idnits 2.17.1 draft-ietf-rtgwg-ipfrr-spec-base-04.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 16. -- Found old boilerplate from RFC 3978, Section 5.5 on line 1159. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 1136. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 1143. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 1149. ** This document has an original RFC 3978 Section 5.4 Copyright Line, instead of the newer IETF Trust Copyright according to RFC 4748. ** This document has an original RFC 3978 Section 5.5 Disclaimer, instead of the newer disclaimer which includes the IETF Trust according to RFC 4748. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == No 'Intended status' indicated for this document; assuming Proposed Standard Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack a both a reference to RFC 2119 and the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. RFC 2119 keyword, line 371: '... specification MUST conform to at le...' RFC 2119 keyword, line 396: '...on to use it. Therefore, it SHOULD be...' RFC 2119 keyword, line 433: '... SHOULD be loop-free with respect to...' RFC 2119 keyword, line 508: '...ernate, a router MUST NOT use an alter...' RFC 2119 keyword, line 519: '...ity, then router S MUST NOT use N_i as...' (18 more instances...) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the RFC 3978 Section 5.4 Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'SHOULD not' in this paragraph: Multicast traffic is out of scope for this specification of IP Fast-Reroute. The alternate next-hops SHOULD not be used for multi-cast RPF checks. -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (July 15, 2005) is 6852 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'E1' is mentioned on line 476, but not defined == Missing Reference: 'E2' is mentioned on line 476, but not defined == Missing Reference: 'A' is mentioned on line 479, but not defined == Missing Reference: 'B' is mentioned on line 479, but not defined == Unused Reference: 'RFC3036' is defined on line 978, but no explicit reference was found in the text == Unused Reference: 'RFC3209' is defined on line 1007, but no explicit reference was found in the text ** Obsolete normative reference: RFC 3036 (Obsoleted by RFC 5036) == Outdated reference: A later version (-13) exists of draft-ietf-rtgwg-ipfrr-framework-03 -- Obsolete informational reference (is this intentional?): RFC 2966 (Obsoleted by RFC 5302) -- Obsolete informational reference (is this intentional?): RFC 3137 (Obsoleted by RFC 6987) Summary: 5 errors (**), 0 flaws (~~), 10 warnings (==), 10 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Atlas, Ed. 3 Internet-Draft Google Inc. 4 Expires: January 16, 2006 A. Zinin, Ed. 5 Alcatel 6 July 15, 2005 8 Basic Specification for IP Fast-Reroute: Loop-free Alternates 9 draft-ietf-rtgwg-ipfrr-spec-base-04 11 Status of this Memo 13 By submitting this Internet-Draft, each author represents that any 14 applicable patent or other IPR claims of which he or she is aware 15 have been or will be disclosed, and any of which he or she becomes 16 aware will be disclosed, in accordance with Section 6 of BCP 79. 18 Internet-Drafts are working documents of the Internet Engineering 19 Task Force (IETF), its areas, and its working groups. Note that 20 other groups may also distribute working documents as Internet- 21 Drafts. 23 Internet-Drafts are draft documents valid for a maximum of six months 24 and may be updated, replaced, or obsoleted by other documents at any 25 time. It is inappropriate to use Internet-Drafts as reference 26 material or to cite them other than as "work in progress." 28 The list of current Internet-Drafts can be accessed at 29 http://www.ietf.org/ietf/1id-abstracts.txt. 31 The list of Internet-Draft Shadow Directories can be accessed at 32 http://www.ietf.org/shadow.html. 34 This Internet-Draft will expire on January 16, 2006. 36 Copyright Notice 38 Copyright (C) The Internet Society (2005). 40 Abstract 42 This document describes the use of loop-free alternates to provide 43 local protection for unicast traffic in pure IP and MPLS/LDP networks 44 in the event of a single failure, whether link, node or shared risk 45 link group (SRLG). The goal of this technology is to reduce the 46 micro-looping and packet loss that happens while routers converge 47 after a topology change due to a failure. Rapid failure repair is 48 achieved through use of precalculated backup next-hops that are loop- 49 free and safe to use until the distributed network convergence 50 process completes. 52 Table of Contents 54 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 55 1.1 Failure Scenarios . . . . . . . . . . . . . . . . . . . . 5 56 2. Applicability of Described Mechanisms . . . . . . . . . . . . 6 57 3. Alternate Next-Hop Calculation . . . . . . . . . . . . . . . . 7 58 3.1 Basic Loop-free Condition . . . . . . . . . . . . . . . . 