" || level == "h4" || level == "h5" || level == "h6") {
118 elem.innerHTML = "<"+level+">"+elem.innerHTML+"</"+level+">";
123 var legend_html = "Colour legend:<br /> <table> <tr><td>Unknown:</td> <td><span class='cplate bgwhite'> </span></td></tr> <tr><td>Draft:</td> <td><span class='cplate bgred'> </span></td></tr> <tr><td>Informational:</td> <td><span class='cplate bgorange'> </span></td></tr> <tr><td>Experimental:</td> <td><span class='cplate bgyellow'> </span></td></tr> <tr><td>Best Common Practice:</td> <td><span class='cplate bgmagenta'> </span></td></tr> <tr><td>Proposed Standard:</td> <td><span class='cplate bgblue'> </span></td></tr> <tr><td>Draft Standard (old designation):</td> <td><span class='cplate bgcyan'> </span></td></tr> <tr><td>Internet Standard:</td> <td><span class='cplate bggreen'> </span></td></tr> <tr><td>Historic:</td> <td><span class='cplate bggrey'> </span></td></tr> <tr><td>Obsolete:</td> <td><span class='cplate bgbrown'> </span></td></tr> </table>";
124 function showElem(id) {
125 var elem = document.getElementById(id);
126 elem.innerHTML = eval(id+"_html");
127 elem.style.visibility='visible';
129 function hideElem(id) {
130 var elem = document.getElementById(id);
131 elem.style.visibility='hidden';
137 <body onload="addHeaderTags()">
138 <div style="height: 13px;">
139 <div onmouseover="this.style.cursor='pointer';" onclick="showElem('legend');" onmouseout="hideElem('legend')" style="height: 6px; position: absolute; cursor: pointer;" class="pre noprint docinfo bgyellow" title="Click for colour legend."> </div>
140 <div id="legend" class="docinfo noprint pre legend" style="position:absolute; top: 4px; left: 4ex; visibility:hidden; background-color: white; padding: 4px 9px 5px 7px; border: solid #345 1px; " onmouseover="showElem('legend');" onmouseout="hideElem('legend');"></div>
142 <span class="pre noprint docinfo top">[<a href="http://tools.ietf.org/html/" title="Document search and retrieval page">Docs</a>] [<a href="http://tools.ietf.org/rfc/rfc6830.txt" title="Plaintext version of this document">txt</a>|<a href="http://tools.ietf.org/pdf/rfc6830" title="PDF version of this document">pdf</a>] [<a href="http://tools.ietf.org/html/draft-ietf-lisp" title="draft-ietf-lisp">draft-ietf-lisp</a>] [<a href="http://tools.ietf.org/rfcdiff?difftype=--hwdiff&url2=rfc6830" title="Inline diff (wdiff)">Diff1</a>] [<a href="http://tools.ietf.org/rfcdiff?url2=rfc6830" title="Side-by-side diff">Diff2</a>] </span><br>
143 <span class="pre noprint docinfo"> </span><br>
144 <span class="pre noprint docinfo"> EXPERIMENTAL</span><br>
145 <span class="pre noprint docinfo"> </span><br>
146 <pre>Internet Engineering Task Force (IETF) D. Farinacci
147 Request for Comments: 6830 Cisco Systems
148 Category: Experimental V. Fuller
156 <span class="h1"><h1>The Locator/ID Separation Protocol (LISP)</h1></span>
160 This document describes a network-layer-based protocol that enables
161 separation of IP addresses into two new numbering spaces: Endpoint
162 Identifiers (EIDs) and Routing Locators (RLOCs). No changes are
163 required to either host protocol stacks or to the "core" of the
164 Internet infrastructure. The Locator/ID Separation Protocol (LISP)
165 can be incrementally deployed, without a "flag day", and offers
166 Traffic Engineering, multihoming, and mobility benefits to early
167 adopters, even when there are relatively few LISP-capable sites.
169 Design and development of LISP was largely motivated by the problem
170 statement produced by the October 2006 IAB Routing and Addressing
175 This document is not an Internet Standards Track specification; it is
176 published for examination, experimental implementation, and
179 This document defines an Experimental Protocol for the Internet
180 community. This document is a product of the Internet Engineering
181 Task Force (IETF). It represents the consensus of the IETF
182 community. It has received public review and has been approved for
183 publication by the Internet Engineering Steering Group (IESG). Not
184 all documents approved by the IESG are a candidate for any level of
185 Internet Standard; see <a href="http://tools.ietf.org/html/rfc5741#section-2">Section 2 of RFC 5741</a>.
187 Information about the current status of this document, any errata,
188 and how to provide feedback on it may be obtained at
189 <a href="http://www.rfc-editor.org/info/rfc6830">http://www.rfc-editor.org/info/rfc6830</a>.
197 <span class="grey">Farinacci, et al. Experimental [Page 1]</span>
198 </pre><!--NewPage--><pre class="newpage"><a name="page-2" id="page-2" href="#page-2" class="invisible"> </a>
199 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
204 Copyright (c) 2013 IETF Trust and the persons identified as the
205 document authors. All rights reserved.
207 This document is subject to <a href="http://tools.ietf.org/html/bcp78">BCP 78</a> and the IETF Trust's Legal
208 Provisions Relating to IETF Documents
209 (<a href="http://trustee.ietf.org/license-info">http://trustee.ietf.org/license-info</a>) in effect on the date of
210 publication of this document. Please review these documents
211 carefully, as they describe your rights and restrictions with respect
212 to this document. Code Components extracted from this document must
213 include Simplified BSD License text as described in <a href="#section-4">Section 4</a>.e of
214 the Trust Legal Provisions and are provided without warranty as
215 described in the Simplified BSD License.
219 <a href="#section-1">1</a>. Introduction ....................................................<a href="#page-3">3</a>
220 <a href="#section-2">2</a>. Requirements Notation ...........................................<a href="#page-5">5</a>
221 <a href="#section-3">3</a>. Definition of Terms .............................................<a href="#page-5">5</a>
222 <a href="#section-4">4</a>. Basic Overview .................................................<a href="#page-10">10</a>
223 <a href="#section-4.1">4.1</a>. Packet Flow Sequence ......................................<a href="#page-13">13</a>
224 <a href="#section-5">5</a>. LISP Encapsulation Details .....................................<a href="#page-15">15</a>
225 <a href="#section-5.1">5.1</a>. LISP IPv4-in-IPv4 Header Format ...........................<a href="#page-16">16</a>
226 <a href="#section-5.2">5.2</a>. LISP IPv6-in-IPv6 Header Format ...........................<a href="#page-17">17</a>
227 <a href="#section-5.3">5.3</a>. Tunnel Header Field Descriptions ..........................<a href="#page-18">18</a>
228 <a href="#section-5.4">5.4</a>. Dealing with Large Encapsulated Packets ...................<a href="#page-22">22</a>
229 <a href="#section-5.4.1">5.4.1</a>. A Stateless Solution to MTU Handling ...............<a href="#page-22">22</a>
230 <a href="#section-5.4.2">5.4.2</a>. A Stateful Solution to MTU Handling ................<a href="#page-23">23</a>
231 <a href="#section-5.5">5.5</a>. Using Virtualization and Segmentation with LISP ...........<a href="#page-24">24</a>
232 <a href="#section-6">6</a>. EID-to-RLOC Mapping ............................................<a href="#page-25">25</a>
233 <a href="#section-6.1">6.1</a>. LISP IPv4 and IPv6 Control-Plane Packet Formats ...........<a href="#page-25">25</a>
234 <a href="#section-6.1.1">6.1.1</a>. LISP Packet Type Allocations .......................<a href="#page-27">27</a>
235 <a href="#section-6.1.2">6.1.2</a>. Map-Request Message Format .........................<a href="#page-27">27</a>
236 <a href="#section-6.1.3">6.1.3</a>. EID-to-RLOC UDP Map-Request Message ................<a href="#page-30">30</a>
237 <a href="#section-6.1.4">6.1.4</a>. Map-Reply Message Format ...........................<a href="#page-31">31</a>
238 <a href="#section-6.1.5">6.1.5</a>. EID-to-RLOC UDP Map-Reply Message ..................<a href="#page-35">35</a>
239 <a href="#section-6.1.6">6.1.6</a>. Map-Register Message Format ........................<a href="#page-37">37</a>
240 <a href="#section-6.1.7">6.1.7</a>. Map-Notify Message Format ..........................<a href="#page-39">39</a>
241 <a href="#section-6.1.8">6.1.8</a>. Encapsulated Control Message Format ................<a href="#page-41">41</a>
242 <a href="#section-6.2">6.2</a>. Routing Locator Selection .................................<a href="#page-42">42</a>
243 <a href="#section-6.3">6.3</a>. Routing Locator Reachability ..............................<a href="#page-44">44</a>
244 <a href="#section-6.3.1">6.3.1</a>. Echo Nonce Algorithm ...............................<a href="#page-46">46</a>
245 <a href="#section-6.3.2">6.3.2</a>. RLOC-Probing Algorithm .............................<a href="#page-48">48</a>
246 <a href="#section-6.4">6.4</a>. EID Reachability within a LISP Site .......................<a href="#page-49">49</a>
247 <a href="#section-6.5">6.5</a>. Routing Locator Hashing ...................................<a href="#page-49">49</a>
253 <span class="grey">Farinacci, et al. Experimental [Page 2]</span>
254 </pre><!--NewPage--><pre class="newpage"><a name="page-3" id="page-3" href="#page-3" class="invisible"> </a>
255 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
258 <a href="#section-6.6">6.6</a>. Changing the Contents of EID-to-RLOC Mappings .............<a href="#page-50">50</a>
259 <a href="#section-6.6.1">6.6.1</a>. Clock Sweep ........................................<a href="#page-51">51</a>
260 <a href="#section-6.6.2">6.6.2</a>. Solicit-Map-Request (SMR) ..........................<a href="#page-52">52</a>
261 <a href="#section-6.6.3">6.6.3</a>. Database Map-Versioning ............................<a href="#page-53">53</a>
262 <a href="#section-7">7</a>. Router Performance Considerations ..............................<a href="#page-54">54</a>
263 <a href="#section-8">8</a>. Deployment Scenarios ...........................................<a href="#page-55">55</a>
264 <a href="#section-8.1">8.1</a>. First-Hop/Last-Hop Tunnel Routers .........................<a href="#page-56">56</a>
265 <a href="#section-8.2">8.2</a>. Border/Edge Tunnel Routers ................................<a href="#page-56">56</a>
266 <a href="#section-8.3">8.3</a>. ISP Provider Edge (PE) Tunnel Routers .....................<a href="#page-57">57</a>
267 <a href="#section-8.4">8.4</a>. LISP Functionality with Conventional NATs .................<a href="#page-58">58</a>
268 <a href="#section-8.5">8.5</a>. Packets Egressing a LISP Site .............................<a href="#page-58">58</a>
269 <a href="#section-9">9</a>. Traceroute Considerations ......................................<a href="#page-58">58</a>
270 <a href="#section-9.1">9.1</a>. IPv6 Traceroute ...........................................<a href="#page-59">59</a>
271 <a href="#section-9.2">9.2</a>. IPv4 Traceroute ...........................................<a href="#page-60">60</a>
272 <a href="#section-9.3">9.3</a>. Traceroute Using Mixed Locators ...........................<a href="#page-60">60</a>
273 <a href="#section-10">10</a>. Mobility Considerations .......................................<a href="#page-61">61</a>
274 <a href="#section-10.1">10.1</a>. Site Mobility ............................................<a href="#page-61">61</a>
275 <a href="#section-10.2">10.2</a>. Slow Endpoint Mobility ...................................<a href="#page-61">61</a>
276 <a href="#section-10.3">10.3</a>. Fast Endpoint Mobility ...................................<a href="#page-61">61</a>
277 <a href="#section-10.4">10.4</a>. Fast Network Mobility ....................................<a href="#page-63">63</a>
278 <a href="#section-10.5">10.5</a>. LISP Mobile Node Mobility ................................<a href="#page-64">64</a>
279 <a href="#section-11">11</a>. Multicast Considerations ......................................<a href="#page-64">64</a>
280 <a href="#section-12">12</a>. Security Considerations .......................................<a href="#page-65">65</a>
281 <a href="#section-13">13</a>. Network Management Considerations .............................<a href="#page-67">67</a>
282 <a href="#section-14">14</a>. IANA Considerations ...........................................<a href="#page-67">67</a>
283 <a href="#section-14.1">14.1</a>. LISP ACT and Flag Fields .................................<a href="#page-67">67</a>
284 <a href="#section-14.2">14.2</a>. LISP Address Type Codes ..................................<a href="#page-68">68</a>
285 <a href="#section-14.3">14.3</a>. LISP UDP Port Numbers ....................................<a href="#page-68">68</a>
286 <a href="#section-14.4">14.4</a>. LISP Key ID Numbers ......................................<a href="#page-68">68</a>
287 <a href="#section-15">15</a>. Known Open Issues and Areas of Future Work ....................<a href="#page-68">68</a>
288 <a href="#section-16">16</a>. References ....................................................<a href="#page-70">70</a>
289 <a href="#section-16.1">16.1</a>. Normative References .....................................<a href="#page-70">70</a>
290 <a href="#section-16.2">16.2</a>. Informative References ...................................<a href="#page-71">71</a>
291 <a href="#appendix-A">Appendix A</a>. Acknowledgments .......................................<a href="#page-74">74</a>
293 <span class="h2"><h2><a class="selflink" name="section-1" href="#section-1">1</a>. Introduction</h2></span>
295 This document describes the Locator/Identifier Separation Protocol
296 (LISP), which provides a set of functions for routers to exchange
297 information used to map from Endpoint Identifiers (EIDs) that are not
298 globally routable to routable Routing Locators (RLOCs). It also
299 defines a mechanism for these LISP routers to encapsulate IP packets
300 addressed with EIDs for transmission across a network infrastructure
301 that uses RLOCs for routing and forwarding.
309 <span class="grey">Farinacci, et al. Experimental [Page 3]</span>
310 </pre><!--NewPage--><pre class="newpage"><a name="page-4" id="page-4" href="#page-4" class="invisible"> </a>
311 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
314 Creation of LISP was initially motivated by discussions during the
315 IAB-sponsored Routing and Addressing Workshop held in Amsterdam in
316 October 2006 (see [<a href="http://tools.ietf.org/html/rfc4984" title=""Report from the IAB Workshop on Routing and Addressing"">RFC4984</a>]). A key conclusion of the workshop was
317 that the Internet routing and addressing system was not scaling well
318 in the face of the explosive growth of new sites; one reason for this
319 poor scaling is the increasing number of multihomed sites and other
320 sites that cannot be addressed as part of topology-based or provider-
321 based aggregated prefixes. Additional work that more completely
322 describes the problem statement may be found in [<a href="#ref-RADIR" title=""On the Scalability of Internet Routing"">RADIR</a>].
324 A basic observation, made many years ago in early networking research
325 such as that documented in [<a href="#ref-CHIAPPA" title=""Endpoints and Endpoint names: A Proposed Enhancement to the Internet Architecture"">CHIAPPA</a>] and [<a href="http://tools.ietf.org/html/rfc4984" title=""Report from the IAB Workshop on Routing and Addressing"">RFC4984</a>], is that using a
326 single address field for both identifying a device and for
327 determining where it is topologically located in the network requires
328 optimization along two conflicting axes: for routing to be efficient,
329 the address must be assigned topologically; for collections of
330 devices to be easily and effectively managed, without the need for
331 renumbering in response to topological change (such as that caused by
332 adding or removing attachment points to the network or by mobility
333 events), the address must explicitly not be tied to the topology.
335 The approach that LISP takes to solving the routing scalability
336 problem is to replace IP addresses with two new types of numbers:
337 Routing Locators (RLOCs), which are topologically assigned to network
338 attachment points (and are therefore amenable to aggregation) and
339 used for routing and forwarding of packets through the network; and
340 Endpoint Identifiers (EIDs), which are assigned independently from
341 the network topology, are used for numbering devices, and are
342 aggregated along administrative boundaries. LISP then defines
343 functions for mapping between the two numbering spaces and for
344 encapsulating traffic originated by devices using non-routable EIDs
345 for transport across a network infrastructure that routes and
346 forwards using RLOCs. Both RLOCs and EIDs are syntactically
347 identical to IP addresses; it is the semantics of how they are used
350 This document describes the protocol that implements these functions.
351 The database that stores the mappings between EIDs and RLOCs is
352 explicitly a separate "module" to facilitate experimentation with a
353 variety of approaches. One database design that is being developed
354 for experimentation as part of the LISP working group work is
355 [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>]. Others that have been described include [<a href="#ref-CONS" title=""LISP-CONS: A Content distribution Overlay Network Service for LISP"">CONS</a>], [<a href="#ref-EMACS" title=""EID Mappings Multicast Across Cooperating Systems for LISP"">EMACS</a>],
356 and [<a href="http://tools.ietf.org/html/rfc6837" title=""NERD: A Not-so-novel Endpoint ID (EID) to Routing Locator (RLOC) Database"">RFC6837</a>]. Finally, [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>] documents a general-purpose
357 service interface for accessing a mapping database; this interface is
358 intended to make the mapping database modular so that different
359 approaches can be tried without the need to modify installed LISP-
360 capable devices in LISP sites.
365 <span class="grey">Farinacci, et al. Experimental [Page 4]</span>
366 </pre><!--NewPage--><pre class="newpage"><a name="page-5" id="page-5" href="#page-5" class="invisible"> </a>
367 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
370 This experimental specification has areas that require additional
371 experience and measurement. It is NOT RECOMMENDED for deployment
372 beyond experimental situations. Results of experimentation may lead
373 to modifications and enhancements of protocol mechanisms defined in
374 this document. See <a href="#section-15">Section 15</a> for specific, known issues that are in
375 need of further work during development, implementation, and
378 An examination of the implications of LISP on Internet traffic,
379 applications, routers, and security is for future study. This
380 analysis will explain what role LISP can play in scalable routing and
381 will also look at scalability and levels of state required for
382 encapsulation, decapsulation, liveness, and so on.
384 <span class="h2"><h2><a class="selflink" name="section-2" href="#section-2">2</a>. Requirements Notation</h2></span>
386 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
387 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
388 document are to be interpreted as described in [<a href="http://tools.ietf.org/html/rfc2119" title=""Key words for use in RFCs to Indicate Requirement Levels"">RFC2119</a>].
390 <span class="h2"><h2><a class="selflink" name="section-3" href="#section-3">3</a>. Definition of Terms</h2></span>
392 Provider-Independent (PI) Addresses: PI addresses are an address
393 block assigned from a pool where blocks are not associated with
394 any particular location in the network (e.g., from a particular
395 service provider) and are therefore not topologically aggregatable
396 in the routing system.
398 Provider-Assigned (PA) Addresses: PA addresses are an address block
399 assigned to a site by each service provider to which a site
400 connects. Typically, each block is a sub-block of a service
401 provider Classless Inter-Domain Routing (CIDR) [<a href="http://tools.ietf.org/html/rfc4632" title=""Classless Inter-domain Routing (CIDR): The Internet Address Assignment and Aggregation Plan"">RFC4632</a>] block and
402 is aggregated into the larger block before being advertised into
403 the global Internet. Traditionally, IP multihoming has been
404 implemented by each multihomed site acquiring its own globally
405 visible prefix. LISP uses only topologically assigned and
406 aggregatable address blocks for RLOCs, eliminating this
407 demonstrably non-scalable practice.
409 Routing Locator (RLOC): An RLOC is an IPv4 [<a href="http://tools.ietf.org/html/rfc0791" title=""Internet Protocol"">RFC0791</a>] or IPv6
410 [<a href="http://tools.ietf.org/html/rfc2460" title=""Internet Protocol, Version 6 (IPv6) Specification"">RFC2460</a>] address of an Egress Tunnel Router (ETR). An RLOC is
411 the output of an EID-to-RLOC mapping lookup. An EID maps to one
412 or more RLOCs. Typically, RLOCs are numbered from topologically
413 aggregatable blocks that are assigned to a site at each point to
414 which it attaches to the global Internet; where the topology is
415 defined by the connectivity of provider networks, RLOCs can be
416 thought of as PA addresses. Multiple RLOCs can be assigned to the
417 same ETR device or to multiple ETR devices at a site.
421 <span class="grey">Farinacci, et al. Experimental [Page 5]</span>
422 </pre><!--NewPage--><pre class="newpage"><a name="page-6" id="page-6" href="#page-6" class="invisible"> </a>
423 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
426 Endpoint ID (EID): An EID is a 32-bit (for IPv4) or 128-bit (for
427 IPv6) value used in the source and destination address fields of
428 the first (most inner) LISP header of a packet. The host obtains
429 a destination EID the same way it obtains a destination address
430 today, for example, through a Domain Name System (DNS) [<a href="http://tools.ietf.org/html/rfc1034" title=""Domain names - concepts and facilities"">RFC1034</a>]
431 lookup or Session Initiation Protocol (SIP) [<a href="http://tools.ietf.org/html/rfc3261" title=""SIP: Session Initiation Protocol"">RFC3261</a>] exchange.
432 The source EID is obtained via existing mechanisms used to set a
433 host's "local" IP address. An EID used on the public Internet
434 must have the same properties as any other IP address used in that
435 manner; this means, among other things, that it must be globally
436 unique. An EID is allocated to a host from an EID-Prefix block
437 associated with the site where the host is located. An EID can be
438 used by a host to refer to other hosts. EIDs MUST NOT be used as
439 LISP RLOCs. Note that EID blocks MAY be assigned in a
440 hierarchical manner, independent of the network topology, to
441 facilitate scaling of the mapping database. In addition, an EID
442 block assigned to a site may have site-local structure
443 (subnetting) for routing within the site; this structure is not
444 visible to the global routing system. In theory, the bit string
445 that represents an EID for one device can represent an RLOC for a
446 different device. As the architecture is realized, if a given bit
447 string is both an RLOC and an EID, it must refer to the same
448 entity in both cases. When used in discussions with other
449 Locator/ID separation proposals, a LISP EID will be called an
450 "LEID". Throughout this document, any references to "EID" refer
453 EID-Prefix: An EID-Prefix is a power-of-two block of EIDs that are
454 allocated to a site by an address allocation authority.
455 EID-Prefixes are associated with a set of RLOC addresses that make
456 up a "database mapping". EID-Prefix allocations can be broken up
457 into smaller blocks when an RLOC set is to be associated with the
458 larger EID-Prefix block. A globally routed address block (whether
459 PI or PA) is not inherently an EID-Prefix. A globally routed
460 address block MAY be used by its assignee as an EID block. The
461 converse is not supported. That is, a site that receives an
462 explicitly allocated EID-Prefix may not use that EID-Prefix as a
463 globally routed prefix. This would require coordination and
464 cooperation with the entities managing the mapping infrastructure.
465 Once this has been done, that block could be removed from the
466 globally routed IP system, if other suitable transition and access
467 mechanisms are in place. Discussion of such transition and access
468 mechanisms can be found in [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>] and [<a href="#ref-LISP-DEPLOY">LISP-DEPLOY</a>].
477 <span class="grey">Farinacci, et al. Experimental [Page 6]</span>
478 </pre><!--NewPage--><pre class="newpage"><a name="page-7" id="page-7" href="#page-7" class="invisible"> </a>
479 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
482 End-system: An end-system is an IPv4 or IPv6 device that originates
483 packets with a single IPv4 or IPv6 header. The end-system
484 supplies an EID value for the destination address field of the IP
485 header when communicating globally (i.e., outside of its routing
486 domain). An end-system can be a host computer, a switch or router
487 device, or any network appliance.
489 Ingress Tunnel Router (ITR): An ITR is a router that resides in a
490 LISP site. Packets sent by sources inside of the LISP site to
491 destinations outside of the site are candidates for encapsulation
492 by the ITR. The ITR treats the IP destination address as an EID
493 and performs an EID-to-RLOC mapping lookup. The router then
494 prepends an "outer" IP header with one of its globally routable
495 RLOCs in the source address field and the result of the mapping
496 lookup in the destination address field. Note that this
497 destination RLOC MAY be an intermediate, proxy device that has
498 better knowledge of the EID-to-RLOC mapping closer to the
499 destination EID. In general, an ITR receives IP packets from site
500 end-systems on one side and sends LISP-encapsulated IP packets
501 toward the Internet on the other side.
503 Specifically, when a service provider prepends a LISP header for
504 Traffic Engineering purposes, the router that does this is also
505 regarded as an ITR. The outer RLOC the ISP ITR uses can be based
506 on the outer destination address (the originating ITR's supplied
507 RLOC) or the inner destination address (the originating host's
510 TE-ITR: A TE-ITR is an ITR that is deployed in a service provider
511 network that prepends an additional LISP header for Traffic
512 Engineering purposes.
514 Egress Tunnel Router (ETR): An ETR is a router that accepts an IP
515 packet where the destination address in the "outer" IP header is
516 one of its own RLOCs. The router strips the "outer" header and
517 forwards the packet based on the next IP header found. In
518 general, an ETR receives LISP-encapsulated IP packets from the
519 Internet on one side and sends decapsulated IP packets to site
520 end-systems on the other side. ETR functionality does not have to
521 be limited to a router device. A server host can be the endpoint
522 of a LISP tunnel as well.
524 TE-ETR: A TE-ETR is an ETR that is deployed in a service provider
525 network that strips an outer LISP header for Traffic Engineering
533 <span class="grey">Farinacci, et al. Experimental [Page 7]</span>
534 </pre><!--NewPage--><pre class="newpage"><a name="page-8" id="page-8" href="#page-8" class="invisible"> </a>
535 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
538 xTR: An xTR is a reference to an ITR or ETR when direction of data
539 flow is not part of the context description. "xTR" refers to the
540 router that is the tunnel endpoint and is used synonymously with
541 the term "Tunnel Router". For example, "An xTR can be located at
542 the Customer Edge (CE) router" indicates both ITR and ETR
543 functionality at the CE router.
545 LISP Router: A LISP router is a router that performs the functions
546 of any or all of the following: ITR, ETR, Proxy-ITR (PITR), or
549 EID-to-RLOC Cache: The EID-to-RLOC Cache is a short-lived,
550 on-demand table in an ITR that stores, tracks, and is responsible
551 for timing out and otherwise validating EID-to-RLOC mappings.
552 This cache is distinct from the full "database" of EID-to-RLOC
553 mappings; it is dynamic, local to the ITR(s), and relatively
554 small, while the database is distributed, relatively static, and
555 much more global in scope.
557 EID-to-RLOC Database: The EID-to-RLOC Database is a global
558 distributed database that contains all known EID-Prefix-to-RLOC
559 mappings. Each potential ETR typically contains a small piece of
560 the database: the EID-to-RLOC mappings for the EID-Prefixes
561 "behind" the router. These map to one of the router's own
562 globally visible IP addresses. The same database mapping entries
563 MUST be configured on all ETRs for a given site. In a steady
564 state, the EID-Prefixes for the site and the Locator-Set for each
565 EID-Prefix MUST be the same on all ETRs. Procedures to enforce
566 and/or verify this are outside the scope of this document. Note
567 that there MAY be transient conditions when the EID-Prefix for the
568 site and Locator-Set for each EID-Prefix may not be the same on
569 all ETRs. This has no negative implications, since a partial set
570 of Locators can be used.
572 Recursive Tunneling: Recursive Tunneling occurs when a packet has
573 more than one LISP IP header. Additional layers of tunneling MAY
574 be employed to implement Traffic Engineering or other re-routing
575 as needed. When this is done, an additional "outer" LISP header
576 is added, and the original RLOCs are preserved in the "inner"
577 header. Any references to tunnels in this specification refer to
578 dynamic encapsulating tunnels; they are never statically
581 Re-encapsulating Tunnels: Re-encapsulating Tunneling occurs when an
582 ETR removes a LISP header, then acts as an ITR to prepend another
583 LISP header. Doing this allows a packet to be re-routed by the
584 re-encapsulating router without adding the overhead of additional
585 tunnel headers. Any references to tunnels in this specification
589 <span class="grey">Farinacci, et al. Experimental [Page 8]</span>
590 </pre><!--NewPage--><pre class="newpage"><a name="page-9" id="page-9" href="#page-9" class="invisible"> </a>
591 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
594 refer to dynamic encapsulating tunnels; they are never statically
595 configured. When using multiple mapping database systems, care
596 must be taken to not create re-encapsulation loops through
599 LISP Header: LISP header is a term used in this document to refer
600 to the outer IPv4 or IPv6 header, a UDP header, and a LISP-
601 specific 8-octet header that follow the UDP header and that an ITR
602 prepends or an ETR strips.
604 Address Family Identifier (AFI): AFI is a term used to describe an
605 address encoding in a packet. An address family currently
606 pertains to an IPv4 or IPv6 address. See [<a href="#ref-AFI" title=""Address Family Numbers"">AFI</a>] and [<a href="http://tools.ietf.org/html/rfc3232" title=""Assigned Numbers: RFC 1700 is Replaced by an On-line Database"">RFC3232</a>] for
607 details. An AFI value of 0 used in this specification indicates
608 an unspecified encoded address where the length of the address is
609 0 octets following the 16-bit AFI value of 0.
611 Negative Mapping Entry: A negative mapping entry, also known as a
612 negative cache entry, is an EID-to-RLOC entry where an EID-Prefix
613 is advertised or stored with no RLOCs. That is, the Locator-Set
614 for the EID-to-RLOC entry is empty or has an encoded Locator count
615 of 0. This type of entry could be used to describe a prefix from
616 a non-LISP site, which is explicitly not in the mapping database.
617 There are a set of well-defined actions that are encoded in a
618 Negative Map-Reply (<a href="#section-6.1.5">Section 6.1.5</a>).
620 Data-Probe: A Data-Probe is a LISP-encapsulated data packet where
621 the inner-header destination address equals the outer-header
622 destination address used to trigger a Map-Reply by a decapsulating
623 ETR. In addition, the original packet is decapsulated and
624 delivered to the destination host if the destination EID is in the
625 EID-Prefix range configured on the ETR. Otherwise, the packet is
626 discarded. A Data-Probe is used in some of the mapping database
627 designs to "probe" or request a Map-Reply from an ETR; in other
628 cases, Map-Requests are used. See each mapping database design
629 for details. When using Data-Probes, by sending Map-Requests on
630 the underlying routing system, EID-Prefixes must be advertised.
631 However, this is discouraged if the core is to scale by having
632 less EID-Prefixes stored in the core router's routing tables.
634 Proxy-ITR (PITR): A PITR is defined and described in [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>]. A
635 PITR acts like an ITR but does so on behalf of non-LISP sites that
636 send packets to destinations at LISP sites.
638 Proxy-ETR (PETR): A PETR is defined and described in [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>]. A
639 PETR acts like an ETR but does so on behalf of LISP sites that
640 send packets to destinations at non-LISP sites.
645 <span class="grey">Farinacci, et al. Experimental [Page 9]</span>
646 </pre><!--NewPage--><pre class="newpage"><a name="page-10" id="page-10" href="#page-10" class="invisible"> </a>
647 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
650 Route-returnability: Route-returnability is an assumption that the
651 underlying routing system will deliver packets to the destination.
652 When combined with a nonce that is provided by a sender and
653 returned by a receiver, this limits off-path data insertion. A
654 route-returnability check is verified when a message is sent with
655 a nonce, another message is returned with the same nonce, and the
656 destination of the original message appears as the source of the
659 LISP site: LISP site is a set of routers in an edge network that are
660 under a single technical administration. LISP routers that reside
661 in the edge network are the demarcation points to separate the
662 edge network from the core network.
664 Client-side: Client-side is a term used in this document to indicate
665 a connection initiation attempt by an EID. The ITR(s) at the LISP
666 site are the first to get involved in obtaining database Map-Cache
667 entries by sending Map-Request messages.
669 Server-side: Server-side is a term used in this document to indicate
670 that a connection initiation attempt is being accepted for a
671 destination EID. The ETR(s) at the destination LISP site are the
672 first to send Map-Replies to the source site initiating the
673 connection. The ETR(s) at this destination site can obtain
674 mappings by gleaning information from Map-Requests, Data-Probes,
675 or encapsulated packets.
677 Locator-Status-Bits (LSBs): Locator-Status-Bits are present in the
678 LISP header. They are used by ITRs to inform ETRs about the up/
679 down status of all ETRs at the local site. These bits are used as
680 a hint to convey up/down router status and not path reachability
681 status. The LSBs can be verified by use of one of the Locator
682 reachability algorithms described in <a href="#section-6.3">Section 6.3</a>.
684 Anycast Address: Anycast Address is a term used in this document to
685 refer to the same IPv4 or IPv6 address configured and used on
686 multiple systems at the same time. An EID or RLOC can be an
687 anycast address in each of their own address spaces.
689 <span class="h2"><h2><a class="selflink" name="section-4" href="#section-4">4</a>. Basic Overview</h2></span>
691 One key concept of LISP is that end-systems (hosts) operate the same
692 way they do today. The IP addresses that hosts use for tracking
693 sockets and connections, and for sending and receiving packets, do
694 not change. In LISP terminology, these IP addresses are called
695 Endpoint Identifiers (EIDs).
701 <span class="grey">Farinacci, et al. Experimental [Page 10]</span>
702 </pre><!--NewPage--><pre class="newpage"><a name="page-11" id="page-11" href="#page-11" class="invisible"> </a>
703 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
706 Routers continue to forward packets based on IP destination
707 addresses. When a packet is LISP encapsulated, these addresses are
708 referred to as Routing Locators (RLOCs). Most routers along a path
709 between two hosts will not change; they continue to perform routing/
710 forwarding lookups on the destination addresses. For routers between
711 the source host and the ITR as well as routers from the ETR to the
712 destination host, the destination address is an EID. For the routers
713 between the ITR and the ETR, the destination address is an RLOC.
715 Another key LISP concept is the "Tunnel Router". A Tunnel Router
716 prepends LISP headers on host-originated packets and strips them
717 prior to final delivery to their destination. The IP addresses in
718 this "outer header" are RLOCs. During end-to-end packet exchange
719 between two Internet hosts, an ITR prepends a new LISP header to each
720 packet, and an ETR strips the new header. The ITR performs
721 EID-to-RLOC lookups to determine the routing path to the ETR, which
722 has the RLOC as one of its IP addresses.
724 Some basic rules governing LISP are:
726 <span class="arch">o End-systems (hosts) only send to addresses that are EIDs. They
727 don't know that addresses are EIDs versus RLOCs but assume that
728 packets get to their intended destinations. In a system where
729 LISP is deployed, LISP routers intercept EID-addressed packets and
730 assist in delivering them across the network core where EIDs
731 cannot be routed. The procedure a host uses to send IP packets
732 does not change.</span>
734 <span class="arch">o EIDs are always IP addresses assigned to hosts.</span>
736 o LISP routers mostly deal with Routing Locator addresses. See
737 details in <a href="#section-4.1">Section 4.1</a> to clarify what is meant by "mostly".
739 o RLOCs are always IP addresses assigned to routers, preferably
740 topologically oriented addresses from provider CIDR (Classless
741 Inter-Domain Routing) blocks.
743 o When a router originates packets, it may use as a source address
744 either an EID or RLOC. When acting as a host (e.g., when
745 terminating a transport session such as Secure SHell (SSH),
746 TELNET, or the Simple Network Management Protocol (SNMP)), it may
747 use an EID that is explicitly assigned for that purpose. An EID
748 that identifies the router as a host MUST NOT be used as an RLOC;
749 an EID is only routable within the scope of a site. A typical BGP
750 configuration might demonstrate this "hybrid" EID/RLOC usage where
751 a router could use its "host-like" EID to terminate iBGP sessions
752 to other routers in a site while at the same time using RLOCs to
753 terminate eBGP sessions to routers outside the site.
757 <span class="grey">Farinacci, et al. Experimental [Page 11]</span>
758 </pre><!--NewPage--><pre class="newpage"><a name="page-12" id="page-12" href="#page-12" class="invisible"> </a>
759 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
762 o Packets with EIDs in them are not expected to be delivered
763 end-to-end in the absence of an EID-to-RLOC mapping operation.
764 They are expected to be used locally for intra-site communication
765 or to be encapsulated for inter-site communication.
767 o EID-Prefixes are likely to be hierarchically assigned in a manner
768 that is optimized for administrative convenience and to facilitate
769 scaling of the EID-to-RLOC mapping database. The hierarchy is
770 based on an address allocation hierarchy that is independent of
771 the network topology.
773 o EIDs may also be structured (subnetted) in a manner suitable for
774 local routing within an Autonomous System (AS).
776 An additional LISP header MAY be prepended to packets by a TE-ITR
777 when re-routing of the path for a packet is desired. A potential
778 use-case for this would be an ISP router that needs to perform
779 Traffic Engineering for packets flowing through its network. In such
780 a situation, termed "Recursive Tunneling", an ISP transit acts as an
781 additional ITR, and the RLOC it uses for the new prepended header
782 would be either a TE-ETR within the ISP (along an intra-ISP traffic
783 engineered path) or a TE-ETR within another ISP (an inter-ISP traffic
784 engineered path, where an agreement to build such a path exists).
786 In order to avoid excessive packet overhead as well as possible
787 encapsulation loops, this document mandates that a maximum of two
788 LISP headers can be prepended to a packet. For initial LISP
789 deployments, it is assumed that two headers is sufficient, where the
790 first prepended header is used at a site for Location/Identity
791 separation and the second prepended header is used inside a service
792 provider for Traffic Engineering purposes.
794 Tunnel Routers can be placed fairly flexibly in a multi-AS topology.
795 For example, the ITR for a particular end-to-end packet exchange
796 might be the first-hop or default router within a site for the source
797 host. Similarly, the ETR might be the last-hop router directly
798 connected to the destination host. Another example, perhaps for a
799 VPN service outsourced to an ISP by a site, the ITR could be the
800 site's border router at the service provider attachment point.
801 Mixing and matching of site-operated, ISP-operated, and other Tunnel
802 Routers is allowed for maximum flexibility. See <a href="#section-8">Section 8</a> for more
813 <span class="grey">Farinacci, et al. Experimental [Page 12]</span>
814 </pre><!--NewPage--><pre class="newpage"><a name="page-13" id="page-13" href="#page-13" class="invisible"> </a>
815 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
818 <span class="h3"><h3><a class="selflink" name="section-4.1" href="#section-4.1">4.1</a>. Packet Flow Sequence</h3></span>
820 This section provides an example of the unicast packet flow with the
821 following conditions:
823 o Source host "host1.abc.example.com" is sending a packet to
824 "host2.xyz.example.com", exactly what host1 would do if the site
827 o Each site is multihomed, so each Tunnel Router has an address
828 (RLOC) assigned from the service provider address block for each
829 provider to which that particular Tunnel Router is attached.
831 o The ITR(s) and ETR(s) are directly connected to the source and
832 destination, respectively, but the source and destination can be
833 located anywhere in the LISP site.
835 o Map-Requests can be sent on the underlying routing system
836 topology, to a mapping database system, or directly over an
837 Alternative Logical Topology [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>]. A Map-Request is sent for
838 an external destination when the destination is not found in the
839 forwarding table or matches a default route.
841 o Map-Replies are sent on the underlying routing system topology.
843 Client host1.abc.example.com wants to communicate with server
844 host2.xyz.example.com:
846 1. host1.abc.example.com wants to open a TCP connection to
847 host2.xyz.example.com. It does a DNS lookup on
848 host2.xyz.example.com. An A/AAAA record is returned. This
849 address is the destination EID. The locally assigned address of
850 host1.abc.example.com is used as the source EID. An IPv4 or IPv6
851 packet is built and forwarded through the LISP site as a normal
852 IP packet until it reaches a LISP ITR.
854 2. The LISP ITR must be able to map the destination EID to an RLOC
855 of one of the ETRs at the destination site. The specific method
856 used to do this is not described in this example. See [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>]
857 or [<a href="#ref-CONS" title=""LISP-CONS: A Content distribution Overlay Network Service for LISP"">CONS</a>] for possible solutions.
859 3. The ITR will send a LISP Map-Request. Map-Requests SHOULD be
869 <span class="grey">Farinacci, et al. Experimental [Page 13]</span>
870 </pre><!--NewPage--><pre class="newpage"><a name="page-14" id="page-14" href="#page-14" class="invisible"> </a>
871 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
874 4. When an alternate mapping system is not in use, the Map-Request
875 packet is routed through the underlying routing system.
876 Otherwise, the Map-Request packet is routed on an alternate
877 logical topology, for example, the [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>] database mapping
878 system. In either case, when the Map-Request arrives at one of
879 the ETRs at the destination site, it will process the packet as a
882 5. The ETR looks at the destination EID of the Map-Request and
883 matches it against the prefixes in the ETR's configured
884 EID-to-RLOC mapping database. This is the list of EID-Prefixes
885 the ETR is supporting for the site it resides in. If there is no
886 match, the Map-Request is dropped. Otherwise, a LISP Map-Reply
887 is returned to the ITR.
889 6. The ITR receives the Map-Reply message, parses the message (to
890 check for format validity), and stores the mapping information
891 from the packet. This information is stored in the ITR's
892 EID-to-RLOC mapping cache. Note that the map-cache is an
893 on-demand cache. An ITR will manage its map-cache in such a way
894 that optimizes for its resource constraints.
896 7. Subsequent packets from host1.abc.example.com to
897 host2.xyz.example.com will have a LISP header prepended by the
898 ITR using the appropriate RLOC as the LISP header destination
899 address learned from the ETR. Note that the packet MAY be sent
900 to a different ETR than the one that returned the Map-Reply due
901 to the source site's hashing policy or the destination site's
904 8. The ETR receives these packets directly (since the destination
905 address is one of its assigned IP addresses), checks the validity
906 of the addresses, strips the LISP header, and forwards packets to
907 the attached destination host.
909 In order to defer the need for a mapping lookup in the reverse
910 direction, an ETR MAY create a cache entry that maps the source EID
911 (inner-header source IP address) to the source RLOC (outer-header
912 source IP address) in a received LISP packet. Such a cache entry is
913 termed a "gleaned" mapping and only contains a single RLOC for the
914 EID in question. More complete information about additional RLOCs
915 SHOULD be verified by sending a LISP Map-Request for that EID. Both
916 the ITR and the ETR may also influence the decision the other makes
917 in selecting an RLOC. See <a href="#section-6">Section 6</a> for more details.
925 <span class="grey">Farinacci, et al. Experimental [Page 14]</span>
926 </pre><!--NewPage--><pre class="newpage"><a name="page-15" id="page-15" href="#page-15" class="invisible"> </a>
927 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
930 <span class="h2"><h2><a class="selflink" name="section-5" href="#section-5">5</a>. LISP Encapsulation Details</h2></span>
932 Since additional tunnel headers are prepended, the packet becomes
933 larger and can exceed the MTU of any link traversed from the ITR to
934 the ETR. It is RECOMMENDED in IPv4 that packets do not get
935 fragmented as they are encapsulated by the ITR. Instead, the packet
936 is dropped and an ICMP Too Big message is returned to the source.
938 This specification RECOMMENDS that implementations provide support
939 for one of the proposed fragmentation and reassembly schemes. Two
940 existing schemes are detailed in <a href="#section-5.4">Section 5.4</a>.
942 Since IPv4 or IPv6 addresses can be either EIDs or RLOCs, the LISP
943 architecture supports IPv4 EIDs with IPv6 RLOCs (where the inner
944 header is in IPv4 packet format and the outer header is in IPv6
945 packet format) or IPv6 EIDs with IPv4 RLOCs (where the inner header
946 is in IPv6 packet format and the outer header is in IPv4 packet
947 format). The next sub-sections illustrate packet formats for the
948 homogeneous case (IPv4-in-IPv4 and IPv6-in-IPv6), but all 4
949 combinations MUST be supported.
981 <span class="grey">Farinacci, et al. Experimental [Page 15]</span>
982 </pre><!--NewPage--><pre class="newpage"><a name="page-16" id="page-16" href="#page-16" class="invisible"> </a>
983 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
986 <span class="h3"><h3><a class="selflink" name="section-5.1" href="#section-5.1">5.1</a>. LISP IPv4-in-IPv4 Header Format</h3></span>
989 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
990 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
991 / |Version| IHL |Type of Service| Total Length |
992 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
993 | | Identification |Flags| Fragment Offset |
994 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
995 OH | Time to Live | Protocol = 17 | Header Checksum |
996 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
997 | | Source Routing Locator |
998 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
999 \ | Destination Routing Locator |
1000 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1001 / | Source Port = xxxx | Dest Port = 4341 |
1002 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1003 \ | UDP Length | UDP Checksum |
1004 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1005 L |N|L|E|V|I|flags| Nonce/Map-Version |
1006 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1007 S / | Instance ID/Locator-Status-Bits |
1008 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1009 / |Version| IHL |Type of Service| Total Length |
1010 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1011 | | Identification |Flags| Fragment Offset |
1012 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1013 IH | Time to Live | Protocol | Header Checksum |
1014 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1016 \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1017 \ | Destination EID |
1018 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1020 IHL = IP-Header-Length
1037 <span class="grey">Farinacci, et al. Experimental [Page 16]</span>
1038 </pre><!--NewPage--><pre class="newpage"><a name="page-17" id="page-17" href="#page-17" class="invisible"> </a>
1039 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1042 <span class="h3"><h3><a class="selflink" name="section-5.2" href="#section-5.2">5.2</a>. LISP IPv6-in-IPv6 Header Format</h3></span>
1045 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
1046 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1047 / |Version| Traffic Class | Flow Label |
1048 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1049 | | Payload Length | Next Header=17| Hop Limit |
1050 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1054 t + Source Routing Locator +
1058 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1062 ^ + Destination Routing Locator +
1066 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1067 / | Source Port = xxxx | Dest Port = 4341 |
1068 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1069 \ | UDP Length | UDP Checksum |
1070 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1071 L |N|L|E|V|I|flags| Nonce/Map-Version |
1072 I \ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1073 S / | Instance ID/Locator-Status-Bits |
1074 P +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1075 / |Version| Traffic Class | Flow Label |
1076 / +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1077 / | Payload Length | Next Header | Hop Limit |
1078 v +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1093 <span class="grey">Farinacci, et al. Experimental [Page 17]</span>
1094 </pre><!--NewPage--><pre class="newpage"><a name="page-18" id="page-18" href="#page-18" class="invisible"> </a>
1095 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1105 H +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1109 ^ + Destination EID +
1113 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1115 <span class="h3"><h3><a class="selflink" name="section-5.3" href="#section-5.3">5.3</a>. Tunnel Header Field Descriptions</h3></span>
1117 Inner Header (IH): The inner header is the header on the datagram
1118 received from the originating host. The source and destination IP
1119 addresses are EIDs [<a href="http://tools.ietf.org/html/rfc0791" title=""Internet Protocol"">RFC0791</a>] [<a href="http://tools.ietf.org/html/rfc2460" title=""Internet Protocol, Version 6 (IPv6) Specification"">RFC2460</a>].
1121 Outer Header: (OH) The outer header is a new header prepended by an
1122 ITR. The address fields contain RLOCs obtained from the ingress
1123 router's EID-to-RLOC Cache. The IP protocol number is "UDP (17)"
1124 from [<a href="http://tools.ietf.org/html/rfc0768" title=""User Datagram Protocol"">RFC0768</a>]. The setting of the Don't Fragment (DF) bit
1125 'Flags' field is according to rules listed in Sections <a href="#section-5.4.1">5.4.1</a> and
1128 UDP Header: The UDP header contains an ITR selected source port when
1129 encapsulating a packet. See <a href="#section-6.5">Section 6.5</a> for details on the hash
1130 algorithm used to select a source port based on the 5-tuple of the
1131 inner header. The destination port MUST be set to the well-known
1132 IANA-assigned port value 4341.
1134 UDP Checksum: The 'UDP Checksum' field SHOULD be transmitted as zero
1135 by an ITR for either IPv4 [<a href="http://tools.ietf.org/html/rfc0768" title=""User Datagram Protocol"">RFC0768</a>] or IPv6 encapsulation
1136 [<a href="#ref-UDP-TUNNELS">UDP-TUNNELS</a>] [<a href="#ref-UDP-ZERO" title=""Applicability Statement for the use of IPv6 UDP Datagrams with Zero Checksums"">UDP-ZERO</a>]. When a packet with a zero UDP checksum
1137 is received by an ETR, the ETR MUST accept the packet for
1138 decapsulation. When an ITR transmits a non-zero value for the UDP
1139 checksum, it MUST send a correctly computed value in this field.