8 59 3.2 Node-Protecting Alternate Next-Hops . . . . . . . . . . . 9 60 3.3 Broadcast and NBMA Links . . . . . . . . . . . . . . . . . 9 61 3.4 ECMP and Alternates . . . . . . . . . . . . . . . . . . . 11 62 3.5 Interactions with ISIS Overload, RFC 3137 and Costed 63 Out Links . . . . . . . . . . . . . . . . . . . . . . . . 11 64 3.6 Selection Procedure . . . . . . . . . . . . . . . . . . . 12 65 4. Using an Alternate . . . . . . . . . . . . . . . . . . . . . . 16 66 4.1 Terminating Use of Alternate . . . . . . . . . . . . . . . 16 67 5. Requirements on LDP Mode . . . . . . . . . . . . . . . . . . . 18 68 6. Routing Aspects . . . . . . . . . . . . . . . . . . . . . . . 18 69 6.1 Multi-Homed Prefixes . . . . . . . . . . . . . . . . . . . 19 70 6.2 OSPF . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 71 6.2.1 OSPF External Routing . . . . . . . . . . . . . . . . 20 72 6.3 BGP Next-Hop Synchronization . . . . . . . . . . . . . . . 20 73 6.4 Multicast Considerations . . . . . . . . . . . . . . . . . 21 74 7. Security Considerations . . . . . . . . . . . . . . . . . . . 21 75 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 21 76 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 21 77 9.1 Normative References . . . . . . . . . . . . . . . . . . . 21 78 9.2 Informative References . . . . . . . . . . . . . . . . . . 22 79 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 23 80 A. OSPF Example Where LFA Based on Local Area Topology is 81 Insufficient . . . . . . . . . . . . . . . . . . . . . . . . . 24 82 Intellectual Property and Copyright Statements . . . . . . . . 26 84 1. Introduction 86 Applications for interactive multimedia services such as VoIP and 87 pseudo-wires can be very sensitive to traffic loss, such as occurs 88 when a link or router in the network fails. A router's convergence 89 time is generally on the order of seconds; the application traffic 90 may be sensitive to losses greater than 10s of milliseconds. 92 As discussed in [FRAMEWORK], minimizing traffic loss requires a 93 mechanism for the router adjacent to a failure to rapidly invoke a 94 repair path, which is minimally affected by any subsequent re- 95 convergence. This specification describes such a mechanism which 96 allows a router whose local link has failed to forward traffic to a 97 pre-computed alternate until the router installs the new primary 98 next-hops based upon the changed network topology. The terminology 99 used in this specification is given in [FRAMEWORK]. The described 100 mechanism assumes that routing in the network is performed using a 101 link-state routing protocol-- OSPF[RFC2328] or ISIS [RFC1195] 102 [RFC2966]. 104 When a local link fails, a router currently must signal the event to 105 its neighbors via the IGP, recompute new primary next-hops for all 106 affected prefixes, and only then install those new primary next-hops 107 into the forwarding plane. Until the new primary next-hops are 108 installed, traffic directed towards the affected prefixes is 109 discarded. This process can take seconds. 111 <-- 112 +-----+ 113 /------| S |--\ 114 / +-----+ \ 115 / 5 8 \ 116 / \ 117 +-----+ +-----+ 118 | E | | N_1 | 119 +-----+ +-----+ 120 \ / 121 \ \ 4 3 / / 122 \| \ / |/ 123 -+ \ +-----+ / +- 124 \---| D |---/ 125 +-----+ 127 Figure 1: Basic Topology 129 The goal of IP Fast-Reroute is to reduce failure reaction time to 10s 130 of milliseconds by using a pre-computed alternate next-hop, in the 131 event that the currently selected primary next-hop fails, so that the 132 alternate can be rapidly used when the failure is detected. A 133 network with this feature experiences less traffic loss and less 134 micro-looping of packets than a network without IPFRR. There are 135 cases where micro-looping is still a possibility since IPFRR coverage 136 varies but in the worst possible situation a network with IPFRR is 137 equivalent with respect to traffic convergence to a network without 138 IPFRR. 140 To clarify the behavior of IP Fast-Reroute, consider the simple 141 topology in Figure 1. When router S computes its shortest path to 142 router D, router S determines to use the link to router E as its 143 primary next-hop. Without IP Fast-Reroute, that link is the only 144 next-hop that router S computes to reach D. With IP Fast-Reroute, S 145 also looks for an alternate next-hop to use. In this example, S 146 would determine that it could send traffic destined to D by using the 147 link to router N_1 and therefore S would install the link to N_1 as 148 its alternate next-hop. At some later time, the link between router 149 S and router E could fail. When that link fails, S and E will be the 150 first to detect it. On detecting the failure, S will stop sending 151 traffic destined for D towards E via the failed link, and instead 152 send the traffic to S's pre-computed alternate next-hop, which is the 153 link to N_1, until a new SPF is run and its results are installed. 