1140 When an ETR receives a packet with a non-zero UDP checksum, it MAY
1141 choose to verify the checksum value. If it chooses to perform
1142 such verification, and the verification fails, the packet MUST be
1143 silently dropped. If the ETR chooses not to perform the
1144 verification, or performs the verification successfully, the
1145 packet MUST be accepted for decapsulation. The handling of UDP
1149 <span class="grey">Farinacci, et al. Experimental [Page 18]</span>
1150 </pre><!--NewPage--><pre class="newpage"><a name="page-19" id="page-19" href="#page-19" class="invisible"> </a>
1151 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1154 checksums for all tunneling protocols, including LISP, is under
1155 active discussion within the IETF. When that discussion
1156 concludes, any necessary changes will be made to align LISP with
1157 the outcome of the broader discussion.
1159 UDP Length: The 'UDP Length' field is set for an IPv4-encapsulated
1160 packet to be the sum of the inner-header IPv4 Total Length plus
1161 the UDP and LISP header lengths. For an IPv6-encapsulated packet,
1162 the 'UDP Length' field is the sum of the inner-header IPv6 Payload
1163 Length, the size of the IPv6 header (40 octets), and the size of
1164 the UDP and LISP headers.
1166 N: The N-bit is the nonce-present bit. When this bit is set to 1,
1167 the low-order 24 bits of the first 32 bits of the LISP header
1168 contain a Nonce. See <a href="#section-6.3.1">Section 6.3.1</a> for details. Both N- and
1169 V-bits MUST NOT be set in the same packet. If they are, a
1170 decapsulating ETR MUST treat the 'Nonce/Map-Version' field as
1171 having a Nonce value present.
1173 L: The L-bit is the 'Locator-Status-Bits' field enabled bit. When
1174 this bit is set to 1, the Locator-Status-Bits in the second
1175 32 bits of the LISP header are in use.
1178 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1179 |N|L|E|V|I|flags| Nonce/Map-Version |
1180 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1181 | Locator-Status-Bits |
1182 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1184 E: The E-bit is the echo-nonce-request bit. This bit MUST be ignored
1185 and has no meaning when the N-bit is set to 0. When the N-bit is
1186 set to 1 and this bit is set to 1, an ITR is requesting that the
1187 nonce value in the 'Nonce' field be echoed back in LISP-
1188 encapsulated packets when the ITR is also an ETR. See
1189 <a href="#section-6.3.1">Section 6.3.1</a> for details.
1191 V: The V-bit is the Map-Version present bit. When this bit is set to
1192 1, the N-bit MUST be 0. Refer to <a href="#section-6.6.3">Section 6.6.3</a> for more details.
1193 This bit indicates that the LISP header is encoded in this
1197 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1198 |N|L|E|V|I|flags| Source Map-Version | Dest Map-Version |
1199 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1200 | Instance ID/Locator-Status-Bits |
1201 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1205 <span class="grey">Farinacci, et al. Experimental [Page 19]</span>
1206 </pre><!--NewPage--><pre class="newpage"><a name="page-20" id="page-20" href="#page-20" class="invisible"> </a>
1207 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1210 I: The I-bit is the Instance ID bit. See <a href="#section-5.5">Section 5.5</a> for more
1211 details. When this bit is set to 1, the 'Locator-Status-Bits'
1212 field is reduced to 8 bits and the high-order 24 bits are used as
1213 an Instance ID. If the L-bit is set to 0, then the low-order
1214 8 bits are transmitted as zero and ignored on receipt. The format
1215 of the LISP header would look like this:
1218 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1219 |N|L|E|V|I|flags| Nonce/Map-Version |
1220 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1221 | Instance ID | LSBs |
1222 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1224 flags: The 'flags' field is a 3-bit field reserved for future flag
1225 use. It MUST be set to 0 on transmit and MUST be ignored on
1228 LISP Nonce: The LISP 'Nonce' field is a 24-bit value that is
1229 randomly generated by an ITR when the N-bit is set to 1. Nonce
1230 generation algorithms are an implementation matter but are
1231 required to generate different nonces when sending to different
1232 destinations. However, the same nonce can be used for a period of
1233 time to the same destination. The nonce is also used when the
1234 E-bit is set to request the nonce value to be echoed by the other
1235 side when packets are returned. When the E-bit is clear but the
1236 N-bit is set, a remote ITR is either echoing a previously
1237 requested echo-nonce or providing a random nonce. See
1238 <a href="#section-6.3.1">Section 6.3.1</a> for more details.
1240 LISP Locator-Status-Bits (LSBs): When the L-bit is also set, the
1241 'Locator-Status-Bits' field in the LISP header is set by an ITR to
1242 indicate to an ETR the up/down status of the Locators in the
1243 source site. Each RLOC in a Map-Reply is assigned an ordinal
1244 value from 0 to n-1 (when there are n RLOCs in a mapping entry).
1245 The Locator-Status-Bits are numbered from 0 to n-1 from the least
1246 significant bit of the field. The field is 32 bits when the I-bit
1247 is set to 0 and is 8 bits when the I-bit is set to 1. When a
1248 Locator-Status-Bit is set to 1, the ITR is indicating to the ETR
1249 that the RLOC associated with the bit ordinal has up status. See
1250 <a href="#section-6.3">Section 6.3</a> for details on how an ITR can determine the status of
1251 the ETRs at the same site. When a site has multiple EID-Prefixes
1252 that result in multiple mappings (where each could have a
1253 different Locator-Set), the Locator-Status-Bits setting in an
1254 encapsulated packet MUST reflect the mapping for the EID-Prefix
1255 that the inner-header source EID address matches. If the LSB for
1256 an anycast Locator is set to 1, then there is at least one RLOC
1257 with that address, and the ETR is considered 'up'.
1261 <span class="grey">Farinacci, et al. Experimental [Page 20]</span>
1262 </pre><!--NewPage--><pre class="newpage"><a name="page-21" id="page-21" href="#page-21" class="invisible"> </a>
1263 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1266 When doing ITR/PITR encapsulation:
1268 o The outer-header 'Time to Live' field (or 'Hop Limit' field, in
1269 the case of IPv6) SHOULD be copied from the inner-header 'Time to
1272 o The outer-header 'Type of Service' field (or the 'Traffic Class'
1273 field, in the case of IPv6) SHOULD be copied from the inner-header
1274 'Type of Service' field (with one exception; see below).
1276 When doing ETR/PETR decapsulation:
1278 o The inner-header 'Time to Live' field (or 'Hop Limit' field, in
1279 the case of IPv6) SHOULD be copied from the outer-header 'Time to
1280 Live' field, when the Time to Live value of the outer header is
1281 less than the Time to Live value of the inner header. Failing to
1282 perform this check can cause the Time to Live of the inner header
1283 to increment across encapsulation/decapsulation cycles. This
1284 check is also performed when doing initial encapsulation, when a
1285 packet comes to an ITR or PITR destined for a LISP site.
1287 o The inner-header 'Type of Service' field (or the 'Traffic Class'
1288 field, in the case of IPv6) SHOULD be copied from the outer-header
1289 'Type of Service' field (with one exception; see below).
1291 Note that if an ETR/PETR is also an ITR/PITR and chooses to
1292 re-encapsulate after decapsulating, the net effect of this is that
1293 the new outer header will carry the same Time to Live as the old
1294 outer header minus 1.
1296 Copying the Time to Live (TTL) serves two purposes: first, it
1297 preserves the distance the host intended the packet to travel;
1298 second, and more importantly, it provides for suppression of looping
1299 packets in the event there is a loop of concatenated tunnels due to
1300 misconfiguration. See <a href="#section-9.3">Section 9.3</a> for TTL exception handling for
1303 The Explicit Congestion Notification ('ECN') field occupies bits 6
1304 and 7 of both the IPv4 'Type of Service' field and the IPv6 'Traffic
1305 Class' field [<a href="http://tools.ietf.org/html/rfc3168" title=""The Addition of Explicit Congestion Notification (ECN) to IP"">RFC3168</a>]. The 'ECN' field requires special treatment
1306 in order to avoid discarding indications of congestion [<a href="http://tools.ietf.org/html/rfc3168" title=""The Addition of Explicit Congestion Notification (ECN) to IP"">RFC3168</a>].
1307 ITR encapsulation MUST copy the 2-bit 'ECN' field from the inner
1308 header to the outer header. Re-encapsulation MUST copy the 2-bit
1309 'ECN' field from the stripped outer header to the new outer header.
1310 If the 'ECN' field contains a congestion indication codepoint (the
1311 value is '11', the Congestion Experienced (CE) codepoint), then ETR
1312 decapsulation MUST copy the 2-bit 'ECN' field from the stripped outer
1313 header to the surviving inner header that is used to forward the
1317 <span class="grey">Farinacci, et al. Experimental [Page 21]</span>
1318 </pre><!--NewPage--><pre class="newpage"><a name="page-22" id="page-22" href="#page-22" class="invisible"> </a>
1319 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1322 packet beyond the ETR. These requirements preserve CE indications
1323 when a packet that uses ECN traverses a LISP tunnel and becomes
1324 marked with a CE indication due to congestion between the tunnel
1327 <span class="h3"><h3><a class="selflink" name="section-5.4" href="#section-5.4">5.4</a>. Dealing with Large Encapsulated Packets</h3></span>
1329 This section proposes two mechanisms to deal with packets that exceed
1330 the path MTU between the ITR and ETR.
1332 It is left to the implementor to decide if the stateless or stateful
1333 mechanism should be implemented. Both or neither can be used, since
1334 it is a local decision in the ITR regarding how to deal with MTU
1335 issues, and sites can interoperate with differing mechanisms.
1337 Both stateless and stateful mechanisms also apply to Re-encapsulating
1338 and Recursive Tunneling, so any actions below referring to an ITR
1339 also apply to a TE-ITR.
1341 <span class="h4"><h4><a class="selflink" name="section-5.4.1" href="#section-5.4.1">5.4.1</a>. A Stateless Solution to MTU Handling</h4></span>
1343 An ITR stateless solution to handle MTU issues is described as
1346 1. Define H to be the size, in octets, of the outer header an ITR
1347 prepends to a packet. This includes the UDP and LISP header
1350 2. Define L to be the size, in octets, of the maximum-sized packet
1351 an ITR can send to an ETR without the need for the ITR or any
1352 intermediate routers to fragment the packet.
1354 3. Define an architectural constant S for the maximum size of a
1355 packet, in octets, an ITR must receive so the effective MTU can
1356 be met. That is, S = L - H.
1358 When an ITR receives a packet from a site-facing interface and adds H
1359 octets worth of encapsulation to yield a packet size greater than L
1360 octets, it resolves the MTU issue by first splitting the original
1361 packet into 2 equal-sized fragments. A LISP header is then prepended
1362 to each fragment. The size of the encapsulated fragments is then
1363 (S/2 + H), which is less than the ITR's estimate of the path MTU
1364 between the ITR and its correspondent ETR.
1366 When an ETR receives encapsulated fragments, it treats them as two
1367 individually encapsulated packets. It strips the LISP headers and
1368 then forwards each fragment to the destination host of the
1369 destination site. The two fragments are reassembled at the
1373 <span class="grey">Farinacci, et al. Experimental [Page 22]</span>
1374 </pre><!--NewPage--><pre class="newpage"><a name="page-23" id="page-23" href="#page-23" class="invisible"> </a>
1375 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1378 destination host into the single IP datagram that was originated by
1379 the source host. Note that reassembly can happen at the ETR if the
1380 encapsulated packet was fragmented at or after the ITR.
1382 This behavior is performed by the ITR when the source host originates
1383 a packet with the 'DF' field of the IP header set to 0. When the
1384 'DF' field of the IP header is set to 1, or the packet is an IPv6
1385 packet originated by the source host, the ITR will drop the packet
1386 when the size is greater than L and send an ICMP Too Big message to
1387 the source with a value of S, where S is (L - H).
1389 When the outer-header encapsulation uses an IPv4 header, an
1390 implementation SHOULD set the DF bit to 1 so ETR fragment reassembly
1391 can be avoided. An implementation MAY set the DF bit in such headers
1392 to 0 if it has good reason to believe there are unresolvable path MTU
1393 issues between the sending ITR and the receiving ETR.
1395 This specification RECOMMENDS that L be defined as 1500.
1397 <span class="h4"><h4><a class="selflink" name="section-5.4.2" href="#section-5.4.2">5.4.2</a>. A Stateful Solution to MTU Handling</h4></span>
1399 An ITR stateful solution to handle MTU issues is described as follows
1400 and was first introduced in [<a href="#ref-OPENLISP" title=""OpenLISP Implementation Report"">OPENLISP</a>]:
1402 1. The ITR will keep state of the effective MTU for each Locator per
1403 Map-Cache entry. The effective MTU is what the core network can
1404 deliver along the path between the ITR and ETR.
1406 2. When an IPv6-encapsulated packet, or an IPv4-encapsulated packet
1407 with the DF bit set to 1, exceeds what the core network can
1408 deliver, one of the intermediate routers on the path will send an
1409 ICMP Too Big message to the ITR. The ITR will parse the ICMP
1410 message to determine which Locator is affected by the effective
1411 MTU change and then record the new effective MTU value in the
1414 3. When a packet is received by the ITR from a source inside of the
1415 site and the size of the packet is greater than the effective MTU
1416 stored with the Map-Cache entry associated with the destination
1417 EID the packet is for, the ITR will send an ICMP Too Big message
1418 back to the source. The packet size advertised by the ITR in the
1419 ICMP Too Big message is the effective MTU minus the LISP
1420 encapsulation length.
1422 Even though this mechanism is stateful, it has advantages over the
1423 stateless IP fragmentation mechanism, by not involving the
1424 destination host with reassembly of ITR fragmented packets.
1429 <span class="grey">Farinacci, et al. Experimental [Page 23]</span>
1430 </pre><!--NewPage--><pre class="newpage"><a name="page-24" id="page-24" href="#page-24" class="invisible"> </a>
1431 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1434 <span class="h3"><h3><a class="selflink" name="section-5.5" href="#section-5.5">5.5</a>. Using Virtualization and Segmentation with LISP</h3></span>
1436 When multiple organizations inside of a LISP site are using private
1437 addresses [<a href="http://tools.ietf.org/html/rfc1918" title=""Address Allocation for Private Internets"">RFC1918</a>] as EID-Prefixes, their address spaces MUST remain
1438 segregated due to possible address duplication. An Instance ID in
1439 the address encoding can aid in making the entire AFI-based address
1440 unique. See IANA Considerations (<a href="#section-14.2">Section 14.2</a>) for details on
1441 possible address encodings.
1443 An Instance ID can be carried in a LISP-encapsulated packet. An ITR
1444 that prepends a LISP header will copy a 24-bit value used by the LISP
1445 router to uniquely identify the address space. The value is copied
1446 to the 'Instance ID' field of the LISP header, and the I-bit is set
1449 When an ETR decapsulates a packet, the Instance ID from the LISP
1450 header is used as a table identifier to locate the forwarding table
1451 to use for the inner destination EID lookup.
1453 For example, an 802.1Q VLAN tag or VPN identifier could be used as a
1485 <span class="grey">Farinacci, et al. Experimental [Page 24]</span>
1486 </pre><!--NewPage--><pre class="newpage"><a name="page-25" id="page-25" href="#page-25" class="invisible"> </a>
1487 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1490 <span class="h2"><h2><a class="selflink" name="section-6" href="#section-6">6</a>. EID-to-RLOC Mapping</h2></span>
1492 <span class="h3"><h3><a class="selflink" name="section-6.1" href="#section-6.1">6.1</a>. LISP IPv4 and IPv6 Control-Plane Packet Formats</h3></span>
1494 The following UDP packet formats are used by the LISP control plane.
1497 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
1498 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1499 |Version| IHL |Type of Service| Total Length |
1500 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1501 | Identification |Flags| Fragment Offset |
1502 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1503 | Time to Live | Protocol = 17 | Header Checksum |
1504 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1505 | Source Routing Locator |
1506 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1507 | Destination Routing Locator |
1508 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1509 / | Source Port | Dest Port |
1510 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1511 \ | UDP Length | UDP Checksum |
1512 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1516 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1541 <span class="grey">Farinacci, et al. Experimental [Page 25]</span>
1542 </pre><!--NewPage--><pre class="newpage"><a name="page-26" id="page-26" href="#page-26" class="invisible"> </a>
1543 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1547 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
1548 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1549 |Version| Traffic Class | Flow Label |
1550 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1551 | Payload Length | Next Header=17| Hop Limit |
1552 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1556 + Source Routing Locator +
1560 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1564 + Destination Routing Locator +
1568 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1569 / | Source Port | Dest Port |
1570 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1571 \ | UDP Length | UDP Checksum |
1572 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1576 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1578 The LISP UDP-based messages are the Map-Request and Map-Reply
1579 messages. When a UDP Map-Request is sent, the UDP source port is
1580 chosen by the sender and the destination UDP port number is set to
1581 4342. <span class="impl">When a UDP Map-Reply is sent, the source UDP port number is
1582 set to 4342 and the destination UDP port number is copied from the
1583 source port of either the Map-Request</span> or the invoking data packet.
1584 Implementations MUST be prepared to accept packets when either the
1585 source port or destination UDP port is set to 4342 due to NATs
1586 changing port number values.
1588 The 'UDP Length' field will reflect the length of the UDP header and
1589 the LISP Message payload.
1597 <span class="grey">Farinacci, et al. Experimental [Page 26]</span>
1598 </pre><!--NewPage--><pre class="newpage"><a name="page-27" id="page-27" href="#page-27" class="invisible"> </a>
1599 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1602 The UDP checksum is computed and set to non-zero for Map-Request,
1603 Map-Reply, Map-Register, and Encapsulated Control Message (ECM)
1604 control messages. It MUST be checked on receipt, and if the checksum
1605 fails, the packet MUST be dropped.
1607 The format of control messages includes the UDP header so the
1608 checksum and length fields can be used to protect and delimit message
1611 <span class="h4"><h4><a class="selflink" name="section-6.1.1" href="#section-6.1.1">6.1.1</a>. LISP Packet Type Allocations</h4></span>
1613 This section will be the authoritative source for allocating LISP
1614 Type values and for defining LISP control message formats. Current
1618 LISP Map-Request: 1 b'0001'
1619 LISP Map-Reply: 2 b'0010'
1620 LISP Map-Register: 3 b'0011'
1621 LISP Map-Notify: 4 b'0100'
1622 LISP Encapsulated Control Message: 8 b'1000'
1624 <span class="h4"><h4><a class="selflink" name="section-6.1.2" href="#section-6.1.2">6.1.2</a>. Map-Request Message Format</h4></span>
1627 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
1628 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1629 |Type=1 |A|M|P|S|p|s| Reserved | IRC | Record Count |
1630 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1632 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1634 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1635 | Source-EID-AFI | Source EID Address ... |
1636 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1637 | ITR-RLOC-AFI 1 | ITR-RLOC Address 1 ... |
1638 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1640 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1641 | ITR-RLOC-AFI n | ITR-RLOC Address n ... |
1642 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1643 / | Reserved | EID mask-len | EID-Prefix-AFI |
1644 Rec +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1645 \ | EID-Prefix ... |
1646 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1647 | Map-Reply Record ... |
1648 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1653 <span class="grey">Farinacci, et al. Experimental [Page 27]</span>
1654 </pre><!--NewPage--><pre class="newpage"><a name="page-28" id="page-28" href="#page-28" class="invisible"> </a>
1655 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1658 Packet field descriptions:
1660 <span class="impl">Type: 1 (Map-Request)</span>
1662 A: This is an authoritative bit, which is set to 0 for UDP-based
1663 Map-Requests sent by an ITR. It is set to 1 when an ITR wants the
1664 destination site to return the Map-Reply rather than the mapping
1667 M: This is the map-data-present bit. When set, it indicates that a
1668 Map-Reply Record segment is included in the Map-Request.
1670 P: This is the probe-bit, which indicates that a Map-Request SHOULD
1671 be treated as a Locator reachability probe. The receiver SHOULD
1672 respond with a Map-Reply with the probe-bit set, indicating that
1673 the Map-Reply is a Locator reachability probe reply, with the
1674 nonce copied from the Map-Request. See <a href="#section-6.3.2">Section 6.3.2</a> for more
1677 S: This is the Solicit-Map-Request (SMR) bit. See <a href="#section-6.6.2">Section 6.6.2</a> for
1680 p: This is the PITR bit. This bit is set to 1 when a PITR sends a
1683 s: This is the SMR-invoked bit. This bit is set to 1 when an xTR is
1684 sending a Map-Request in response to a received SMR-based
1687 <span class="impl">Reserved: This field MUST be set to 0 on transmit and MUST be
1688 ignored on receipt.</span>
1690 IRC: This 5-bit field is the ITR-RLOC Count, which encodes the
1691 additional number of ('ITR-RLOC-AFI', 'ITR-RLOC Address') fields
1692 present in this message. At least one (ITR-RLOC-AFI,
1693 ITR-RLOC-Address) pair MUST be encoded. Multiple 'ITR-RLOC
1694 Address' fields are used, so a Map-Replier can select which
1695 destination address to use for a Map-Reply. The IRC value ranges
1696 from 0 to 31. For a value of 0, there is 1 ITR-RLOC address
1697 encoded; for a value of 1, there are 2 ITR-RLOC addresses encoded,
1698 and so on up to 31, which encodes a total of 32 ITR-RLOC
1701 Record Count: This is the number of records in this Map-Request
1702 message. A record is comprised of the portion of the packet that
1703 is labeled 'Rec' above and occurs the number of times equal to
1704 Record Count. For this version of the protocol, a receiver MUST
1705 accept and process Map-Requests that contain one or more records,
1709 <span class="grey">Farinacci, et al. Experimental [Page 28]</span>
1710 </pre><!--NewPage--><pre class="newpage"><a name="page-29" id="page-29" href="#page-29" class="invisible"> </a>
1711 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1714 but a sender MUST only send Map-Requests containing one record.
1715 Support for requesting multiple EIDs in a single Map-Request
1716 message will be specified in a future version of the protocol.
1718 <span class="impl"> Nonce: This is an 8-octet random value created by the sender of the
1719 Map-Request. This nonce will be returned in the Map-Reply. The
1720 security of the LISP mapping protocol critically depends on the
1721 strength of the nonce in the Map-Request message. The nonce
1722 SHOULD be generated by a properly seeded pseudo-random (or strong
1723 random) source. See [<a href="http://tools.ietf.org/html/rfc4086" title=""Randomness Requirements for Security"">RFC4086</a>] for advice on generating security-
1724 sensitive random data.</span>
1726 Source-EID-AFI: This is the address family of the 'Source EID
1729 Source EID Address: This is the EID of the source host that
1730 originated the packet that caused the Map-Request. When
1731 Map-Requests are used for refreshing a Map-Cache entry or for
1732 RLOC-Probing, an AFI value 0 is used and this field is of zero
1735 ITR-RLOC-AFI: This is the address family of the 'ITR-RLOC Address'
1736 field that follows this field.
1738 ITR-RLOC Address: This is used to give the ETR the option of
1739 selecting the destination address from any address family for the
1740 Map-Reply message. This address MUST be a routable RLOC address
1741 of the sender of the Map-Request message.
1743 EID mask-len: This is the mask length for the EID-Prefix.
1745 EID-Prefix-AFI: This is the address family of the EID-Prefix
1746 according to [<a href="#ref-AFI" title=""Address Family Numbers"">AFI</a>].
1748 <span class="part">EID-Prefix: This prefix is 4 octets for an IPv4 address family and
1749 16 octets for an IPv6 address family. When a Map-Request is sent
1750 by an ITR because a data packet is received for a destination
1751 where there is no mapping entry, the EID-Prefix is set to the
1752 destination IP address of the data packet, and the 'EID mask-len'
1753 is set to 32 or 128 for IPv4 or IPv6, respectively. When an xTR
1754 wants to query a site about the status of a mapping it already has
1755 cached, the EID-Prefix used in the Map-Request has the same mask
1756 length as the EID-Prefix returned from the site when it sent a
1757 Map-Reply message.</span>
1765 <span class="grey">Farinacci, et al. Experimental [Page 29]</span>
1766 </pre><!--NewPage--><pre class="newpage"><a name="page-30" id="page-30" href="#page-30" class="invisible"> </a>
1767 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1770 Map-Reply Record: When the M-bit is set, this field is the size of a
1771 single "Record" in the Map-Reply format. This Map-Reply record
1772 contains the EID-to-RLOC mapping entry associated with the Source
1773 EID. This allows the ETR that will receive this Map-Request to
1774 cache the data if it chooses to do so.
1776 <span class="h4"><h4><a class="selflink" name="section-6.1.3" href="#section-6.1.3">6.1.3</a>. EID-to-RLOC UDP Map-Request Message</h4></span>
1778 A Map-Request is sent from an ITR when it needs a mapping for an EID,
1779 wants to test an RLOC for reachability, or wants to refresh a mapping
1780 before TTL expiration. For the initial case, the destination IP
1781 address used for the Map-Request is the data packet's destination
1782 address (i.e., the destination EID) that had a mapping cache lookup
1783 failure. For the latter two cases, the destination IP address used
1784 for the Map-Request is one of the RLOC addresses from the Locator-Set
1785 of the Map-Cache entry. The source address is either an IPv4 or IPv6
1786 RLOC address, depending on whether the Map-Request is using an IPv4
1787 or IPv6 header, respectively. In all cases, the UDP source port
1788 number for the Map-Request message is a 16-bit value selected by the
1789 ITR/PITR, and the UDP destination port number is set to the well-
1790 known destination port number 4342. A successful Map-Reply, which is
1791 one that has a nonce that matches an outstanding Map-Request nonce,
1792 will update the cached set of RLOCs associated with the EID-Prefix
1795 One or more Map-Request ('ITR-RLOC-AFI', 'ITR-RLOC-Address') fields
1796 MUST be filled in by the ITR. The number of fields (minus 1) encoded
1797 MUST be placed in the 'IRC' field. The ITR MAY include all locally
1798 configured Locators in this list or just provide one locator address
1799 from each address family it supports. If the ITR erroneously
1800 provides no ITR-RLOC addresses, the Map-Replier MUST drop the
1803 Map-Requests can also be LISP encapsulated using UDP destination
1804 port 4342 with a LISP Type value set to "Encapsulated Control
1805 Message", when sent from an ITR to a Map-Resolver. Likewise,
1806 Map-Requests are LISP encapsulated the same way from a Map-Server to
1807 an ETR. Details on Encapsulated Map-Requests and Map-Resolvers can
1808 be found in [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>].
1810 Map-Requests MUST be rate-limited. It is RECOMMENDED that a
1811 Map-Request for the same EID-Prefix be sent no more than once per
1814 An ITR that is configured with mapping database information (i.e., it
1815 is also an ETR) MAY optionally include those mappings in a
1816 Map-Request. When an ETR configured to accept and verify such
1817 "piggybacked" mapping data receives such a Map-Request and it does
1821 <span class="grey">Farinacci, et al. Experimental [Page 30]</span>
1822 </pre><!--NewPage--><pre class="newpage"><a name="page-31" id="page-31" href="#page-31" class="invisible"> </a>
1823 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1826 not have this mapping in the map-cache, it MAY originate a "verifying
1827 Map-Request", addressed to the map-requesting ITR and the ETR MAY add
1828 a Map-Cache entry. If the ETR has a Map-Cache entry that matches the
1829 "piggybacked" EID and the RLOC is in the Locator-Set for the entry,
1830 then it may send the "verifying Map-Request" directly to the
1831 originating Map-Request source. If the RLOC is not in the
1832 Locator-Set, then the ETR MUST send the "verifying Map-Request" to
1833 the "piggybacked" EID. Doing this forces the "verifying Map-Request"
1834 to go through the mapping database system to reach the authoritative
1835 source of information about that EID, guarding against RLOC-spoofing
1836 in the "piggybacked" mapping data.
1838 <span class="h4"><h4><a class="selflink" name="section-6.1.4" href="#section-6.1.4">6.1.4</a>. Map-Reply Message Format</h4></span>
1841 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
1842 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1843 |Type=2 |P|E|S| Reserved | Record Count |
1844 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1846 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1848 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1850 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1851 R | Locator Count | EID mask-len | ACT |A| Reserved |
1852 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1853 c | Rsvd | Map-Version Number |<span class="part"> EID-Prefix-AFI </span>|
1854 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1855 r |<span class="part"> EID-Prefix </span>|
1856 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1857 | /| Priority | Weight | M Priority | M Weight |
1858 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1859 | o | Unused Flags |L|p|R|<span class="part"> Loc-AFI </span>|
1860 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1861 | \|<span class="part"> Locator </span>|
1862 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1877 <span class="grey">Farinacci, et al. Experimental [Page 31]</span>
1878 </pre><!--NewPage--><pre class="newpage"><a name="page-32" id="page-32" href="#page-32" class="invisible"> </a>
1879 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1882 Packet field descriptions:
1884 <span class="impl">Type: 2 (Map-Reply)</span>
1886 P: This is the probe-bit, which indicates that the Map-Reply is in
1887 response to a Locator reachability probe Map-Request. The 'Nonce'
1888 field MUST contain a copy of the nonce value from the original
1889 Map-Request. See <a href="#section-6.3.2">Section 6.3.2</a> for more details.
1891 E: This bit indicates that the ETR that sends this Map-Reply message
1892 is advertising that the site is enabled for the Echo-Nonce Locator
1893 reachability algorithm. See <a href="#section-6.3.1">Section 6.3.1</a> for more details.
1895 S: This is the Security bit. When set to 1, the following
1896 authentication information will be appended to the end of the
1897 Map-Reply. The detailed format of the Authentication Data Content
1898 is for further study.
1901 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
1902 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1903 | AD Type | Authentication Data Content . . . |
1904 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1906 Reserved: This field MUST be set to 0 on transmit and MUST be
1909 <span class="impl">Record Count: This is the number of records in this reply message.
1910 A record is comprised of that portion of the packet labeled
1911 'Record' above and occurs the number of times equal to Record
1914 Nonce: This is a 24-bit value set in a Data-Probe packet, or a
1915 64-bit value from the Map-Request is echoed in this 'Nonce' field
1916 of the Map-Reply. When a 24-bit value is supplied, it resides in
1917 the low-order 64 bits of the 'Nonce' field.</span>
1919 Record TTL: This is the time in minutes the recipient of the
1920 Map-Reply will store the mapping. If the TTL is 0, the entry
1921 SHOULD be removed from the cache immediately. If the value is
1922 0xffffffff, the recipient can decide locally how long to store the
1925 <span class="impl">Locator Count: This is the number of Locator entries. A Locator
1926 entry comprises what is labeled above as 'Loc'. The Locator count
1927 can be 0, indicating that there are no Locators for the
1933 <span class="grey">Farinacci, et al. Experimental [Page 32]</span>
1934 </pre><!--NewPage--><pre class="newpage"><a name="page-33" id="page-33" href="#page-33" class="invisible"> </a>
1935 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1938 EID mask-len: This is the mask length for the EID-Prefix.
1940 ACT: This 3-bit field describes Negative Map-Reply actions. In any
1941 other message type, these bits are set to 0 and ignored on
1942 receipt. These bits are used only when the 'Locator Count' field
1943 is set to 0. The action bits are encoded only in Map-Reply
1944 messages. The actions defined are used by an ITR or PITR when a
1945 destination EID matches a negative Map-Cache entry. Unassigned
1946 values should cause a Map-Cache entry to be created, and when
1947 packets match this negative cache entry, they will be dropped.
1948 The current assigned values are:
1950 (0) No-Action: The map-cache is kept alive, and no packet
1951 encapsulation occurs.
1953 (1) Natively-Forward: The packet is not encapsulated or dropped
1954 but natively forwarded.
1956 (2) Send-Map-Request: The packet invokes sending a Map-Request.
1958 (3) Drop: A packet that matches this map-cache entry is dropped.
1959 An ICMP Destination Unreachable message SHOULD be sent.
1961 A: The Authoritative bit, when sent, is always set to 1 by an ETR.
1962 When a Map-Server is proxy Map-Replying [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>] for a LISP site,
1963 the Authoritative bit is set to 0. This indicates to requesting
1964 ITRs that the Map-Reply was not originated by a LISP node managed
1965 at the site that owns the EID-Prefix.
1967 Map-Version Number: When this 12-bit value is non-zero, the
1968 Map-Reply sender is informing the ITR what the version number is
1969 for the EID record contained in the Map-Reply. The ETR can
1970 allocate this number internally but MUST coordinate this value
1971 with other ETRs for the site. When this value is 0, there is no
1972 versioning information conveyed. The Map-Version Number can be
1973 included in Map-Request and Map-Register messages. See
1974 <a href="#section-6.6.3">Section 6.6.3</a> for more details.
1976 EID-Prefix-AFI: Address family of the EID-Prefix according to [<a href="#ref-AFI" title=""Address Family Numbers"">AFI</a>].
1978 EID-Prefix: This prefix is 4 octets for an IPv4 address family and
1979 16 octets for an IPv6 address family.
1981 Priority: Each RLOC is assigned a unicast Priority. Lower values
1982 are more preferable. When multiple RLOCs have the same Priority,
1983 they MAY be used in a load-split fashion. A value of 255 means
1984 the RLOC MUST NOT be used for unicast forwarding.
1989 <span class="grey">Farinacci, et al. Experimental [Page 33]</span>
1990 </pre><!--NewPage--><pre class="newpage"><a name="page-34" id="page-34" href="#page-34" class="invisible"> </a>
1991 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
1994 Weight: When priorities are the same for multiple RLOCs, the Weight
1995 indicates how to balance unicast traffic between them. Weight is
1996 encoded as a relative weight of total unicast packets that match
1997 the mapping entry. For example, if there are 4 Locators in a
1998 Locator-Set, where the Weights assigned are 30, 20, 20, and 10,
1999 the first Locator will get 37.5% of the traffic, the 2nd and 3rd
2000 Locators will get 25% of the traffic, and the 4th Locator will get
2001 12.5% of the traffic. If all Weights for a Locator-Set are equal,
2002 the receiver of the Map-Reply will decide how to load-split the
2003 traffic. See <a href="#section-6.5">Section 6.5</a> for a suggested hash algorithm to
2004 distribute the load across Locators with the same Priority and
2005 equal Weight values.
2007 M Priority: Each RLOC is assigned a multicast Priority used by an
2008 ETR in a receiver multicast site to select an ITR in a source
2009 multicast site for building multicast distribution trees. A value
2010 of 255 means the RLOC MUST NOT be used for joining a multicast
2011 distribution tree. For more details, see [<a href="http://tools.ietf.org/html/rfc6831" title=""The Locator/ID Separation Protocol (LISP) for Multicast Environments"">RFC6831</a>].
2013 M Weight: When priorities are the same for multiple RLOCs, the
2014 Weight indicates how to balance building multicast distribution
2015 trees across multiple ITRs. The Weight is encoded as a relative
2016 weight (similar to the unicast Weights) of the total number of
2017 trees built to the source site identified by the EID-Prefix. If
2018 all Weights for a Locator-Set are equal, the receiver of the
2019 Map-Reply will decide how to distribute multicast state across
2020 ITRs. For more details, see [<a href="http://tools.ietf.org/html/rfc6831" title=""The Locator/ID Separation Protocol (LISP) for Multicast Environments"">RFC6831</a>].
2022 Unused Flags: These are set to 0 when sending and ignored on
2025 L: When this bit is set, the Locator is flagged as a local Locator to
2026 the ETR that is sending the Map-Reply. When a Map-Server is doing
2027 proxy Map-Replying [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>] for a LISP site, the L-bit is set to
2028 0 for all Locators in this Locator-Set.
2030 p: When this bit is set, an ETR informs the RLOC-Probing ITR that the
2031 locator address for which this bit is set is the one being
2032 RLOC-probed and MAY be different from the source address of the
2033 Map-Reply. An ITR that RLOC-probes a particular Locator MUST use
2034 this Locator for retrieving the data structure used to store the
2035 fact that the Locator is reachable. The p-bit is set for a single
2036 Locator in the same Locator-Set. If an implementation sets more
2037 than one p-bit erroneously, the receiver of the Map-Reply MUST
2038 select the first Locator. The p-bit MUST NOT be set for
2039 Locator-Set records sent in Map-Request and Map-Register messages.
2045 <span class="grey">Farinacci, et al. Experimental [Page 34]</span>
2046 </pre><!--NewPage--><pre class="newpage"><a name="page-35" id="page-35" href="#page-35" class="invisible"> </a>
2047 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2050 R: This is set when the sender of a Map-Reply has a route to the
2051 Locator in the Locator data record. This receiver may find this
2052 useful to know if the Locator is up but not necessarily reachable
2053 from the receiver's point of view. See also <a href="#section-6.4">Section 6.4</a> for
2054 another way the R-bit may be used.
2056 Locator: This is an IPv4 or IPv6 address (as encoded by the
2057 'Loc-AFI' field) assigned to an ETR. Note that the destination
2058 RLOC address MAY be an anycast address. A source RLOC can be an
2059 anycast address as well. The source or destination RLOC MUST NOT
2060 be the broadcast address (255.255.255.255 or any subnet broadcast
2061 address known to the router) and MUST NOT be a link-local
2062 multicast address. The source RLOC MUST NOT be a multicast
2063 address. The destination RLOC SHOULD be a multicast address if it
2064 is being mapped from a multicast destination EID.
2066 <span class="h4"><h4><a class="selflink" name="section-6.1.5" href="#section-6.1.5">6.1.5</a>. EID-to-RLOC UDP Map-Reply Message</h4></span>
2068 A Map-Reply returns an EID-Prefix with a prefix length that is less
2069 than or equal to the EID being requested. The EID being requested is
2070 either from the destination field of an IP header of a Data-Probe or
2071 the EID record of a Map-Request. The RLOCs in the Map-Reply are
2072 globally routable IP addresses of all ETRs for the LISP site. Each
2073 RLOC conveys status reachability but does not convey path
2074 reachability from a requester's perspective. Separate testing of
2075 path reachability is required. See <a href="#section-6.3">Section 6.3</a> for details.
2077 Note that a Map-Reply may contain different EID-Prefix granularity
2078 (prefix + length) than the Map-Request that triggers it. This might
2079 occur if a Map-Request were for a prefix that had been returned by an
2080 earlier Map-Reply. In such a case, the requester updates its cache
2081 with the new prefix information and granularity. For example, a
2082 requester with two cached EID-Prefixes that are covered by a
2083 Map-Reply containing one less-specific prefix replaces the entry with
2084 the less-specific EID-Prefix. Note that the reverse, replacement of
2085 one less-specific prefix with multiple more-specific prefixes, can
2086 also occur, not by removing the less-specific prefix but rather by
2087 adding the more-specific prefixes that, during a lookup, will
2088 override the less-specific prefix.
2101 <span class="grey">Farinacci, et al. Experimental [Page 35]</span>
2102 </pre><!--NewPage--><pre class="newpage"><a name="page-36" id="page-36" href="#page-36" class="invisible"> </a>
2103 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2106 When an ETR is configured with overlapping EID-Prefixes, a
2107 Map-Request with an EID that best matches any EID-Prefix MUST be
2108 returned in a single Map-Reply message. For instance, if an ETR had
2109 database mapping entries for EID-Prefixes:
2116 A Map-Request for EID 10.1.1.1 would cause a Map-Reply with a record
2117 count of 1 to be returned with a mapping record EID-Prefix of
2120 A Map-Request for EID 10.1.5.5 would cause a Map-Reply with a record
2121 count of 3 to be returned with mapping records for EID-Prefixes
2122 10.1.0.0/16, 10.1.1.0/24, and 10.1.2.0/24.
2124 Note that not all overlapping EID-Prefixes need to be returned but
2125 only the more-specific entries (note that in the second example above
2126 10.0.0.0/8 was not returned for requesting EID 10.1.5.5) for the
2127 matching EID-Prefix of the requesting EID. When more than one
2128 EID-Prefix is returned, all SHOULD use the same Time to Live value so
2129 they can all time out at the same time. When a more-specific
2130 EID-Prefix is received later, its Time to Live value in the Map-Reply
2131 record can be stored even when other less-specific entries exist.
2132 When a less-specific EID-Prefix is received later, its map-cache
2133 expiration time SHOULD be set to the minimum expiration time of any
2134 more-specific EID-Prefix in the map-cache. This is done so the
2135 integrity of the EID-Prefix set is wholly maintained and so no more-
2136 specific entries are removed from the map-cache while keeping less-
2139 Map-Replies SHOULD be sent for an EID-Prefix no more often than once
2140 per second to the same requesting router. For scalability, it is
2141 expected that aggregation of EID addresses into EID-Prefixes will
2142 allow one Map-Reply to satisfy a mapping for the EID addresses in the
2143 prefix range, thereby reducing the number of Map-Request messages.
2145 Map-Reply records can have an empty Locator-Set. A Negative
2146 Map-Reply is a Map-Reply with an empty Locator-Set. Negative
2147 Map-Replies convey special actions by the sender to the ITR or PITR
2148 that have solicited the Map-Reply. There are two primary
2149 applications for Negative Map-Replies. The first is for a
2150 Map-Resolver to instruct an ITR or PITR when a destination is for a
2151 LISP site versus a non-LISP site, and the other is to source quench
2152 Map-Requests that are sent for non-allocated EIDs.
2157 <span class="grey">Farinacci, et al. Experimental [Page 36]</span>
2158 </pre><!--NewPage--><pre class="newpage"><a name="page-37" id="page-37" href="#page-37" class="invisible"> </a>
2159 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2162 For each Map-Reply record, the list of Locators in a Locator-Set MUST
2163 appear in the same order for each ETR that originates a Map-Reply
2164 message. The Locator-Set MUST be sorted in order of ascending IP
2165 address where an IPv4 locator address is considered numerically 'less
2166 than' an IPv6 locator address.
2168 When sending a Map-Reply message, the destination address is copied
2169 from one of the 'ITR-RLOC' fields from the Map-Request. The ETR can
2170 choose a locator address from one of the address families it
2171 supports. For Data-Probes, the destination address of the Map-Reply
2172 is copied from the source address of the Data-Probe message that is
2173 invoking the reply. The source address of the Map-Reply is one of
2174 the local IP addresses chosen to allow Unicast Reverse Path
2175 Forwarding (uRPF) checks to succeed in the upstream service provider.
2176 The destination port of a Map-Reply message is copied from the source
2177 port of the Map-Request or Data-Probe, and the source port of the
2178 Map-Reply message is set to the well-known UDP port 4342.
2180 <span class="h5"><h5><a class="selflink" name="section-6.1.5.1" href="#section-6.1.5.1">6.1.5.1</a>. Traffic Redirection with Coarse EID-Prefixes</h5></span>
2182 When an ETR is misconfigured or compromised, it could return coarse
2183 EID-Prefixes in Map-Reply messages it sends. The EID-Prefix could
2184 cover EID-Prefixes that are allocated to other sites, redirecting
2185 their traffic to the Locators of the compromised site.
2187 To solve this problem, there are two basic solutions that could be
2188 used. The first is to have Map-Servers proxy Map-Reply on behalf of
2189 ETRs so their registered EID-Prefixes are the ones returned in
2190 Map-Replies. Since the interaction between an ETR and Map-Server is
2191 secured with shared keys, it is easier for an ETR to detect
2192 misbehavior. The second solution is to have ITRs and PITRs cache
2193 EID-Prefixes with mask lengths that are greater than or equal to a
2194 configured prefix length. This limits the damage to a specific width
2195 of any EID-Prefix advertised but needs to be coordinated with the
2196 allocation of site prefixes. These solutions can be used
2197 independently or at the same time.
2199 At the time of this writing, other approaches are being considered
2202 <span class="h4"><h4><a class="selflink" name="section-6.1.6" href="#section-6.1.6">6.1.6</a>. Map-Register Message Format</h4></span>
2204 The usage details of the Map-Register message can be found in
2205 specification [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>]. This section solely defines the message
2208 The message is sent in UDP with a destination UDP port of 4342 and a
2209 randomly selected UDP source port number.
2213 <span class="grey">Farinacci, et al. Experimental [Page 37]</span>
2214 </pre><!--NewPage--><pre class="newpage"><a name="page-38" id="page-38" href="#page-38" class="invisible"> </a>
2215 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2218 The Map-Register message format is:
2221 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
2222 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2223 |Type=3 |P| Reserved |<span class="impl">M</span>| Record Count |
2224 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2226 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2228 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2229 | Key ID | Authentication Data Length |
2230 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2231 ~ Authentication Data ~
2232 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2234 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2235 R | Locator Count | EID mask-len | ACT |A| Reserved |
2236 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2237 c | Rsvd | Map-Version Number | EID-Prefix-AFI |
2238 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2240 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2241 | /| Priority | Weight | M Priority | M Weight |
2242 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2243 | o | Unused Flags |L|p|R| Loc-AFI |
2244 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2246 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2248 Packet field descriptions:
2250 <span class="impl"> Type: 3 (Map-Register)</span>
2252 P: This is the proxy Map-Reply bit. When set to 1, an ETR sends a
2253 Map-Register message requesting the Map-Server to proxy a
2254 Map-Reply. The Map-Server will send non-authoritative Map-Replies
2255 on behalf of the ETR. Details on this usage can be found in
2256 [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>].
2258 Reserved: This field MUST be set to 0 on transmit and MUST be
2261 <span class="impl">M: This is the want-map-notify bit. When set to 1, an ETR is
2262 requesting a Map-Notify message to be returned in response to
2263 sending a Map-Register message. The Map-Notify message sent by a
2264 Map-Server is used to acknowledge receipt of a Map-Register
2269 <span class="grey">Farinacci, et al. Experimental [Page 38]</span>
2270 </pre><!--NewPage--><pre class="newpage"><a name="page-39" id="page-39" href="#page-39" class="invisible"> </a>
2271 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2274 Record Count: This is the number of records in this Map-Register
2275 message. A record is comprised of that portion of the packet
2276 labeled 'Record' above and occurs the number of times equal to
2279 <span class="impl"> Nonce: This 8-octet 'Nonce' field is set to 0 in Map-Register
2280 messages. Since the Map-Register message is authenticated, the
2281 'Nonce' field is not currently used for any security function but
2282 may be in the future as part of an anti-replay solution.</span>
2284 Key ID: This is a configured ID to find the configured Message
2285 Authentication Code (MAC) algorithm and key value used for the
2286 authentication function. See <a href="#section-14.4">Section 14.4</a> for codepoint
2289 Authentication Data Length: This is the length in octets of the
2290 'Authentication Data' field that follows this field. The length
2291 of the 'Authentication Data' field is dependent on the MAC
2292 algorithm used. The length field allows a device that doesn't
2293 know the MAC algorithm to correctly parse the packet.
2295 Authentication Data: This is the message digest used from the output
2296 of the MAC algorithm. The entire Map-Register payload is
2297 authenticated with this field preset to 0. After the MAC is
2298 computed, it is placed in this field. Implementations of this
2299 specification MUST include support for HMAC-SHA-1-96 [<a href="http://tools.ietf.org/html/rfc2404" title=""The Use of HMAC-SHA-1-96 within ESP and AH"">RFC2404</a>],
2300 and support for HMAC-SHA-256-128 [<a href="http://tools.ietf.org/html/rfc4868" title=""Using HMAC-SHA-256, HMAC-SHA-384, and HMAC-SHA-512 with IPsec"">RFC4868</a>] is RECOMMENDED.
2302 The definition of the rest of the Map-Register can be found in
2303 <a href="#section-6.1.4">Section 6.1.4</a>.