154 As with the primary next-hop, an alternate next-hop is computed for 155 each destination. The process of computing an alternate next-hop 156 does not alter the primary next-hop computed via a standard SPF. 158 If in the example of Figure 1, the link cost from N_1 to D increased 159 to 30 from 3, then N_1 would not be a loop-free alternate, because 160 the cost of the path from N_1 to D via S would be 17 while the cost 161 from N_1 directly to D would be 30. In real networks, we may often 162 face this situation. The existence of a suitable loop-free alternate 163 next-hop is topology dependent and the nature of the failure the 164 alternate is calculated for. 166 A neighbor N can provide a loop-free alternate (LFA) if and only if 168 Distance_opt(N, D) < Distance_opt(N, S) + Distance_opt(S, D) 170 Equation 1: Loop-Free Criterion 172 A sub-set of loop-free alternate are downstream paths which must meet 173 a more restrictive condition that is applicable to more complex 174 failure scenarios: 176 Distance_opt(N, D) < Distance_opt(S, D) 178 Equation 2: Downstream Path Criterion 180 1.1 Failure Scenarios 182 The alternate next-hop can protect against a single link failure, a 183 single node failure, one or more shared risk link group failure, or a 184 combination of these. Whenever a failure occurs that is more 185 extensive than what the alternate was intended to protect, there is 186 the possibility of temporarily looping traffic (note again, that such 187 a loop would only last until the next complete SPF calculation). The 188 example where a node fails when the alternate provided only link 189 protection is illustrated below. If unexpected simultaneous failures 190 occur, then micro-looping may occur since the alternates are not pre- 191 computed to avoid the set of failed links. 193 If only link protection is provided and the node fails, it is 194 possible for traffic using the alternates to experience micro- 195 looping. This issue is illustrated in Figure 2. If Link(S->E) 196 fails, then the link-protecting alternate via N will work correctly. 197 However, if router E fails, then both S and N will detect a failure 198 and switch to their alternates. In this example, that would cause S 199 to redirect the traffic to N and N to redirect the traffic to S and 200 thus causing a forwarding loop. Such a scenario can arise because 201 the key assumption, that all other routers in the network are 202 forwarding based upon the shortest path, is violated because of a 203 second simultaneous correlated failure - another link connected to 204 the same primary neighbor. If there are not other protection 205 mechanisms a node failure is still a concern when only using link 206 protection. 208 <@@@ 209 @@@> 210 +-----+ +-----+ 211 | S |-------| N | 212 +-+---+ 5 +-----+ 213 | | 214 | 5 4 | | 215 | | | \|/ 216 \|/ | | 217 | +-----+ | 218 +----| E |---+ 219 +--+--+ 220 | 221 | 222 | 10 223 | 224 +--+--+ 225 | D | 226 +-----+ 228 Figure 2: Link-Protecting Alternates Causing Loop on Node Failure 230 Micro-looping of traffic via the alternates caused when a more 231 extensive failure than planned for can be prevented via selection of 232 only downstream paths as alternates. In Figure 2, S would be able to 233 use N as an alternate, but N could not use S; therefore N would have 234 no alternate and would discard the traffic, thus avoiding the micro- 235 loop. A micro-loop due to the use of alternates can be avoided by 236 using downstream paths because each router in the path to the 237 destination must be closer to the destination (according to the 238 topology prior to the failures). Although use of downstream paths 239 ensures that the micro-looping via alternates does not occur, such a 240 restriction can severely limit the coverage of alternates. 242 It may be desirable to find an alternate that can protect against 243 other correlated failures (of which node failure is a specific 244 instance). In the general case, these are handled by shared risk 245 link groups (SRLGs) where any links in the network can belong to the 246 SRLG. General SRLGs may add unacceptably to the computational 247 complexity of finding a loop-free alternate. 249 However, a sub-category of SRLGs is of interest and can be applied 250 only during the selection of an acceptable alternate. This sub- 251 category is to express correlated failures of links that are 252 connected to the same router. For example, if there are multiple 253 logical sub-interfaces on the same physical interface, such as VLANs 254 on an Ethernet interface, if multiple interfaces use the same 255 physical port because of channelization, or if multiple interfaces 256 share a correlated failure because they are on the same line-card. 257 This sub-category of SRLGs will be referred to as local-SRLGs. A 258 local-SRLG has all of its member links with one end connected to the 259 same router. Thus, router S could select a loop-free alternate which 260 does not use a link in the same local-SRLG as the primary next-hop. 261 The local-SRLGs belonging to E can be protected against via node- 262 protection; i.e. picking a loop-free node-protecting alternate. 