2305 <span class="h4"><h4><a class="selflink" name="section-6.1.7" href="#section-6.1.7">6.1.7</a>. Map-Notify Message Format</h4></span>
2307 The usage details of the Map-Notify message can be found in
2308 specification [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>]. This section solely defines the message
2311 The message is sent inside a UDP packet with source and destination
2312 UDP ports equal to 4342.
2325 <span class="grey">Farinacci, et al. Experimental [Page 39]</span>
2326 </pre><!--NewPage--><pre class="newpage"><a name="page-40" id="page-40" href="#page-40" class="invisible"> </a>
2327 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2330 The Map-Notify message format is:
2333 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
2334 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2335 |Type=4 | Reserved | Record Count |
2336 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2338 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2340 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2341 | Key ID | Authentication Data Length |
2342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2343 ~ Authentication Data ~
2344 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2346 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2347 R | Locator Count | EID mask-len | ACT |A| Reserved |
2348 e +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2349 c | Rsvd | Map-Version Number | EID-Prefix-AFI |
2350 o +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2352 d +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2353 | /| Priority | Weight | M Priority | M Weight |
2354 | L +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2355 | o | Unused Flags |L|p|R| Loc-AFI |
2356 | c +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2358 +-> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2360 Packet field descriptions:
2362 <span class="impl">Type: 4 (Map-Notify)</span>
2364 The Map-Notify message has the same contents as a Map-Register
2365 message. See the Map-Register section for field descriptions.
2381 <span class="grey">Farinacci, et al. Experimental [Page 40]</span>
2382 </pre><!--NewPage--><pre class="newpage"><a name="page-41" id="page-41" href="#page-41" class="invisible"> </a>
2383 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2386 <span class="h4"><h4><a class="selflink" name="section-6.1.8" href="#section-6.1.8">6.1.8</a>. Encapsulated Control Message Format</h4></span>
2388 An Encapsulated Control Message (ECM) is used to encapsulate control
2389 packets sent between xTRs and the mapping database system described
2390 in [<a href="http://tools.ietf.org/html/rfc6833" title=""Locator/ID Separation Protocol (LISP) Map-Server Interface"">RFC6833</a>].
2393 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
2394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2395 / | IPv4 or IPv6 Header |
2396 OH | (uses RLOC addresses) |
2398 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2399 / | Source Port = xxxx | Dest Port = 4342 |
2400 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2401 \ | UDP Length | UDP Checksum |
2402 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2403 LH |Type=8 |S| Reserved |
2404 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2405 / | IPv4 or IPv6 Header |
2406 IH | (uses RLOC or EID addresses) |
2408 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2409 / | Source Port = xxxx | Dest Port = yyyy |
2410 UDP +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2411 \ | UDP Length | UDP Checksum |
2412 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2413 LCM | LISP Control Message |
2414 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2416 Packet header descriptions:
2418 OH: The outer IPv4 or IPv6 header, which uses RLOC addresses in the
2419 source and destination header address fields.
2421 UDP: The outer UDP header with destination port 4342. The source
2422 port is randomly allocated. The checksum field MUST be
2425 LH: Type 8 is defined to be a "LISP Encapsulated Control Message",
2426 and what follows is either an IPv4 or IPv6 header as encoded by
2427 the first 4 bits after the 'Reserved' field.
2429 S: This is the Security bit. When set to 1, the field following
2430 the 'Reserved' field will have the following format. The
2431 detailed format of the Authentication Data Content is for
2437 <span class="grey">Farinacci, et al. Experimental [Page 41]</span>
2438 </pre><!--NewPage--><pre class="newpage"><a name="page-42" id="page-42" href="#page-42" class="invisible"> </a>
2439 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2443 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
2444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2445 | AD Type | Authentication Data Content . . . |
2446 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2448 IH: The inner IPv4 or IPv6 header, which can use either RLOC or EID
2449 addresses in the header address fields. When a Map-Request is
2450 encapsulated in this packet format, the destination address in
2451 this header is an EID.
2453 UDP: The inner UDP header, where the port assignments depend on the
2454 control packet being encapsulated. When the control packet is
2455 a Map-Request or Map-Register, the source port is selected by
2456 the ITR/PITR and the destination port is 4342. When the
2457 control packet is a Map-Reply, the source port is 4342 and the
2458 destination port is assigned from the source port of the
2459 invoking Map-Request. Port number 4341 MUST NOT be assigned to
2460 either port. The checksum field MUST be non-zero.
2462 LCM: The format is one of the control message formats described in
2463 this section. At this time, only Map-Request messages are
2464 allowed to be encapsulated. In the future, PIM Join/Prune
2465 messages [<a href="http://tools.ietf.org/html/rfc6831" title=""The Locator/ID Separation Protocol (LISP) for Multicast Environments"">RFC6831</a>] might be allowed. Encapsulating other types
2466 of LISP control messages is for further study. When
2467 Map-Requests are sent for RLOC-Probing purposes (i.e., the
2468 probe-bit is set), they MUST NOT be sent inside Encapsulated
2471 <span class="h3"><h3><a class="selflink" name="section-6.2" href="#section-6.2">6.2</a>. Routing Locator Selection</h3></span>
2473 Both the client-side and server-side may need control over the
2474 selection of RLOCs for conversations between them. This control is
2475 achieved by manipulating the 'Priority' and 'Weight' fields in
2476 EID-to-RLOC Map-Reply messages. Alternatively, RLOC information MAY
2477 be gleaned from received tunneled packets or EID-to-RLOC Map-Request
2480 The following are different scenarios for choosing RLOCs and the
2481 controls that are available:
2483 o The server-side returns one RLOC. The client-side can only use
2484 one RLOC. The server-side has complete control of the selection.
2486 o The server-side returns a list of RLOCs where a subset of the list
2487 has the same best Priority. The client can only use the subset
2488 list according to the weighting assigned by the server-side. In
2489 this case, the server-side controls both the subset list and
2493 <span class="grey">Farinacci, et al. Experimental [Page 42]</span>
2494 </pre><!--NewPage--><pre class="newpage"><a name="page-43" id="page-43" href="#page-43" class="invisible"> </a>
2495 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2498 load-splitting across its members. The client-side can use RLOCs
2499 outside of the subset list if it determines that the subset list
2500 is unreachable (unless RLOCs are set to a Priority of 255). Some
2501 sharing of control exists: the server-side determines the
2502 destination RLOC list and load distribution while the client-side
2503 has the option of using alternatives to this list if RLOCs in the
2504 list are unreachable.
2506 o The server-side sets a Weight of 0 for the RLOC subset list. In
2507 this case, the client-side can choose how the traffic load is
2508 spread across the subset list. Control is shared by the server-
2509 side determining the list and the client determining load
2510 distribution. Again, the client can use alternative RLOCs if the
2511 server-provided list of RLOCs is unreachable.
2513 o Either side (more likely the server-side ETR) decides not to send
2514 a Map-Request. For example, if the server-side ETR does not send
2515 Map-Requests, it gleans RLOCs from the client-side ITR, giving the
2516 client-side ITR responsibility for bidirectional RLOC reachability
2517 and preferability. Server-side ETR gleaning of the client-side
2518 ITR RLOC is done by caching the inner-header source EID and the
2519 outer-header source RLOC of received packets. The client-side ITR
2520 controls how traffic is returned and can alternate using an outer-
2521 header source RLOC, which then can be added to the list the
2522 server-side ETR uses to return traffic. Since no Priority or
2523 Weights are provided using this method, the server-side ETR MUST
2524 assume that each client-side ITR RLOC uses the same best Priority
2525 with a Weight of zero. In addition, since EID-Prefix encoding
2526 cannot be conveyed in data packets, the EID-to-RLOC Cache on
2527 Tunnel Routers can grow to be very large.
2529 o A "gleaned" Map-Cache entry, one learned from the source RLOC of a
2530 received encapsulated packet, is only stored and used for a few
2531 seconds, pending verification. Verification is performed by
2532 sending a Map-Request to the source EID (the inner-header IP
2533 source address) of the received encapsulated packet. A reply to
2534 this "verifying Map-Request" is used to fully populate the
2535 Map-Cache entry for the "gleaned" EID and is stored and used for
2536 the time indicated from the 'TTL' field of a received Map-Reply.
2537 When a verified Map-Cache entry is stored, data gleaning no longer
2538 occurs for subsequent packets that have a source EID that matches
2539 the EID-Prefix of the verified entry.
2541 RLOCs that appear in EID-to-RLOC Map-Reply messages are assumed to be
2542 reachable when the R-bit for the Locator record is set to 1. When
2543 the R-bit is set to 0, an ITR or PITR MUST NOT encapsulate to the
2544 RLOC. Neither the information contained in a Map-Reply nor that
2545 stored in the mapping database system provides reachability
2549 <span class="grey">Farinacci, et al. Experimental [Page 43]</span>
2550 </pre><!--NewPage--><pre class="newpage"><a name="page-44" id="page-44" href="#page-44" class="invisible"> </a>
2551 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2554 information for RLOCs. Note that reachability is not part of the
2555 mapping system and is determined using one or more of the Routing
2556 Locator reachability algorithms described in the next section.
2558 <span class="h3"><h3><a class="selflink" name="section-6.3" href="#section-6.3">6.3</a>. Routing Locator Reachability</h3></span>
2560 Several mechanisms for determining RLOC reachability are currently
2563 1. An ETR may examine the Locator-Status-Bits in the LISP header of
2564 an encapsulated data packet received from an ITR. If the ETR is
2565 also acting as an ITR and has traffic to return to the original
2566 ITR site, it can use this status information to help select an
2569 2. An ITR may receive an ICMP Network Unreachable or Host
2570 Unreachable message for an RLOC it is using. This indicates that
2571 the RLOC is likely down. Note that trusting ICMP messages may
2572 not be desirable, but neither is ignoring them completely.
2573 Implementations are encouraged to follow current best practices
2574 in treating these conditions.
2576 3. An ITR that participates in the global routing system can
2577 determine that an RLOC is down if no BGP Routing Information Base
2578 (RIB) route exists that matches the RLOC IP address.
2580 4. An ITR may receive an ICMP Port Unreachable message from a
2581 destination host. This occurs if an ITR attempts to use
2582 interworking [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>] and LISP-encapsulated data is sent to a
2583 non-LISP-capable site.
2585 5. An ITR may receive a Map-Reply from an ETR in response to a
2586 previously sent Map-Request. The RLOC source of the Map-Reply is
2587 likely up, since the ETR was able to send the Map-Reply to the
2590 6. When an ETR receives an encapsulated packet from an ITR, the
2591 source RLOC from the outer header of the packet is likely up.
2593 7. An ITR/ETR pair can use the Locator reachability algorithms
2594 described in this section, namely Echo-Noncing or RLOC-Probing.
2605 <span class="grey">Farinacci, et al. Experimental [Page 44]</span>
2606 </pre><!--NewPage--><pre class="newpage"><a name="page-45" id="page-45" href="#page-45" class="invisible"> </a>
2607 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2610 When determining Locator up/down reachability by examining the
2611 Locator-Status-Bits from the LISP-encapsulated data packet, an ETR
2612 will receive up-to-date status from an encapsulating ITR about
2613 reachability for all ETRs at the site. CE-based ITRs at the source
2614 site can determine reachability relative to each other using the site
2617 o Under normal circumstances, each ITR will advertise a default
2618 route into the site IGP.
2620 o If an ITR fails or if the upstream link to its PE fails, its
2621 default route will either time out or be withdrawn.
2623 Each ITR can thus observe the presence or lack of a default route
2624 originated by the others to determine the Locator-Status-Bits it sets
2627 RLOCs listed in a Map-Reply are numbered with ordinals 0 to n-1. The
2628 Locator-Status-Bits in a LISP-encapsulated packet are numbered from 0
2629 to n-1 starting with the least significant bit. For example, if an
2630 RLOC listed in the 3rd position of the Map-Reply goes down (ordinal
2631 value 2), then all ITRs at the site will clear the 3rd least
2632 significant bit (xxxx x0xx) of the 'Locator-Status-Bits' field for
2633 the packets they encapsulate.
2635 When an ETR decapsulates a packet, it will check for any change in
2636 the 'Locator-Status-Bits' field. When a bit goes from 1 to 0, the
2637 ETR, if acting also as an ITR, will refrain from encapsulating
2638 packets to an RLOC that is indicated as down. It will only resume
2639 using that RLOC if the corresponding Locator-Status-Bit returns to a
2640 value of 1. Locator-Status-Bits are associated with a Locator-Set
2641 per EID-Prefix. Therefore, when a Locator becomes unreachable, the
2642 Locator-Status-Bit that corresponds to that Locator's position in the
2643 list returned by the last Map-Reply will be set to zero for that
2644 particular EID-Prefix.
2646 When ITRs at the site are not deployed in CE routers, the IGP can
2647 still be used to determine the reachability of Locators, provided
2648 they are injected into the IGP. This is typically done when a /32
2649 address is configured on a loopback interface.
2651 When ITRs receive ICMP Network Unreachable or Host Unreachable
2652 messages as a method to determine unreachability, they will refrain
2653 from using Locators that are described in Locator lists of
2654 Map-Replies. However, using this approach is unreliable because many
2655 network operators turn off generation of ICMP Destination Unreachable
2661 <span class="grey">Farinacci, et al. Experimental [Page 45]</span>
2662 </pre><!--NewPage--><pre class="newpage"><a name="page-46" id="page-46" href="#page-46" class="invisible"> </a>
2663 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2666 If an ITR does receive an ICMP Network Unreachable or Host
2667 Unreachable message, it MAY originate its own ICMP Destination
2668 Unreachable message destined for the host that originated the data
2669 packet the ITR encapsulated.
2671 Also, BGP-enabled ITRs can unilaterally examine the RIB to see if a
2672 locator address from a Locator-Set in a mapping entry matches a
2673 prefix. If it does not find one and BGP is running in the Default-
2674 Free Zone (DFZ), it can decide to not use the Locator even though the
2675 Locator-Status-Bits indicate that the Locator is up. In this case,
2676 the path from the ITR to the ETR that is assigned the Locator is not
2677 available. More details are in [<a href="#ref-LOC-ID-ARCH">LOC-ID-ARCH</a>].
2679 Optionally, an ITR can send a Map-Request to a Locator, and if a
2680 Map-Reply is returned, reachability of the Locator has been
2681 determined. Obviously, sending such probes increases the number of
2682 control messages originated by Tunnel Routers for active flows, so
2683 Locators are assumed to be reachable when they are advertised.
2685 This assumption does create a dependency: Locator unreachability is
2686 detected by the receipt of ICMP Host Unreachable messages. When a
2687 Locator has been determined to be unreachable, it is not used for
2688 active traffic; this is the same as if it were listed in a Map-Reply
2691 The ITR can test the reachability of the unreachable Locator by
2692 sending periodic Requests. Both Requests and Replies MUST be rate-
2693 limited. Locator reachability testing is never done with data
2694 packets, since that increases the risk of packet loss for end-to-end
2697 When an ETR decapsulates a packet, it knows that it is reachable from
2698 the encapsulating ITR because that is how the packet arrived. In
2699 most cases, the ETR can also reach the ITR but cannot assume this to
2700 be true, due to the possibility of path asymmetry. In the presence
2701 of unidirectional traffic flow from an ITR to an ETR, the ITR SHOULD
2702 NOT use the lack of return traffic as an indication that the ETR is
2703 unreachable. Instead, it MUST use an alternate mechanism to
2704 determine reachability.
2706 <span class="h4"><h4><a class="selflink" name="section-6.3.1" href="#section-6.3.1">6.3.1</a>. Echo Nonce Algorithm</h4></span>
2708 When data flows bidirectionally between Locators from different
2709 sites, a data-plane mechanism called "nonce echoing" can be used to
2710 determine reachability between an ITR and ETR. When an ITR wants to
2711 solicit a nonce echo, it sets the N- and E-bits and places a 24-bit
2712 nonce [<a href="http://tools.ietf.org/html/rfc4086" title=""Randomness Requirements for Security"">RFC4086</a>] in the LISP header of the next encapsulated data
2717 <span class="grey">Farinacci, et al. Experimental [Page 46]</span>
2718 </pre><!--NewPage--><pre class="newpage"><a name="page-47" id="page-47" href="#page-47" class="invisible"> </a>
2719 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2722 <span class="nrel">When this packet is received by the ETR, the encapsulated packet is
2723 forwarded as normal. When the ETR next sends a data packet to the
2724 ITR, it includes the nonce received earlier with the N-bit set and
2725 E-bit cleared. The ITR sees this "echoed nonce" and knows that the
2726 path to and from the ETR is up.
2728 The ITR will set the E-bit and N-bit for every packet it sends while
2729 in the echo-nonce-request state. The time the ITR waits to process
2730 the echoed nonce before it determines the path is unreachable is
2731 variable and is a choice left for the implementation.
2733 If the ITR is receiving packets from the ETR but does not see the
2734 nonce echoed while being in the echo-nonce-request state, then the
2735 path to the ETR is unreachable. This decision may be overridden by
2736 other Locator reachability algorithms. Once the ITR determines that
2737 the path to the ETR is down, it can switch to another Locator for
2740 Note that "ITR" and "ETR" are relative terms here. Both devices MUST
2741 be implementing both ITR and ETR functionality for the echo nonce
2742 mechanism to operate.
2744 The ITR and ETR may both go into the echo-nonce-request state at the
2745 same time. The number of packets sent or the time during which echo
2746 nonce requests are sent is an implementation-specific setting.
2747 However, when an ITR is in the echo-nonce-request state, it can echo
2748 the ETR's nonce in the next set of packets that it encapsulates and
2749 subsequently continue sending echo-nonce-request packets.
2751 This mechanism does not completely solve the forward path
2752 reachability problem, as traffic may be unidirectional. That is, the
2753 ETR receiving traffic at a site may not be the same device as an ITR
2754 that transmits traffic from that site, or the site-to-site traffic is
2755 unidirectional so there is no ITR returning traffic.
2757 The echo-nonce algorithm is bilateral. That is, if one side sets the
2758 E-bit and the other side is not enabled for echo-noncing, then the
2759 echoing of the nonce does not occur and the requesting side may
2760 erroneously consider the Locator unreachable. An ITR SHOULD only set
2761 the E-bit in an encapsulated data packet when it knows the ETR is
2762 enabled for echo-noncing. This is conveyed by the E-bit in the
2765 Note that other Locator reachability mechanisms are being researched
2766 and can be used to compliment or even override the echo nonce
2767 algorithm. See the next section for an example of control-plane
2773 <span class="grey">Farinacci, et al. Experimental [Page 47]</span>
2774 </pre><!--NewPage--><pre class="newpage"><a name="page-48" id="page-48" href="#page-48" class="invisible"> </a>
2775 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2778 <span class="h4"><h4><a class="selflink" name="section-6.3.2" href="#section-6.3.2">6.3.2</a>. RLOC-Probing Algorithm</h4></span>
2780 RLOC-Probing is a method that an ITR or PITR can use to determine the
2781 reachability status of one or more Locators that it has cached in a
2782 Map-Cache entry. The probe-bit of the Map-Request and Map-Reply
2783 messages is used for RLOC-Probing.
2785 RLOC-Probing is done in the control plane on a timer basis, where an
2786 ITR or PITR will originate a Map-Request destined to a locator
2787 address from one of its own locator addresses. A Map-Request used as
2788 an RLOC-probe is NOT encapsulated and NOT sent to a Map-Server or to
2789 the mapping database system as one would when soliciting mapping
2790 data. The EID record encoded in the Map-Request is the EID-Prefix of
2791 the Map-Cache entry cached by the ITR or PITR. The ITR may include a
2792 mapping data record for its own database mapping information that
2793 contains the local EID-Prefixes and RLOCs for its site. RLOC-probes
2794 are sent periodically using a jittered timer interval.
2796 When an ETR receives a Map-Request message with the probe-bit set, it
2797 returns a Map-Reply with the probe-bit set. The source address of
2798 the Map-Reply is set according to the procedure described in
2799 <a href="#section-6.1.5">Section 6.1.5</a>. The Map-Reply SHOULD contain mapping data for the
2800 EID-Prefix contained in the Map-Request. This provides the
2801 opportunity for the ITR or PITR that sent the RLOC-probe to get
2802 mapping updates if there were changes to the ETR's database mapping
2805 There are advantages and disadvantages of RLOC-Probing. The greatest
2806 benefit of RLOC-Probing is that it can handle many failure scenarios
2807 allowing the ITR to determine when the path to a specific Locator is
2808 reachable or has become unreachable, thus providing a robust
2809 mechanism for switching to using another Locator from the cached
2810 Locator. RLOC-Probing can also provide rough Round-Trip Time (RTT)
2811 estimates between a pair of Locators, which can be useful for network
2812 management purposes as well as for selecting low delay paths. The
2813 major disadvantage of RLOC-Probing is in the number of control
2814 messages required and the amount of bandwidth used to obtain those
2815 benefits, especially if the requirement for failure detection times
2818 Continued research and testing will attempt to characterize the
2819 tradeoffs of failure detection times versus message overhead.</span>
2829 <span class="grey">Farinacci, et al. Experimental [Page 48]</span>
2830 </pre><!--NewPage--><pre class="newpage"><a name="page-49" id="page-49" href="#page-49" class="invisible"> </a>
2831 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2834 <span class="h3"><h3><a class="selflink" name="section-6.4" href="#section-6.4">6.4</a>. EID Reachability within a LISP Site</h3></span>
2836 A site may be multihomed using two or more ETRs. The hosts and
2837 infrastructure within a site will be addressed using one or more
2838 EID-Prefixes that are mapped to the RLOCs of the relevant ETRs in the
2839 mapping system. One possible failure mode is for an ETR to lose
2840 reachability to one or more of the EID-Prefixes within its own site.
2841 When this occurs when the ETR sends Map-Replies, it can clear the
2842 R-bit associated with its own Locator. And when the ETR is also an
2843 ITR, it can clear its Locator-Status-Bit in the encapsulation data
2846 It is recognized that there are no simple solutions to the site
2847 partitioning problem because it is hard to know which part of the
2848 EID-Prefix range is partitioned and which Locators can reach any
2849 sub-ranges of the EID-Prefixes. This problem is under investigation
2850 with the expectation that experiments will tell us more. Note that
2851 this is not a new problem introduced by the LISP architecture. The
2852 problem exists today when a multihomed site uses BGP to advertise its
2853 reachability upstream.
2855 <span class="h3"><h3><a class="selflink" name="section-6.5" href="#section-6.5">6.5</a>. Routing Locator Hashing</h3></span>
2857 When an ETR provides an EID-to-RLOC mapping in a Map-Reply message to
2858 a requesting ITR, the Locator-Set for the EID-Prefix may contain
2859 different Priority values for each locator address. When more than
2860 one best Priority Locator exists, the ITR can decide how to load-
2861 share traffic against the corresponding Locators.
2863 The following hash algorithm may be used by an ITR to select a
2864 Locator for a packet destined to an EID for the EID-to-RLOC mapping:
2866 1. Either a source and destination address hash or the traditional
2867 5-tuple hash can be used. The traditional 5-tuple hash includes
2868 the source and destination addresses; source and destination TCP,
2869 UDP, or Stream Control Transmission Protocol (SCTP) port numbers;
2870 and the IP protocol number field or IPv6 next-protocol fields of
2871 a packet that a host originates from within a LISP site. When a
2872 packet is not a TCP, UDP, or SCTP packet, the source and
2873 destination addresses only from the header are used to compute
2876 2. Take the hash value and divide it by the number of Locators
2877 stored in the Locator-Set for the EID-to-RLOC mapping.
2879 3. The remainder will yield a value of 0 to "number of Locators
2880 minus 1". Use the remainder to select the Locator in the
2885 <span class="grey">Farinacci, et al. Experimental [Page 49]</span>
2886 </pre><!--NewPage--><pre class="newpage"><a name="page-50" id="page-50" href="#page-50" class="invisible"> </a>
2887 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2890 Note that when a packet is LISP encapsulated, the source port number
2891 in the outer UDP header needs to be set. Selecting a hashed value
2892 allows core routers that are attached to Link Aggregation Groups
2893 (LAGs) to load-split the encapsulated packets across member links of
2894 such LAGs. Otherwise, core routers would see a single flow, since
2895 packets have a source address of the ITR, for packets that are
2896 originated by different EIDs at the source site. A suggested setting
2897 for the source port number computed by an ITR is a 5-tuple hash
2898 function on the inner header, as described above.
2900 Many core router implementations use a 5-tuple hash to decide how to
2901 balance packet load across members of a LAG. The 5-tuple hash
2902 includes the source and destination addresses of the packet and the
2903 source and destination ports when the protocol number in the packet
2904 is TCP or UDP. For this reason, UDP encoding is used for LISP
2907 <span class="h3"><h3><a class="selflink" name="section-6.6" href="#section-6.6">6.6</a>. Changing the Contents of EID-to-RLOC Mappings</h3></span>
2909 Since the LISP architecture uses a caching scheme to retrieve and
2910 store EID-to-RLOC mappings, the only way an ITR can get a more up-to-
2911 date mapping is to re-request the mapping. However, the ITRs do not
2912 know when the mappings change, and the ETRs do not keep track of
2913 which ITRs requested its mappings. For scalability reasons, we want
2914 to maintain this approach but need to provide a way for ETRs to
2915 change their mappings and inform the sites that are currently
2916 communicating with the ETR site using such mappings.
2918 When adding a new Locator record in lexicographic order to the end of
2919 a Locator-Set, it is easy to update mappings. We assume that new
2920 mappings will maintain the same Locator ordering as the old mapping
2921 but will just have new Locators appended to the end of the list. So,
2922 some ITRs can have a new mapping while other ITRs have only an old
2923 mapping that is used until they time out. When an ITR has only an
2924 old mapping but detects bits set in the Locator-Status-Bits that
2925 correspond to Locators beyond the list it has cached, it simply
2926 ignores them. However, this can only happen for locator addresses
2927 that are lexicographically greater than the locator addresses in the
2928 existing Locator-Set.
2930 When a Locator record is inserted in the middle of a Locator-Set, to
2931 maintain lexicographic order, the SMR procedure in <a href="#section-6.6.2">Section 6.6.2</a> is
2932 used to inform ITRs and PITRs of the new Locator-Status-Bit mappings.
2934 When a Locator record is removed from a Locator-Set, ITRs that have
2935 the mapping cached will not use the removed Locator because the xTRs
2936 will set the Locator-Status-Bit to 0. So, even if the Locator is in
2937 the list, it will not be used. For new mapping requests, the xTRs
2941 <span class="grey">Farinacci, et al. Experimental [Page 50]</span>
2942 </pre><!--NewPage--><pre class="newpage"><a name="page-51" id="page-51" href="#page-51" class="invisible"> </a>
2943 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
2946 can set the Locator AFI to 0 (indicating an unspecified address), as
2947 well as setting the corresponding Locator-Status-Bit to 0. This
2948 forces ITRs with old or new mappings to avoid using the removed
2951 If many changes occur to a mapping over a long period of time, one
2952 will find empty record slots in the middle of the Locator-Set and new
2953 records appended to the Locator-Set. At some point, it would be
2954 useful to compact the Locator-Set so the Locator-Status-Bit settings
2955 can be efficiently packed.
2957 We propose here three approaches for Locator-Set compaction: one
2958 operational mechanism and two protocol mechanisms. The operational
2959 approach uses a clock sweep method. The protocol approaches use the
2960 concept of Solicit-Map-Requests and Map-Versioning.
2962 <span class="h4"><h4><a class="selflink" name="section-6.6.1" href="#section-6.6.1">6.6.1</a>. Clock Sweep</h4></span>
2964 The clock sweep approach uses planning in advance and the use of
2965 count-down TTLs to time out mappings that have already been cached.
2966 The default setting for an EID-to-RLOC mapping TTL is 24 hours. So,
2967 there is a 24-hour window to time out old mappings. The following
2968 clock sweep procedure is used:
2970 1. 24 hours before a mapping change is to take effect, a network
2971 administrator configures the ETRs at a site to start the clock
2974 2. During the clock sweep window, ETRs continue to send Map-Reply
2975 messages with the current (unchanged) mapping records. The TTL
2976 for these mappings is set to 1 hour.
2978 3. 24 hours later, all previous cache entries will have timed out,
2979 and any active cache entries will time out within 1 hour. During
2980 this 1-hour window, the ETRs continue to send Map-Reply messages
2981 with the current (unchanged) mapping records with the TTL set to
2984 4. At the end of the 1-hour window, the ETRs will send Map-Reply
2985 messages with the new (changed) mapping records. So, any active
2986 caches can get the new mapping contents right away if not cached,
2987 or in 1 minute if they had the mapping cached. The new mappings
2988 are cached with a TTL equal to the TTL in the Map-Reply.
2997 <span class="grey">Farinacci, et al. Experimental [Page 51]</span>
2998 </pre><!--NewPage--><pre class="newpage"><a name="page-52" id="page-52" href="#page-52" class="invisible"> </a>
2999 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3002 <span class="h4"><h4><a class="selflink" name="section-6.6.2" href="#section-6.6.2">6.6.2</a>. Solicit-Map-Request (SMR)</h4></span>
3004 Soliciting a Map-Request is a selective way for ETRs, at the site
3005 where mappings change, to control the rate they receive requests for
3006 Map-Reply messages. SMRs are also used to tell remote ITRs to update
3007 the mappings they have cached.
3009 Since the ETRs don't keep track of remote ITRs that have cached their
3010 mappings, they do not know which ITRs need to have their mappings
3011 updated. As a result, an ETR will solicit Map-Requests (called an
3012 SMR message) from those sites to which it has been sending
3013 encapsulated data for the last minute. In particular, an ETR will
3014 send an SMR to an ITR to which it has recently sent encapsulated
3017 An SMR message is simply a bit set in a Map-Request message. An ITR
3018 or PITR will send a Map-Request when they receive an SMR message.
3019 Both the SMR sender and the Map-Request responder MUST rate-limit
3020 these messages. Rate-limiting can be implemented as a global rate-
3021 limiter or one rate-limiter per SMR destination.
3023 The following procedure shows how an SMR exchange occurs when a site
3024 is doing Locator-Set compaction for an EID-to-RLOC mapping:
3026 1. When the database mappings in an ETR change, the ETRs at the site
3027 begin to send Map-Requests with the SMR bit set for each Locator
3028 in each Map-Cache entry the ETR caches.
3030 2. A remote ITR that receives the SMR message will schedule sending
3031 a Map-Request message to the source locator address of the SMR
3032 message or to the mapping database system. A newly allocated
3033 random nonce is selected, and the EID-Prefix used is the one
3034 copied from the SMR message. If the source Locator is the only
3035 Locator in the cached Locator-Set, the remote ITR SHOULD send a
3036 Map-Request to the database mapping system just in case the
3037 single Locator has changed and may no longer be reachable to
3038 accept the Map-Request.
3040 3. The remote ITR MUST rate-limit the Map-Request until it gets a
3041 Map-Reply while continuing to use the cached mapping. When
3042 Map-Versioning as described in <a href="#section-6.6.3">Section 6.6.3</a> is used, an SMR
3043 sender can detect if an ITR is using the most up-to-date database
3046 4. The ETRs at the site with the changed mapping will reply to the
3047 Map-Request with a Map-Reply message that has a nonce from the
3048 SMR-invoked Map-Request. The Map-Reply messages SHOULD be rate-
3049 limited. This is important to avoid Map-Reply implosion.
3053 <span class="grey">Farinacci, et al. Experimental [Page 52]</span>
3054 </pre><!--NewPage--><pre class="newpage"><a name="page-53" id="page-53" href="#page-53" class="invisible"> </a>
3055 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3058 5. The ETRs at the site with the changed mapping record the fact
3059 that the site that sent the Map-Request has received the new
3060 mapping data in the Map-Cache entry for the remote site so the
3061 Locator-Status-Bits are reflective of the new mapping for packets
3062 going to the remote site. The ETR then stops sending SMR
3065 Experimentation is in progress to determine the appropriate rate-
3068 For security reasons, an ITR MUST NOT process unsolicited
3069 Map-Replies. To avoid Map-Cache entry corruption by a third party, a
3070 sender of an SMR-based Map-Request MUST be verified. If an ITR
3071 receives an SMR-based Map-Request and the source is not in the
3072 Locator-Set for the stored Map-Cache entry, then the responding
3073 Map-Request MUST be sent with an EID destination to the mapping
3074 database system. Since the mapping database system is a more secure
3075 way to reach an authoritative ETR, it will deliver the Map-Request to
3076 the authoritative source of the mapping data.
3078 When an ITR receives an SMR-based Map-Request for which it does not
3079 have a cached mapping for the EID in the SMR message, it MAY not send
3080 an SMR-invoked Map-Request. This scenario can occur when an ETR
3081 sends SMR messages to all Locators in the Locator-Set it has stored
3082 in its map-cache but the remote ITRs that receive the SMR may not be
3083 sending packets to the site. There is no point in updating the ITRs
3084 until they need to send, in which case they will send Map-Requests to
3085 obtain a Map-Cache entry.
3087 <span class="h4"><h4><a class="selflink" name="section-6.6.3" href="#section-6.6.3">6.6.3</a>. Database Map-Versioning</h4></span>
3089 When there is unidirectional packet flow between an ITR and ETR, and
3090 the EID-to-RLOC mappings change on the ETR, it needs to inform the
3091 ITR so encapsulation to a removed Locator can stop and can instead be
3092 started to a new Locator in the Locator-Set.
3094 An ETR, when it sends Map-Reply messages, conveys its own Map-Version
3095 Number. This is known as the Destination Map-Version Number. ITRs
3096 include the Destination Map-Version Number in packets they
3097 encapsulate to the site. When an ETR decapsulates a packet and
3098 detects that the Destination Map-Version Number is less than the
3099 current version for its mapping, the SMR procedure described in
3100 <a href="#section-6.6.2">Section 6.6.2</a> occurs.
3109 <span class="grey">Farinacci, et al. Experimental [Page 53]</span>
3110 </pre><!--NewPage--><pre class="newpage"><a name="page-54" id="page-54" href="#page-54" class="invisible"> </a>
3111 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3114 An ITR, when it encapsulates packets to ETRs, can convey its own
3115 Map-Version Number. This is known as the Source Map-Version Number.
3116 When an ETR decapsulates a packet and detects that the Source
3117 Map-Version Number is greater than the last Map-Version Number sent
3118 in a Map-Reply from the ITR's site, the ETR will send a Map-Request
3119 to one of the ETRs for the source site.
3121 A Map-Version Number is used as a sequence number per EID-Prefix, so
3122 values that are greater are considered to be more recent. A value of
3123 0 for the Source Map-Version Number or the Destination Map-Version
3124 Number conveys no versioning information, and an ITR does no
3125 comparison with previously received Map-Version Numbers.
3127 A Map-Version Number can be included in Map-Register messages as
3128 well. This is a good way for the Map-Server to assure that all ETRs
3129 for a site registering to it will be synchronized according to
3132 See [<a href="http://tools.ietf.org/html/rfc6834" title=""Locator/ID Separation Protocol (LISP) Map-Versioning"">RFC6834</a>] for a more detailed analysis and description of
3133 Database Map-Versioning.
3135 <span class="h2"><h2><a class="selflink" name="section-7" href="#section-7">7</a>. Router Performance Considerations</h2></span>
3137 LISP is designed to be very "hardware-based forwarding friendly". A
3138 few implementation techniques can be used to incrementally implement
3141 o When a tunnel-encapsulated packet is received by an ETR, the outer
3142 destination address may not be the address of the router. This
3143 makes it challenging for the control plane to get packets from the
3144 hardware. This may be mitigated by creating special Forwarding
3145 Information Base (FIB) entries for the EID-Prefixes of EIDs served
3146 by the ETR (those for which the router provides an RLOC
3147 translation). These FIB entries are marked with a flag indicating
3148 that control-plane processing should be performed. The forwarding
3149 logic of testing for particular IP protocol number values is not
3150 necessary. There are a few proven cases where no changes to
3151 existing deployed hardware were needed to support the LISP data-
3154 o On an ITR, prepending a new IP header consists of adding more
3155 octets to a MAC rewrite string and prepending the string as part
3156 of the outgoing encapsulation procedure. Routers that support
3157 Generic Routing Encapsulation (GRE) tunneling [<a href="http://tools.ietf.org/html/rfc2784" title=""Generic Routing Encapsulation (GRE)"">RFC2784</a>] or 6to4
3158 tunneling [<a href="http://tools.ietf.org/html/rfc3056" title=""Connection of IPv6 Domains via IPv4 Clouds"">RFC3056</a>] may already support this action.
3165 <span class="grey">Farinacci, et al. Experimental [Page 54]</span>
3166 </pre><!--NewPage--><pre class="newpage"><a name="page-55" id="page-55" href="#page-55" class="invisible"> </a>
3167 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3170 o A packet's source address or interface the packet was received on
3171 can be used to select VRF (Virtual Routing/Forwarding). The VRF's
3172 routing table can be used to find EID-to-RLOC mappings.
3174 For performance issues related to map-cache management, see
3175 <a href="#section-12">Section 12</a>.
3177 <span class="h2"><h2><a class="selflink" name="section-8" href="#section-8">8</a>. Deployment Scenarios</h2></span>
3179 This section will explore how and where ITRs and ETRs can be deployed
3180 and will discuss the pros and cons of each deployment scenario. For
3181 a more detailed deployment recommendation, refer to [<a href="#ref-LISP-DEPLOY">LISP-DEPLOY</a>].
3183 There are two basic deployment tradeoffs to consider: centralized
3184 versus distributed caches; and flat, Recursive, or Re-encapsulating
3185 Tunneling. When deciding on centralized versus distributed caching,
3186 the following issues should be considered:
3188 o Are the Tunnel Routers spread out so that the caches are spread
3189 across all the memories of each router? A centralized cache is
3190 when an ITR keeps a cache for all the EIDs it is encapsulating to.
3191 The packet takes a direct path to the destination Locator. A
3192 distributed cache is when an ITR needs help from other
3193 re-encapsulating routers because it does not store all the cache
3194 entries for the EIDs it is encapsulating to. So, the packet takes
3195 a path through re-encapsulating routers that have a different set
3198 o Should management "touch points" be minimized by only choosing a
3199 few Tunnel Routers, just enough for redundancy?
3201 o In general, using more ITRs doesn't increase management load,
3202 since caches are built and stored dynamically. On the other hand,
3203 using more ETRs does require more management, since EID-Prefix-to-
3204 RLOC mappings need to be explicitly configured.
3206 When deciding on flat, Recursive, or Re-encapsulating Tunneling, the
3207 following issues should be considered:
3209 o Flat tunneling implements a single tunnel between the source site
3210 and destination site. This generally offers better paths between
3211 sources and destinations with a single tunnel path.
3213 o Recursive Tunneling is when tunneled traffic is again further
3214 encapsulated in another tunnel, either to implement VPNs or to
3215 perform Traffic Engineering. When doing VPN-based tunneling, the
3216 site has some control, since the site is prepending a new tunnel
3217 header. In the case of TE-based tunneling, the site may have
3221 <span class="grey">Farinacci, et al. Experimental [Page 55]</span>
3222 </pre><!--NewPage--><pre class="newpage"><a name="page-56" id="page-56" href="#page-56" class="invisible"> </a>
3223 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3226 control if it is prepending a new tunnel header, but if the site's
3227 ISP is doing the TE, then the site has no control. Recursive
3228 Tunneling generally will result in suboptimal paths but with the
3229 benefit of steering traffic to parts of the network that have more
3230 resources available.
3232 o The technique of re-encapsulation ensures that packets only
3233 require one tunnel header. So, if a packet needs to be re-routed,
3234 it is first decapsulated by the ETR and then re-encapsulated with
3235 a new tunnel header using a new RLOC.
3237 The next sub-sections will examine where Tunnel Routers can reside in
3240 <span class="h3"><h3><a class="selflink" name="section-8.1" href="#section-8.1">8.1</a>. First-Hop/Last-Hop Tunnel Routers</h3></span>
3242 By locating Tunnel Routers close to hosts, the EID-Prefix set is at
3243 the granularity of an IP subnet. So, at the expense of more
3244 EID-Prefix-to-RLOC sets for the site, the caches in each Tunnel
3245 Router can remain relatively small. But caches always depend on the
3246 number of non-aggregated EID destination flows active through these
3249 With more Tunnel Routers doing encapsulation, the increase in control
3250 traffic grows as well: since the EID granularity is greater, more
3251 Map-Requests and Map-Replies are traveling between more routers.
3253 The advantage of placing the caches and databases at these stub
3254 routers is that the products deployed in this part of the network
3255 have better price-memory ratios than their core router counterparts.
3256 Memory is typically less expensive in these devices, and fewer routes
3257 are stored (only IGP routes). These devices tend to have excess
3258 capacity, both for forwarding and routing states.
3260 LISP functionality can also be deployed in edge switches. These
3261 devices generally have layer-2 ports facing hosts and layer-3 ports
3262 facing the Internet. Spare capacity is also often available in these
3265 <span class="h3"><h3><a class="selflink" name="section-8.2" href="#section-8.2">8.2</a>. Border/Edge Tunnel Routers</h3></span>
3267 Using Customer Edge (CE) routers for tunnel endpoints allows the EID
3268 space associated with a site to be reachable via a small set of RLOCs
3269 assigned to the CE routers for that site. This is the default
3270 behavior envisioned in the rest of this specification.
3277 <span class="grey">Farinacci, et al. Experimental [Page 56]</span>
3278 </pre><!--NewPage--><pre class="newpage"><a name="page-57" id="page-57" href="#page-57" class="invisible"> </a>
3279 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3282 This offers the opposite benefit of the first-hop/last-hop Tunnel
3283 Router scenario: the number of mapping entries and network management
3284 touch points is reduced, allowing better scaling.
3286 One disadvantage is that fewer network resources are used to reach
3287 host endpoints, thereby centralizing the point-of-failure domain and
3288 creating network choke points at the CE router.
3290 Note that more than one CE router at a site can be configured with
3291 the same IP address. In this case, an RLOC is an anycast address.
3292 This allows resilience between the CE routers. That is, if a CE
3293 router fails, traffic is automatically routed to the other routers
3294 using the same anycast address. However, this comes with the
3295 disadvantage where the site cannot control the entrance point when
3296 the anycast route is advertised out from all border routers. Another
3297 disadvantage of using anycast Locators is the limited advertisement
3298 scope of /32 (or /128 for IPv6) routes.
3300 <span class="h3"><h3><a class="selflink" name="section-8.3" href="#section-8.3">8.3</a>. ISP Provider Edge (PE) Tunnel Routers</h3></span>
3302 The use of ISP PE routers as tunnel endpoint routers is not the
3303 typical deployment scenario envisioned in this specification. This
3304 section attempts to capture some of the reasoning behind this
3305 preference for implementing LISP on CE routers.
3307 The use of ISP PE routers as tunnel endpoint routers gives an ISP,
3308 rather than a site, control over the location of the egress tunnel
3309 endpoints. That is, the ISP can decide whether the tunnel endpoints
3310 are in the destination site (in either CE routers or last-hop routers
3311 within a site) or at other PE edges. The advantage of this case is
3312 that two tunnel headers can be avoided. By having the PE be the
3313 first router on the path to encapsulate, it can choose a TE path
3314 first, and the ETR can decapsulate and re-encapsulate for a tunnel to
3315 the destination end site.
3317 An obvious disadvantage is that the end site has no control over
3318 where its packets flow or over the RLOCs used. Other disadvantages
3319 include difficulty in synchronizing path liveness updates between CE
3322 As mentioned in earlier sections, a combination of these scenarios is
3323 possible at the expense of extra packet header overhead; if both site
3324 and provider want control, then Recursive or Re-encapsulating Tunnels
3333 <span class="grey">Farinacci, et al. Experimental [Page 57]</span>
3334 </pre><!--NewPage--><pre class="newpage"><a name="page-58" id="page-58" href="#page-58" class="invisible"> </a>
3335 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3338 <span class="h3"><h3><a class="selflink" name="section-8.4" href="#section-8.4">8.4</a>. LISP Functionality with Conventional NATs</h3></span>
3340 LISP routers can be deployed behind Network Address Translator (NAT)
3341 devices to provide the same set of packet services hosts have today
3342 when they are addressed out of private address space.
3344 It is important to note that a locator address in any LISP control
3345 message MUST be a globally routable address and therefore SHOULD NOT
3346 contain [<a href="http://tools.ietf.org/html/rfc1918" title=""Address Allocation for Private Internets"">RFC1918</a>] addresses. If a LISP router is configured with
3347 private addresses, they MUST be used only in the outer IP header so
3348 the NAT device can translate properly. Otherwise, EID addresses MUST
3349 be translated before encapsulation is performed. Both NAT
3350 translation and LISP encapsulation functions could be co-located in
3353 More details on LISP address translation can be found in [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>].
3355 <span class="h3"><h3><a class="selflink" name="section-8.5" href="#section-8.5">8.5</a>. Packets Egressing a LISP Site</h3></span>
3357 When a LISP site is using two ITRs for redundancy, the failure of one
3358 ITR will likely shift outbound traffic to the second. This second
3359 ITR's cache may not be populated with the same EID-to-RLOC mapping
3360 entries as the first. If this second ITR does not have these
3361 mappings, traffic will be dropped while the mappings are retrieved
3362 from the mapping system. The retrieval of these messages may
3363 increase the load of requests being sent into the mapping system.
3364 Deployment and experimentation will determine whether this issue
3365 requires more attention.
3367 <span class="h2"><h2><a class="selflink" name="section-9" href="#section-9">9</a>. Traceroute Considerations</h2></span>
3369 When a source host in a LISP site initiates a traceroute to a
3370 destination host in another LISP site, it is highly desirable for it
3371 to see the entire path. Since packets are encapsulated from the ITR
3372 to the ETR, the hop across the tunnel could be viewed as a single
3373 hop. However, LISP traceroute will provide the entire path so the
3374 user can see 3 distinct segments of the path from a source LISP host
3375 to a destination LISP host:
3389 <span class="grey">Farinacci, et al. Experimental [Page 58]</span>
3390 </pre><!--NewPage--><pre class="newpage"><a name="page-59" id="page-59" href="#page-59" class="invisible"> </a>
3391 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3394 Segment 1 (in source LISP site based on EIDs):
3396 source host ---> first hop ... next hop ---> ITR
3398 Segment 2 (in the core network based on RLOCs):
3400 ITR ---> next hop ... next hop ---> ETR
3402 Segment 3 (in the destination LISP site based on EIDs):
3404 ETR ---> next hop ... last hop ---> destination host
3406 For segment 1 of the path, ICMP Time Exceeded messages are returned
3407 in the normal manner as they are today. The ITR performs a TTL
3408 decrement and tests for 0 before encapsulating. Therefore, the ITR's
3409 hop is seen by the traceroute source as having an EID address (the
3410 address of the site-facing interface).
3412 For segment 2 of the path, ICMP Time Exceeded messages are returned
3413 to the ITR because the TTL decrement to 0 is done on the outer
3414 header, so the destinations of the ICMP messages are the ITR RLOC
3415 address and the source RLOC address of the encapsulated traceroute
3416 packet. The ITR looks inside of the ICMP payload to inspect the
3417 traceroute source so it can return the ICMP message to the address of
3418 the traceroute client and also retain the core router IP address in
3419 the ICMP message. This is so the traceroute client can display the
3420 core router address (the RLOC address) in the traceroute output. The
3421 ETR returns its RLOC address and responds to the TTL decrement to 0,
3422 as the previous core routers did.
3424 For segment 3, the next-hop router downstream from the ETR will be
3425 decrementing the TTL for the packet that was encapsulated, sent into
3426 the core, decapsulated by the ETR, and forwarded because it isn't the
3427 final destination. If the TTL is decremented to 0, any router on the
3428 path to the destination of the traceroute, including the next-hop
3429 router or destination, will send an ICMP Time Exceeded message to the
3430 source EID of the traceroute client. The ICMP message will be
3431 encapsulated by the local ITR and sent back to the ETR in the
3432 originated traceroute source site, where the packet will be delivered
3435 <span class="h3"><h3><a class="selflink" name="section-9.1" href="#section-9.1">9.1</a>. IPv6 Traceroute</h3></span>
3437 IPv6 traceroute follows the procedure described above, since the
3438 entire traceroute data packet is included in the ICMP Time Exceeded
3439 message payload. Therefore, only the ITR needs to pay special
3440 attention to forwarding ICMP messages back to the traceroute source.
3445 <span class="grey">Farinacci, et al. Experimental [Page 59]</span>
3446 </pre><!--NewPage--><pre class="newpage"><a name="page-60" id="page-60" href="#page-60" class="invisible"> </a>
3447 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3450 <span class="h3"><h3><a class="selflink" name="section-9.2" href="#section-9.2">9.2</a>. IPv4 Traceroute</h3></span>
3452 For IPv4 traceroute, we cannot follow the above procedure, since IPv4
3453 ICMP Time Exceeded messages only include the invoking IP header and
3454 8 octets that follow the IP header. Therefore, when a core router
3455 sends an IPv4 Time Exceeded message to an ITR, all the ITR has in the
3456 ICMP payload is the encapsulated header it prepended, followed by a
3457 UDP header. The original invoking IP header, and therefore the
3458 identity of the traceroute source, is lost.
3460 The solution we propose to solve this problem is to cache traceroute
3461 IPv4 headers in the ITR and to match them up with corresponding IPv4
3462 Time Exceeded messages received from core routers and the ETR. The
3463 ITR will use a circular buffer for caching the IPv4 and UDP headers
3464 of traceroute packets. It will select a 16-bit number as a key to
3465 find them later when the IPv4 Time Exceeded messages are received.
3466 When an ITR encapsulates an IPv4 traceroute packet, it will use the
3467 16-bit number as the UDP source port in the encapsulating header.
3468 When the ICMP Time Exceeded message is returned to the ITR, the UDP
3469 header of the encapsulating header is present in the ICMP payload,
3470 thereby allowing the ITR to find the cached headers for the
3471 traceroute source. The ITR puts the cached headers in the payload
3472 and sends the ICMP Time Exceeded message to the traceroute source
3473 retaining the source address of the original ICMP Time Exceeded
3474 message (a core router or the ETR of the site of the traceroute
3477 The signature of a traceroute packet comes in two forms. The first
3478 form is encoded as a UDP message where the destination port is
3479 inspected for a range of values. The second form is encoded as an
3480 ICMP message where the IP identification field is inspected for a
3483 <span class="h3"><h3><a class="selflink" name="section-9.3" href="#section-9.3">9.3</a>. Traceroute Using Mixed Locators</h3></span>
3485 When either an IPv4 traceroute or IPv6 traceroute is originated and
3486 the ITR encapsulates it in the other address family header, one
3487 cannot get all 3 segments of the traceroute. Segment 2 of the
3488 traceroute cannot be conveyed to the traceroute source, since it is
3489 expecting addresses from intermediate hops in the same address format
3490 for the type of traceroute it originated. Therefore, in this case,
3491 segment 2 will make the tunnel look like one hop. All the ITR has to
3492 do to make this work is to not copy the inner TTL to the outer,
3493 encapsulating header's TTL when a traceroute packet is encapsulated
3494 using an RLOC from a different address family. This will cause no
3495 TTL decrement to 0 to occur in core routers between the ITR and ETR.
3501 <span class="grey">Farinacci, et al. Experimental [Page 60]</span>
3502 </pre><!--NewPage--><pre class="newpage"><a name="page-61" id="page-61" href="#page-61" class="invisible"> </a>
3503 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3506 <span class="h2"><h2><a class="selflink" name="section-10" href="#section-10">10</a>. Mobility Considerations</h2></span>
3508 There are several kinds of mobility, of which only some might be of
3509 concern to LISP. Essentially, they are as follows.
3511 <span class="h3"><h3><a class="selflink" name="section-10.1" href="#section-10.1">10.1</a>. Site Mobility</h3></span>
3513 A site wishes to change its attachment points to the Internet, and
3514 its LISP Tunnel Routers will have new RLOCs when it changes upstream
3515 providers. Changes in EID-to-RLOC mappings for sites are expected to
3516 be handled by configuration, outside of LISP.
3518 <span class="h3"><h3><a class="selflink" name="section-10.2" href="#section-10.2">10.2</a>. Slow Endpoint Mobility</h3></span>
3520 An individual endpoint wishes to move but is not concerned about
3521 maintaining session continuity. Renumbering is involved. LISP can
3522 help with the issues surrounding renumbering [<a href="http://tools.ietf.org/html/rfc4192" title=""Procedures for Renumbering an IPv6 Network without a Flag Day"">RFC4192</a>] [<a href="#ref-LISA96" title=""Renumbering: Threat or Menace?"">LISA96</a>] by
3523 decoupling the address space used by a site from the address spaces
3524 used by its ISPs [<a href="http://tools.ietf.org/html/rfc4984" title=""Report from the IAB Workshop on Routing and Addressing"">RFC4984</a>].
3526 <span class="h3"><h3><a class="selflink" name="section-10.3" href="#section-10.3">10.3</a>. Fast Endpoint Mobility</h3></span>
3528 Fast endpoint mobility occurs when an endpoint moves relatively
3529 rapidly, changing its IP-layer network attachment point. Maintenance
3530 of session continuity is a goal. This is where the Mobile IPv4
3531 [<a href="http://tools.ietf.org/html/rfc5944" title=""IP Mobility Support for IPv4, Revised"">RFC5944</a>] and Mobile IPv6 [<a href="http://tools.ietf.org/html/rfc6275" title=""Mobility Support in IPv6"">RFC6275</a>] [<a href="http://tools.ietf.org/html/rfc4866" title=""Enhanced Route Optimization for Mobile IPv6"">RFC4866</a>] mechanisms are used and
3532 primarily where interactions with LISP need to be explored.
3534 The problem is that as an endpoint moves, it may require changes to
3535 the mapping between its EID and a set of RLOCs for its new network
3536 location. When this is added to the overhead of Mobile IP binding
3537 updates, some packets might be delayed or dropped.
3539 In IPv4 mobility, when an endpoint is away from home, packets to it
3540 are encapsulated and forwarded via a home agent that resides in the
3541 home area the endpoint's address belongs to. The home agent will
3542 encapsulate and forward packets either directly to the endpoint or to
3543 a foreign agent that resides where the endpoint has moved to.
3544 Packets from the endpoint may be sent directly to the correspondent
3545 node, may be sent via the foreign agent, or may be reverse-tunneled
3546 back to the home agent for delivery to the mobile node. As the
3547 mobile node's EID or available RLOC changes, LISP EID-to-RLOC
3557 <span class="grey">Farinacci, et al. Experimental [Page 61]</span>
3558 </pre><!--NewPage--><pre class="newpage"><a name="page-62" id="page-62" href="#page-62" class="invisible"> </a>
3559 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3562 mappings are required for communication between the mobile node and
3563 the home agent, whether via the foreign agent or not. As a mobile
3564 endpoint changes networks, up to three LISP mapping changes may be
3567 o The mobile node moves from an old location to a new visited
3568 network location and notifies its home agent that it has done so.
3569 The Mobile IPv4 control packets the mobile node sends pass through
3570 one of the new visited network's ITRs, which needs an EID-to-RLOC
3571 mapping for the home agent.
3573 o The home agent might not have the EID-to-RLOC mappings for the
3574 mobile node's "care-of" address or its foreign agent in the new
3575 visited network, in which case it will need to acquire them.
3577 o When packets are sent directly to the correspondent node, it may
3578 be that no traffic has been sent from the new visited network to
3579 the correspondent node's network, and the new visited network's
3580 ITR will need to obtain an EID-to-RLOC mapping for the
3581 correspondent node's site.
3583 In addition, if the IPv4 endpoint is sending packets from the new
3584 visited network using its original EID, then LISP will need to
3585 perform a route-returnability check on the new EID-to-RLOC mapping
3588 In IPv6 mobility, packets can flow directly between the mobile node
3589 and the correspondent node in either direction. The mobile node uses
3590 its "care-of" address (EID). In this case, the route-returnability
3591 check would not be needed but one more LISP mapping lookup may be
3594 o As above, three mapping changes may be needed for the mobile node
3595 to communicate with its home agent and to send packets to the
3598 o In addition, another mapping will be needed in the correspondent
3599 node's ITR, in order for the correspondent node to send packets to
3600 the mobile node's "care-of" address (EID) at the new network
3603 When both endpoints are mobile, the number of potential mapping
3604 lookups increases accordingly.
3606 As a mobile node moves, there are not only mobility state changes in
3607 the mobile node, correspondent node, and home agent, but also state
3608 changes in the ITRs and ETRs for at least some EID-Prefixes.
3613 <span class="grey">Farinacci, et al. Experimental [Page 62]</span>
3614 </pre><!--NewPage--><pre class="newpage"><a name="page-63" id="page-63" href="#page-63" class="invisible"> </a>
3615 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3618 The goal is to support rapid adaptation, with little delay or packet
3619 loss for the entire system. Also, IP mobility can be modified to
3620 require fewer mapping changes. In order to increase overall system
3621 performance, there may be a need to reduce the optimization of one
3622 area in order to place fewer demands on another.
3624 In LISP, one possibility is to "glean" information. When a packet
3625 arrives, the ETR could examine the EID-to-RLOC mapping and use that
3626 mapping for all outgoing traffic to that EID. It can do this after
3627 performing a route-returnability check, to ensure that the new
3628 network location does have an internal route to that endpoint.
3629 However, this does not cover the case where an ITR (the node assigned
3630 the RLOC) at the mobile-node location has been compromised.
3632 Mobile IP packet exchange is designed for an environment in which all
3633 routing information is disseminated before packets can be forwarded.
3634 In order to allow the Internet to grow to support expected future
3635 use, we are moving to an environment where some information may have
3636 to be obtained after packets are in flight. Modifications to IP
3637 mobility should be considered in order to optimize the behavior of
3638 the overall system. Anything that decreases the number of new
3639 EID-to-RLOC mappings needed when a node moves, or maintains the
3640 validity of an EID-to-RLOC mapping for a longer time, is useful.
3642 <span class="h3"><h3><a class="selflink" name="section-10.4" href="#section-10.4">10.4</a>. Fast Network Mobility</h3></span>
3644 In addition to endpoints, a network can be mobile, possibly changing
3645 xTRs. A "network" can be as small as a single router and as large as
3646 a whole site. This is different from site mobility in that it is
3647 fast and possibly short-lived, but different from endpoint mobility
3648 in that a whole prefix is changing RLOCs. However, the mechanisms
3649 are the same, and there is no new overhead in LISP. A map request
3650 for any endpoint will return a binding for the entire mobile prefix.
3652 If mobile networks become a more common occurrence, it may be useful
3653 to revisit the design of the mapping service and allow for dynamic
3654 updates of the database.
3656 The issue of interactions between mobility and LISP needs to be
3657 explored further. Specific improvements to the entire system will
3658 depend on the details of mapping mechanisms. Mapping mechanisms
3659 should be evaluated on how well they support session continuity for
3669 <span class="grey">Farinacci, et al. Experimental [Page 63]</span>
3670 </pre><!--NewPage--><pre class="newpage"><a name="page-64" id="page-64" href="#page-64" class="invisible"> </a>
3671 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3674 <span class="h3"><h3><a class="selflink" name="section-10.5" href="#section-10.5">10.5</a>. LISP Mobile Node Mobility</h3></span>
3676 A mobile device can use the LISP infrastructure to achieve mobility
3677 by implementing the LISP encapsulation and decapsulation functions
3678 and acting as a simple ITR/ETR. By doing this, such a "LISP mobile
3679 node" can use topologically independent EID IP addresses that are not
3680 advertised into and do not impose a cost on the global routing
3681 system. These EIDs are maintained at the edges of the mapping system
3682 (in LISP Map-Servers and Map-Resolvers) and are provided on demand to
3683 only the correspondents of the LISP mobile node.
3685 Refer to [<a href="#ref-LISP-MN" title=""LISP Mobile Node"">LISP-MN</a>] for more details.
3687 <span class="h2"><h2><a class="selflink" name="section-11" href="#section-11">11</a>. Multicast Considerations</h2></span>
3689 A multicast group address, as defined in the original Internet
3690 architecture, is an identifier of a grouping of topologically
3691 independent receiver host locations. The address encoding itself
3692 does not determine the location of the receiver(s). The multicast
3693 routing protocol, and the network-based state the protocol creates,
3694 determine where the receivers are located.
3696 In the context of LISP, a multicast group address is both an EID and
3697 a Routing Locator. Therefore, no specific semantic or action needs
3698 to be taken for a destination address, as it would appear in an IP
3699 header. Therefore, a group address that appears in an inner IP
3700 header built by a source host will be used as the destination EID.
3701 The outer IP header (the destination Routing Locator address),
3702 prepended by a LISP router, will use the same group address as the
3703 destination Routing Locator.
3705 Having said that, only the source EID and source Routing Locator need
3706 to be dealt with. Therefore, an ITR merely needs to put its own IP
3707 address in the source 'Routing Locator' field when prepending the
3708 outer IP header. This source Routing Locator address, like any other
3709 Routing Locator address, MUST be globally routable.
3711 Therefore, an EID-to-RLOC mapping does not need to be performed by an
3712 ITR when a received data packet is a multicast data packet or when
3713 processing a source-specific Join (either by IGMPv3 or PIM). But the
3714 source Routing Locator is decided by the multicast routing protocol
3715 in a receiver site. That is, an EID-to-RLOC translation is done at
3718 Another approach is to have the ITR not encapsulate a multicast
3719 packet and allow the packet built by the host to flow into the core
3720 even if the source address is allocated out of the EID namespace. If
3721 the RPF-Vector TLV [<a href="http://tools.ietf.org/html/rfc5496" title=""The Reverse Path Forwarding (RPF) Vector TLV"">RFC5496</a>] is used by PIM in the core, then core
3725 <span class="grey">Farinacci, et al. Experimental [Page 64]</span>
3726 </pre><!--NewPage--><pre class="newpage"><a name="page-65" id="page-65" href="#page-65" class="invisible"> </a>
3727 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3730 routers can RPF to the ITR (the locator address, which is injected
3731 into core routing) rather than the host source address (the EID
3732 address, which is not injected into core routing).
3734 To avoid any EID-based multicast state in the network core, the first
3735 approach is chosen for LISP-Multicast. Details for LISP-Multicast
3736 and interworking with non-LISP sites are described in [<a href="http://tools.ietf.org/html/rfc6831" title=""The Locator/ID Separation Protocol (LISP) for Multicast Environments"">RFC6831</a>] and
3737 [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>].
3739 <span class="h2"><h2><a class="selflink" name="section-12" href="#section-12">12</a>. Security Considerations</h2></span>
3741 It is believed that most of the security mechanisms will be part of
3742 the mapping database service when using control-plane procedures for
3743 obtaining EID-to-RLOC mappings. For data-plane-triggered mappings,
3744 as described in this specification, protection is provided against
3745 ETR spoofing by using route-returnability (see <a href="#section-3">Section 3</a>) mechanisms
3746 evidenced by the use of a 24-bit 'Nonce' field in the LISP
3747 encapsulation header and a 64-bit 'Nonce' field in the LISP control
3750 The nonce, coupled with the ITR accepting only solicited Map-Replies,
3751 provides a basic level of security, in many ways similar to the
3752 security experienced in the current Internet routing system. It is
3753 hard for off-path attackers to launch attacks against these LISP
3754 mechanisms, as they do not have the nonce values. Sending a large
3755 number of packets to accidentally find the right nonce value is
3756 possible but would already by itself be a denial-of-service (DoS)
3757 attack. On-path attackers can perform far more serious attacks, but
3758 on-path attackers can launch serious attacks in the current Internet
3759 as well, including eavesdropping, blocking, or redirecting traffic.
3760 See more discussion on this topic in <a href="#section-6.1.5.1">Section 6.1.5.1</a>.
3762 LISP does not rely on a PKI or a more heavyweight authentication
3763 system. These systems challenge one of the primary design goals of
3764 LISP -- scalability.
3766 DoS attack prevention will depend on implementations rate-limiting
3767 Map-Requests and Map-Replies to the control plane as well as
3768 rate-limiting the number of data-triggered Map-Replies.
3770 An incorrectly implemented or malicious ITR might choose to ignore
3771 the Priority and Weights provided by the ETR in its Map-Reply. This
3772 traffic-steering would be limited to the traffic that is sent by this
3773 ITR's site and no more severe than if the site initiated a bandwidth
3774 DoS attack on (one of) the ETR's ingress links. The ITR's site would
3775 typically gain no benefit from not respecting the Weights and would
3776 likely receive better service by abiding by them.
3781 <span class="grey">Farinacci, et al. Experimental [Page 65]</span>
3782 </pre><!--NewPage--><pre class="newpage"><a name="page-66" id="page-66" href="#page-66" class="invisible"> </a>
3783 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3786 To deal with map-cache exhaustion attempts in an ITR/PITR, the
3787 implementation should consider putting a maximum cap on the number of
3788 entries stored with a reserve list for special or frequently accessed
3789 sites. This should be a configuration policy control set by the
3790 network administrator who manages ITRs and PITRs. When overlapping
3791 EID-Prefixes occur across multiple Map-Cache entries, the integrity
3792 of the set must be wholly maintained. So, if a more-specific entry
3793 cannot be added due to reaching the maximum cap, then none of the
3794 less-specific entries should be stored in the map-cache.
3796 Given that the ITR/PITR maintains a cache of EID-to-RLOC mappings,
3797 cache sizing and maintenance are issues to be kept in mind during
3798 implementation. It is a good idea to have instrumentation in place
3799 to detect thrashing of the cache. Implementation experimentation
3800 will be used to determine which cache management strategies work
3801 best. In general, it is difficult to defend against cache-thrashing
3802 attacks. It should be noted that an undersized cache in an ITR/PITR
3803 not only causes adverse effects on the site or region it supports but
3804 may also cause increased Map-Request loads on the mapping system.
3806 "Piggybacked" mapping data as discussed in <a href="#section-6.1.3">Section 6.1.3</a> specifies
3807 how to handle such mappings and includes the possibility for an ETR
3808 to temporarily accept such a mapping before verification when running
3809 in "trusted" environments. In such cases, there is a potential
3810 threat that a fake mapping could be inserted (even if only for a
3811 short period) into a map-cache. As noted in <a href="#section-6.1.3">Section 6.1.3</a>, an ETR
3812 MUST be specifically configured to run in such a mode and might
3813 usefully only consider some specific ITRs as also running in that
3814 same trusted environment.
3816 There is a security risk implicit in the fact that ETRs generate the
3817 EID-Prefix to which they are responding. An ETR can claim a shorter
3818 prefix than it is actually responsible for. Various mechanisms to
3819 ameliorate or resolve this issue will be examined in the future
3820 [<a href="#ref-LISP-SEC" title=""LISP-Security (LISP-SEC)"">LISP-SEC</a>].
3822 Spoofing of inner-header addresses of LISP-encapsulated packets is
3823 possible, as with any tunneling mechanism. ITRs MUST verify the
3824 source address of a packet to be an EID that belongs to the site's
3825 EID-Prefix range prior to encapsulation. An ETR must only
3826 decapsulate and forward datagrams with an inner-header destination
3827 that matches one of its EID-Prefix ranges. If, upon receipt and
3828 decapsulation, the destination EID of a datagram does not match one
3829 of the ETR's configured EID-Prefixes, the ETR MUST drop the datagram.
3830 If a LISP-encapsulated packet arrives at an ETR, it SHOULD compare
3831 the inner-header source EID address and the outer-header source RLOC
3832 address with the mapping that exists in the mapping database. Then,
3837 <span class="grey">Farinacci, et al. Experimental [Page 66]</span>
3838 </pre><!--NewPage--><pre class="newpage"><a name="page-67" id="page-67" href="#page-67" class="invisible"> </a>
3839 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3842 when spoofing attacks occur, the outer-header source RLOC address can
3843 be used to trace back the attack to the source site, using existing
3846 This experimental specification does not address automated key
3847 management (AKM). <a href="http://tools.ietf.org/html/bcp107">BCP 107</a> [<a href="http://tools.ietf.org/html/rfc4107" title=""Guidelines for Cryptographic Key Management"">RFC4107</a>] provides guidance in this area.
3848 In addition, at the time of this writing, substantial work is being
3849 undertaken to improve security of the routing system [<a href="http://tools.ietf.org/html/rfc6518" title=""Keying and Authentication for Routing Protocols (KARP) Design Guidelines"">RFC6518</a>]
3850 [<a href="http://tools.ietf.org/html/rfc6480" title=""An Infrastructure to Support Secure Internet Routing"">RFC6480</a>] [<a href="#ref-BGP-SEC" title=""An Overview of BGPSEC"">BGP-SEC</a>] [<a href="#ref-LISP-SEC" title=""LISP-Security (LISP-SEC)"">LISP-SEC</a>]. Future work on LISP should address
3851 the issues discussed in <a href="http://tools.ietf.org/html/bcp107">BCP 107</a> as well as other open security
3852 considerations, which may require changes to this specification.
3854 <span class="h2"><h2><a class="selflink" name="section-13" href="#section-13">13</a>. Network Management Considerations</h2></span>
3856 Considerations for network management tools exist so the LISP
3857 protocol suite can be operationally managed. These mechanisms can be
3858 found in [<a href="#ref-LISP-MIB" title=""LISP MIB"">LISP-MIB</a>] and [<a href="http://tools.ietf.org/html/rfc6835" title=""The Locator/ID Separation Protocol Internet Groper (LIG)"">RFC6835</a>].
3860 <span class="h2"><h2><a class="selflink" name="section-14" href="#section-14">14</a>. IANA Considerations</h2></span>
3862 This section provides guidance to the Internet Assigned Numbers
3863 Authority (IANA) regarding registration of values related to the LISP
3864 specification, in accordance with <a href="http://tools.ietf.org/html/bcp26">BCP 26</a> [<a href="http://tools.ietf.org/html/rfc5226" title=""Guidelines for Writing an IANA Considerations Section in RFCs"">RFC5226</a>].
3866 There are four namespaces (listed in the sub-sections below) in LISP
3867 that have been registered.
3869 o LISP IANA registry allocations should not be made for purposes
3870 unrelated to LISP routing or transport protocols.
3872 o The following policies are used here with the meanings defined in
3873 <a href="http://tools.ietf.org/html/bcp26">BCP 26</a>: "Specification Required", "IETF Review", "Experimental
3874 Use", and "First Come First Served".
3876 <span class="h3"><h3><a class="selflink" name="section-14.1" href="#section-14.1">14.1</a>. LISP ACT and Flag Fields</h3></span>
3878 New ACT values (<a href="#section-6.1.4">Section 6.1.4</a>) can be allocated through IETF review
3879 or IESG approval. Four values have already been allocated by this
3880 specification (<a href="#section-6.1.4">Section 6.1.4</a>).
3882 In addition, LISP has a number of flag fields and reserved fields,
3883 such as the LISP header flags field (<a href="#section-5.3">Section 5.3</a>). New bits for
3884 flags in these fields can be implemented after IETF review or IESG
3885 approval, but these need not be managed by IANA.
3893 <span class="grey">Farinacci, et al. Experimental [Page 67]</span>
3894 </pre><!--NewPage--><pre class="newpage"><a name="page-68" id="page-68" href="#page-68" class="invisible"> </a>
3895 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3898 <span class="h3"><h3><a class="selflink" name="section-14.2" href="#section-14.2">14.2</a>. LISP Address Type Codes</h3></span>
3900 LISP Address [<a href="#ref-LCAF" title=""LISP Canonical Address Format (LCAF)"">LCAF</a>] type codes have a range from 0 to 255. New type
3901 codes MUST be allocated consecutively, starting at 0. Type Codes
3902 0-127 are to be assigned by IETF review or IESG approval.
3904 Type Codes 128-255 are available according to the [<a href="http://tools.ietf.org/html/rfc5226" title=""Guidelines for Writing an IANA Considerations Section in RFCs"">RFC5226</a>] First
3905 Come First Served policy.
3907 This registry, initially empty, is constructed for future use in
3908 experimental work related to LISP Canonical Address Format (LCAF)
3909 values. See [<a href="#ref-LCAF" title=""LISP Canonical Address Format (LCAF)"">LCAF</a>] for details of other possible unapproved address
3910 encodings. The unapproved LCAF encodings are an area for further
3911 study and experimentation.
3913 <span class="h3"><h3><a class="selflink" name="section-14.3" href="#section-14.3">14.3</a>. LISP UDP Port Numbers</h3></span>
3915 The IANA registry has allocated UDP port numbers 4341 and 4342 for
3916 lisp-data and lisp-control operation, respectively. IANA has updated
3917 the description for UDP ports 4341 and 4342 as follows:
3919 lisp-data 4341 udp LISP Data Packets
3920 lisp-control 4342 udp LISP Control Packets
3922 <span class="h3"><h3><a class="selflink" name="section-14.4" href="#section-14.4">14.4</a>. LISP Key ID Numbers</h3></span>
3924 The following Key ID values are defined by this specification as used
3925 in any packet type that references a 'Key ID' field:
3927 Name Number Defined in
3928 -----------------------------------------------
3930 HMAC-SHA-1-96 1 [<a href="http://tools.ietf.org/html/rfc2404" title=""The Use of HMAC-SHA-1-96 within ESP and AH"">RFC2404</a>]
3931 HMAC-SHA-256-128 2 [<a href="http://tools.ietf.org/html/rfc4868" title=""Using HMAC-SHA-256, HMAC-SHA-384, and HMAC-SHA-512 with IPsec"">RFC4868</a>]
3933 Number values are in the range of 0 to 65535. The allocation of
3934 values is on a first come first served basis.
3936 <span class="h2"><h2><a class="selflink" name="section-15" href="#section-15">15</a>. Known Open Issues and Areas of Future Work</h2></span>
3938 As an experimental specification, this work is, by definition,
3939 incomplete. Specific areas where additional experience and work are
3940 needed include the following:
3942 o At present, only [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>] is defined for implementing a database
3943 of EID-to-RLOC mapping information. Additional research on other
3944 mapping database systems is strongly encouraged.
3949 <span class="grey">Farinacci, et al. Experimental [Page 68]</span>
3950 </pre><!--NewPage--><pre class="newpage"><a name="page-69" id="page-69" href="#page-69" class="invisible"> </a>
3951 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
3954 o Failure and recovery of LISP site partitioning (see <a href="#section-6.4">Section 6.4</a>)
3955 in the presence of redundant configuration (see <a href="#section-8.5">Section 8.5</a>) needs
3956 further research and experimentation.
3958 o The characteristics of map-cache management under exceptional
3959 conditions, such as denial-of-service attacks, are not fully
3960 understood. Further experience is needed to determine whether
3961 current caching methods are practical or in need of further
3962 development. In particular, the performance, scaling, and
3963 security characteristics of the map-cache will be discovered as
3964 part of this experiment. Performance metrics to be observed are
3965 packet reordering associated with the LISP Data-Probe and loss of
3966 the first packet in a flow associated with map-caching. The
3967 impact of these upon TCP will be observed. See <a href="#section-12">Section 12</a> for
3968 additional thoughts and considerations.
3970 o Preliminary work has been done to ensure that sites employing LISP
3971 can interconnect with the rest of the Internet. This work is
3972 documented in [<a href="http://tools.ietf.org/html/rfc6832" title=""Interworking between Locator/ID Separation Protocol (LISP) and Non-LISP Sites"">RFC6832</a>], but further experimentation and
3973 experience are needed.
3975 o At present, no mechanism for automated key management for message
3976 authentication is defined. Addressing automated key management is
3977 necessary before this specification can be developed into a
3978 Standards Track RFC. See <a href="#section-12">Section 12</a> for further details regarding
3979 security considerations.
3981 o In order to maintain security and stability, Internet protocols
3982 typically isolate the control and data planes. Therefore, user
3983 activity cannot cause control-plane state to be created or
3984 destroyed. LISP does not maintain this separation. The degree to
3985 which the loss of separation impacts security and stability is a
3986 topic for experimental observation.
3988 o LISP allows for the use of different mapping database systems.
3989 While only one [<a href="http://tools.ietf.org/html/rfc6836" title=""Locator/ID Separation Protocol Alternative Logical Topology (LISP+ALT)"">RFC6836</a>] is currently well defined, each mapping
3990 database will likely have some impact on the security of the
3991 EID-to-RLOC mappings. How each mapping database system's security
3992 properties impact LISP overall is for further study.
3994 o An examination of the implications of LISP on Internet traffic,
3995 applications, routers, and security is needed. This will help
3996 implementors understand the consequences for network stability,
3997 routing protocol function, routing scalability, migration and
3998 backward compatibility, and implementation scalability (as
3999 influenced by additional protocol components; additional state;
4000 and additional processing for encapsulation, decapsulation, and
4005 <span class="grey">Farinacci, et al. Experimental [Page 69]</span>
4006 </pre><!--NewPage--><pre class="newpage"><a name="page-70" id="page-70" href="#page-70" class="invisible"> </a>
4007 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4010 o Experiments need to verify that LISP produces no significant
4011 change in the behavior of protocols run between end-systems over a
4012 LISP infrastructure versus being run directly between those same
4015 o Experiments need to verify that the issues raised in the Critique
4016 section of [<a href="http://tools.ietf.org/html/rfc6115" title=""Recommendation for a Routing Architecture"">RFC6115</a>] are either insignificant or have been
4017 addressed by updates to LISP.
4019 Other LISP documents may also include open issues and areas for
4022 <span class="h2"><h2><a class="selflink" name="section-16" href="#section-16">16</a>. References</h2></span>
4024 <span class="h3"><h3><a class="selflink" name="section-16.1" href="#section-16.1">16.1</a>. Normative References</h3></span>
4026 [<a name="ref-RFC0768" id="ref-RFC0768">RFC0768</a>] Postel, J., "User Datagram Protocol", STD 6, <a href="http://tools.ietf.org/html/rfc768">RFC 768</a>,
4029 [<a name="ref-RFC0791" id="ref-RFC0791">RFC0791</a>] Postel, J., "Internet Protocol", STD 5, <a href="http://tools.ietf.org/html/rfc791">RFC 791</a>,
4032 [<a name="ref-RFC1918" id="ref-RFC1918">RFC1918</a>] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and
4033 E. Lear, "Address Allocation for Private Internets",
4034 <a href="http://tools.ietf.org/html/bcp5">BCP 5</a>, <a href="http://tools.ietf.org/html/rfc1918">RFC 1918</a>, February 1996.
4036 [<a name="ref-RFC2119" id="ref-RFC2119">RFC2119</a>] Bradner, S., "Key words for use in RFCs to Indicate
4037 Requirement Levels", <a href="http://tools.ietf.org/html/bcp14">BCP 14</a>, <a href="http://tools.ietf.org/html/rfc2119">RFC 2119</a>, March 1997.
4039 [<a name="ref-RFC2404" id="ref-RFC2404">RFC2404</a>] Madson, C. and R. Glenn, "The Use of HMAC-SHA-1-96 within
4040 ESP and AH", <a href="http://tools.ietf.org/html/rfc2404">RFC 2404</a>, November 1998.
4042 [<a name="ref-RFC2460" id="ref-RFC2460">RFC2460</a>] Deering, S. and R. Hinden, "Internet Protocol, Version 6
4043 (IPv6) Specification", <a href="http://tools.ietf.org/html/rfc2460">RFC 2460</a>, December 1998.
4045 [<a name="ref-RFC3168" id="ref-RFC3168">RFC3168</a>] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition
4046 of Explicit Congestion Notification (ECN) to IP",
4047 <a href="http://tools.ietf.org/html/rfc3168">RFC 3168</a>, September 2001.
4049 [<a name="ref-RFC3232" id="ref-RFC3232">RFC3232</a>] Reynolds, J., "Assigned Numbers: <a href="http://tools.ietf.org/html/rfc1700">RFC 1700</a> is Replaced by
4050 an On-line Database", <a href="http://tools.ietf.org/html/rfc3232">RFC 3232</a>, January 2002.
4052 [<a name="ref-RFC4086" id="ref-RFC4086">RFC4086</a>] Eastlake, D., Schiller, J., and S. Crocker, "Randomness
4053 Requirements for Security", <a href="http://tools.ietf.org/html/bcp106">BCP 106</a>, <a href="http://tools.ietf.org/html/rfc4086">RFC 4086</a>, June 2005.
4055 [<a name="ref-RFC4632" id="ref-RFC4632">RFC4632</a>] Fuller, V. and T. Li, "Classless Inter-domain Routing
4056 (CIDR): The Internet Address Assignment and Aggregation
4057 Plan", <a href="http://tools.ietf.org/html/bcp122">BCP 122</a>, <a href="http://tools.ietf.org/html/rfc4632">RFC 4632</a>, August 2006.
4061 <span class="grey">Farinacci, et al. Experimental [Page 70]</span>
4062 </pre><!--NewPage--><pre class="newpage"><a name="page-71" id="page-71" href="#page-71" class="invisible"> </a>
4063 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4066 [<a name="ref-RFC4868" id="ref-RFC4868">RFC4868</a>] Kelly, S. and S. Frankel, "Using HMAC-SHA-256,
4067 HMAC-SHA-384, and HMAC-SHA-512 with IPsec", <a href="http://tools.ietf.org/html/rfc4868">RFC 4868</a>,
4070 [<a name="ref-RFC5226" id="ref-RFC5226">RFC5226</a>] Narten, T. and H. Alvestrand, "Guidelines for Writing an
4071 IANA Considerations Section in RFCs", <a href="http://tools.ietf.org/html/bcp26">BCP 26</a>, <a href="http://tools.ietf.org/html/rfc5226">RFC 5226</a>,
4074 [<a name="ref-RFC5496" id="ref-RFC5496">RFC5496</a>] Wijnands, IJ., Boers, A., and E. Rosen, "The Reverse Path
4075 Forwarding (RPF) Vector TLV", <a href="http://tools.ietf.org/html/rfc5496">RFC 5496</a>, March 2009.
4077 [<a name="ref-RFC5944" id="ref-RFC5944">RFC5944</a>] Perkins, C., "IP Mobility Support for IPv4, Revised",
4078 <a href="http://tools.ietf.org/html/rfc5944">RFC 5944</a>, November 2010.
4080 [<a name="ref-RFC6115" id="ref-RFC6115">RFC6115</a>] Li, T., "Recommendation for a Routing Architecture",
4081 <a href="http://tools.ietf.org/html/rfc6115">RFC 6115</a>, February 2011.
4083 [<a name="ref-RFC6275" id="ref-RFC6275">RFC6275</a>] Perkins, C., Johnson, D., and J. Arkko, "Mobility Support
4084 in IPv6", <a href="http://tools.ietf.org/html/rfc6275">RFC 6275</a>, July 2011.
4086 [<a name="ref-RFC6833" id="ref-RFC6833">RFC6833</a>] Farinacci, D. and V. Fuller, "Locator/ID Separation
4087 Protocol (LISP) Map-Server Interface", <a href="http://tools.ietf.org/html/rfc6833">RFC 6833</a>,
4090 [<a name="ref-RFC6834" id="ref-RFC6834">RFC6834</a>] Iannone, L., Saucez, D., and O. Bonaventure, "Locator/ID
4091 Separation Protocol (LISP) Map-Versioning", <a href="http://tools.ietf.org/html/rfc6834">RFC 6834</a>,
4094 [<a name="ref-RFC6836" id="ref-RFC6836">RFC6836</a>] Farinacci, D., Fuller, V., Meyer, D., and D. Lewis,
4095 "Locator/ID Separation Protocol Alternative Logical
4096 Topology (LISP+ALT)", <a href="http://tools.ietf.org/html/rfc6836">RFC 6836</a>, January 2013.
4098 <span class="h3"><h3><a class="selflink" name="section-16.2" href="#section-16.2">16.2</a>. Informative References</h3></span>
4100 [<a name="ref-AFI" id="ref-AFI">AFI</a>] IANA, "Address Family Numbers",
4101 <<a href="http://www.iana.org/assignments/address-family-numbers">http://www.iana.org/assignments/address-family-numbers</a>>.
4103 [<a name="ref-BGP-SEC" id="ref-BGP-SEC">BGP-SEC</a>] Lepinski, M. and S. Turner, <a style="text-decoration: none" href="http://www.google.com/search?sitesearch=tools.ietf.org%2Fhtml%2F&q=inurl:draft-+%22An+Overview+of+BGPSEC%22">"An Overview of BGPSEC"</a>, Work
4104 in Progress, May 2012.
4106 [<a name="ref-CHIAPPA" id="ref-CHIAPPA">CHIAPPA</a>] Chiappa, J., "Endpoints and Endpoint names: A Proposed
4107 Enhancement to the Internet Architecture", 1999,
4108 <<a href="http://mercury.lcs.mit.edu/%7Ejnc/tech/endpoints.txt">http://mercury.lcs.mit.edu/~jnc/tech/endpoints.txt</a>>.
4110 [<a name="ref-CONS" id="ref-CONS">CONS</a>] Brim, S., Chiappa, N., Farinacci, D., Fuller, V., Lewis,
4111 D., and D. Meyer, "LISP-CONS: A Content distribution
4112 Overlay Network Service for LISP", Work in Progress,
4117 <span class="grey">Farinacci, et al. Experimental [Page 71]</span>
4118 </pre><!--NewPage--><pre class="newpage"><a name="page-72" id="page-72" href="#page-72" class="invisible"> </a>
4119 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4122 [<a name="ref-EMACS" id="ref-EMACS">EMACS</a>] Brim, S., Farinacci, D., Meyer, D., and J. Curran, "EID
4123 Mappings Multicast Across Cooperating Systems for LISP",
4124 Work in Progress, November 2007.
4126 [<a name="ref-LCAF" id="ref-LCAF">LCAF</a>] Farinacci, D., Meyer, D., and J. Snijders, "LISP Canonical
4127 Address Format (LCAF)", Work in Progress, January 2013.
4129 [<a name="ref-LISA96" id="ref-LISA96">LISA96</a>] Lear, E., Tharp, D., Katinsky, J., and J. Coffin,
4130 "Renumbering: Threat or Menace?", Usenix Tenth System
4131 Administration Conference (LISA 96), October 1996.
4133 [<a name="ref-LISP-DEPLOY" id="ref-LISP-DEPLOY">LISP-DEPLOY</a>]
4134 Jakab, L., Cabellos-Aparicio, A., Coras, F.,
4135 Domingo-Pascual, J., and D. Lewis, "LISP Network Element
4136 Deployment Considerations", Work in Progress,
4139 [<a name="ref-LISP-MIB" id="ref-LISP-MIB">LISP-MIB</a>] Schudel, G., Jain, A., and V. Moreno, <a style="text-decoration: none" href="http://www.google.com/search?sitesearch=tools.ietf.org%2Fhtml%2F&q=inurl:draft-+%22LISP+MIB%22">"LISP MIB"</a>, Work
4140 in Progress, January 2013.
4142 [<a name="ref-LISP-MN" id="ref-LISP-MN">LISP-MN</a>] Farinacci, D., Lewis, D., Meyer, D., and C. White, "LISP
4143 Mobile Node", Work in Progress, October 2012.
4145 [<a name="ref-LISP-SEC" id="ref-LISP-SEC">LISP-SEC</a>] Maino, F., Ermagan, V., Cabellos, A., Saucez, D., and O.
4146 Bonaventure, "LISP-Security (LISP-SEC)", Work in Progress,
4149 [<a name="ref-LOC-ID-ARCH" id="ref-LOC-ID-ARCH">LOC-ID-ARCH</a>]
4150 Meyer, D. and D. Lewis, "Architectural Implications of
4151 Locator/ID Separation", Work in Progress, January 2009.
4153 [<a name="ref-OPENLISP" id="ref-OPENLISP">OPENLISP</a>] Iannone, L., Saucez, D., and O. Bonaventure, "OpenLISP
4154 Implementation Report", Work in Progress, July 2008.
4156 [<a name="ref-RADIR" id="ref-RADIR">RADIR</a>] Narten, T., <a style="text-decoration: none" href="http://www.google.com/search?sitesearch=tools.ietf.org%2Fhtml%2F&q=inurl:draft-+%22On+the+Scalability+of+Internet+Routing%22">"On the Scalability of Internet Routing"</a>, Work
4157 in Progress, February 2010.
4159 [<a name="ref-RFC1034" id="ref-RFC1034">RFC1034</a>] Mockapetris, P., "Domain names - concepts and facilities",
4160 STD 13, <a href="http://tools.ietf.org/html/rfc1034">RFC 1034</a>, November 1987.
4162 [<a name="ref-RFC2784" id="ref-RFC2784">RFC2784</a>] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P.
4163 Traina, "Generic Routing Encapsulation (GRE)", <a href="http://tools.ietf.org/html/rfc2784">RFC 2784</a>,
4166 [<a name="ref-RFC3056" id="ref-RFC3056">RFC3056</a>] Carpenter, B. and K. Moore, "Connection of IPv6 Domains
4167 via IPv4 Clouds", <a href="http://tools.ietf.org/html/rfc3056">RFC 3056</a>, February 2001.
4173 <span class="grey">Farinacci, et al. Experimental [Page 72]</span>
4174 </pre><!--NewPage--><pre class="newpage"><a name="page-73" id="page-73" href="#page-73" class="invisible"> </a>
4175 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4178 [<a name="ref-RFC3261" id="ref-RFC3261">RFC3261</a>] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston,
4179 A., Peterson, J., Sparks, R., Handley, M., and E.
4180 Schooler, "SIP: Session Initiation Protocol", <a href="http://tools.ietf.org/html/rfc3261">RFC 3261</a>,
4183 [<a name="ref-RFC4107" id="ref-RFC4107">RFC4107</a>] Bellovin, S. and R. Housley, "Guidelines for Cryptographic
4184 Key Management", <a href="http://tools.ietf.org/html/bcp107">BCP 107</a>, <a href="http://tools.ietf.org/html/rfc4107">RFC 4107</a>, June 2005.
4186 [<a name="ref-RFC4192" id="ref-RFC4192">RFC4192</a>] Baker, F., Lear, E., and R. Droms, "Procedures for
4187 Renumbering an IPv6 Network without a Flag Day", <a href="http://tools.ietf.org/html/rfc4192">RFC 4192</a>,
4190 [<a name="ref-RFC4866" id="ref-RFC4866">RFC4866</a>] Arkko, J., Vogt, C., and W. Haddad, "Enhanced Route
4191 Optimization for Mobile IPv6", <a href="http://tools.ietf.org/html/rfc4866">RFC 4866</a>, May 2007.
4193 [<a name="ref-RFC4984" id="ref-RFC4984">RFC4984</a>] Meyer, D., Zhang, L., and K. Fall, "Report from the IAB
4194 Workshop on Routing and Addressing", <a href="http://tools.ietf.org/html/rfc4984">RFC 4984</a>,
4197 [<a name="ref-RFC6480" id="ref-RFC6480">RFC6480</a>] Lepinski, M. and S. Kent, "An Infrastructure to Support
4198 Secure Internet Routing", <a href="http://tools.ietf.org/html/rfc6480">RFC 6480</a>, February 2012.
4200 [<a name="ref-RFC6518" id="ref-RFC6518">RFC6518</a>] Lebovitz, G. and M. Bhatia, "Keying and Authentication for
4201 Routing Protocols (KARP) Design Guidelines", <a href="http://tools.ietf.org/html/rfc6518">RFC 6518</a>,
4204 [<a name="ref-RFC6831" id="ref-RFC6831">RFC6831</a>] Farinacci, D., Meyer, D., Zwiebel, J., and S. Venaas, "The
4205 Locator/ID Separation Protocol (LISP) for Multicast
4206 Environments", <a href="http://tools.ietf.org/html/rfc6831">RFC 6831</a>, January 2013.
4208 [<a name="ref-RFC6832" id="ref-RFC6832">RFC6832</a>] Lewis, D., Meyer, D., Farinacci, D., and V. Fuller,
4209 "Interworking between Locator/ID Separation Protocol
4210 (LISP) and Non-LISP Sites", <a href="http://tools.ietf.org/html/rfc6832">RFC 6832</a>, January 2013.
4212 [<a name="ref-RFC6835" id="ref-RFC6835">RFC6835</a>] Farinacci, D. and D. Meyer, "The Locator/ID Separation
4213 Protocol Internet Groper (LIG)", <a href="http://tools.ietf.org/html/rfc6835">RFC 6835</a>, January 2013.
4215 [<a name="ref-RFC6837" id="ref-RFC6837">RFC6837</a>] Lear, E., "NERD: A Not-so-novel Endpoint ID (EID) to
4216 Routing Locator (RLOC) Database", <a href="http://tools.ietf.org/html/rfc6837">RFC 6837</a>, January 2013.
4218 [<a name="ref-UDP-TUNNELS" id="ref-UDP-TUNNELS">UDP-TUNNELS</a>]
4219 Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and
4220 UDP Checksums for Tunneled Packets", Work in Progress,
4223 [<a name="ref-UDP-ZERO" id="ref-UDP-ZERO">UDP-ZERO</a>] Fairhurst, G. and M. Westerlund, "Applicability Statement
4224 for the use of IPv6 UDP Datagrams with Zero Checksums",
4225 Work in Progress, December 2012.
4229 <span class="grey">Farinacci, et al. Experimental [Page 73]</span>
4230 </pre><!--NewPage--><pre class="newpage"><a name="page-74" id="page-74" href="#page-74" class="invisible"> </a>
4231 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4234 <span class="h2"><h2><a class="selflink" name="appendix-A" href="#appendix-A">Appendix A</a>. Acknowledgments</h2></span>
4236 An initial thank you goes to Dave Oran for planting the seeds for the
4237 initial ideas for LISP. His consultation continues to provide value
4238 to the LISP authors.
4240 A special and appreciative thank you goes to Noel Chiappa for
4241 providing architectural impetus over the past decades on separation
4242 of location and identity, as well as detailed reviews of the LISP
4243 architecture and documents, coupled with enthusiasm for making LISP a
4244 practical and incremental transition for the Internet.
4246 The authors would like to gratefully acknowledge many people who have
4247 contributed discussions and ideas to the making of this proposal.
4248 They include Scott Brim, Andrew Partan, John Zwiebel, Jason Schiller,
4249 Lixia Zhang, Dorian Kim, Peter Schoenmaker, Vijay Gill, Geoff Huston,
4250 David Conrad, Mark Handley, Ron Bonica, Ted Seely, Mark Townsley,
4251 Chris Morrow, Brian Weis, Dave McGrew, Peter Lothberg, Dave Thaler,
4252 Eliot Lear, Shane Amante, Ved Kafle, Olivier Bonaventure, Luigi
4253 Iannone, Robin Whittle, Brian Carpenter, Joel Halpern, Terry
4254 Manderson, Roger Jorgensen, Ran Atkinson, Stig Venaas, Iljitsch van
4255 Beijnum, Roland Bless, Dana Blair, Bill Lynch, Marc Woolward, Damien
4256 Saucez, Damian Lezama, Attilla De Groot, Parantap Lahiri, David
4257 Black, Roque Gagliano, Isidor Kouvelas, Jesper Skriver, Fred Templin,
4258 Margaret Wasserman, Sam Hartman, Michael Hofling, Pedro Marques, Jari
4259 Arkko, Gregg Schudel, Srinivas Subramanian, Amit Jain, Xu Xiaohu,
4260 Dhirendra Trivedi, Yakov Rekhter, John Scudder, John Drake, Dimitri
4261 Papadimitriou, Ross Callon, Selina Heimlich, Job Snijders, Vina
4262 Ermagan, Albert Cabellos, Fabio Maino, Victor Moreno, Chris White,
4263 Clarence Filsfils, and Alia Atlas.
4265 This work originated in the Routing Research Group (RRG) of the IRTF.
4266 An individual submission was converted into the IETF LISP working
4267 group document that became this RFC.
4269 The LISP working group would like to give a special thanks to Jari
4270 Arkko, the Internet Area AD at the time that the set of LISP
4271 documents were being prepared for IESG last call, and for his
4272 meticulous reviews and detailed commentaries on the 7 working group
4273 last call documents progressing toward experimental RFCs.
4285 <span class="grey">Farinacci, et al. Experimental [Page 74]</span>
4286 </pre><!--NewPage--><pre class="newpage"><a name="page-75" id="page-75" href="#page-75" class="invisible"> </a>
4287 <span class="grey"><a href="http://tools.ietf.org/html/rfc6830">RFC 6830</a> LISP January 2013</span>
4298 EMail: farinacci@gmail.com
4312 EMail: dmm@1-4-5.net
4321 EMail: darlewis@cisco.com
4341 Farinacci, et al. Experimental [Page 75]
4344 <span class="noprint"><small><small>Html markup produced by rfcmarkup 1.104, available from
4345 <a href="http://tools.ietf.org/tools/rfcmarkup/">http://tools.ietf.org/tools/rfcmarkup/</a>
4346 </small></small></span>