264 Where SRLG protection is provided, it is in the context of the 265 particular OSPF or ISIS area, whose topology is used the SPF 266 computations to compute the loop-free alternates. If an SRLG 267 contains links in multiple areas, then separate SRLG-protecting 268 alternates would be required in each area that is traversed by the 269 affected traffic. 271 2. Applicability of Described Mechanisms 273 IP Fast Reroute mechanisms described in this memo cover intra-domain 274 routing only, with OSPF[RFC2328] or ISIS [RFC1195] [RFC2966] as the 275 IGP. Specifically, Fast Reroute for BGP inter-domain routing is not 276 part of this specification. 278 Certain aspects of OSPF inter-area routing behavior explained in 279 Section 6.2 and Appendix A impact the ability of the router 280 calculating the backup next-hops to assess traffic trajectories. In 281 order to avoid micro-looping and ensure required coverage, certain 282 constrains are applied to multi-area OSPF networks: 284 a. Loop-free alternates should not be used in the backbone area if 285 there are any virtual links configured unless, for each transit 286 area, there is a full mesh of virtual links between all ABRs in 287 that area. Loop-free alternates may be used in non-backbone 288 areas regardless of whether there are virtual links configured. 290 b. Loop-free alternates should not be used for inter-area routes in 291 an area that contains more than one alternate ABR. [RFC3509] 293 c. Loop-free alternates should not be used for AS External routes or 294 ASBR routes in a non-backbone area of a network where there 295 exists an ABR that is announced as an ASBR in multiple non- 296 backbone areas and there exists another ABR that is in at least 297 two of the same non-backbone areas. 299 d. Loop-free alternates should not be used in a non-backbone area of 300 a network for AS External routes where an AS External prefix is 301 advertised with the same type of external metric by multiple 302 ASBRs, which are in different non-backbone areas, with a 303 forwarding address of 0.0.0.0 or by one or more ASBRs with 304 forwarding addresses in multiple non-backbone areas when an ABR 305 exists simultaneously in two or more of those non-backbone areas. 307 3. Alternate Next-Hop Calculation 309 In addition to the set of primary next-hops obtained through a 310 shortest path tree (SPT) computation that is part of standard link- 311 state routing functionality, routers supporting IP Fast Reroute also 312 calculate a set of backup next hops that are engaged when a local 313 failure occurs. These backup next hops are calculated to provide 314 required type of protection (i.e. link-protecting and/or node- 315 protecting) and to guarantee that when the expected failure occurs, 316 forwarding traffic through them will not result in a loop. Such next 317 hops are called loop-free alternates or LFAs throughout this 318 specification. 320 In general, to be able to calculate the set of LFAs for a specific 321 destination D, a router needs to know the following basic pieces of 322 information: 324 o Shortest-path distance from the calculating router to the 325 destination (Distance_opt(S, D)) 327 o Shortest-path distance from the router's IGP neighbors to the 328 destination (Distance_opt(N, D)) 330 o Shortest path distance from the router's IGP neighbors to itself 331 (Distance_opt(N, S)) 333 o Distance_opt(S, D) is normally available from the regular SPF 334 calculation performed by the link-state routing protocols. 335 Distance_opt(N, D) and Distance_opt(N, S) can be obtained by 336 performing additional SPF calculations from the perspective of 337 each IGP neighbor (i.e. considering the neighbor's vertex as the 338 root of the SPT--called SPT(N) hereafter--rather than the 339 calculating router's one, called SPT(S)). 341 This specification defines a form of SRLG protection limited to those 342 SRLGs that include a link that the calculating router is directly 343 connected to. Information about local link SRLG membership is 344 manually configured. Information about remote link SRLG membership 345 is dynamically obtained using [ISIS-SRLG] or [OSPF-SRLG]. In order 346 to choose among all available LFAs that provide required SRLG 347 protection for a given destination, the calculating router needs to 348 track the set of SRLGs that the path through a specific IGP neighbor 349 involves. To do so, each node D in the network topology is 350 associated with SRLG_set(N, D), which is the set of SRLGs that would 351 be crossed if traffic to D was forwarded through N. To calculate this 352 set, the router initializes SRLG_set(N, N) for each of its IGP 353 neighbors to be empty. During the SPT(N) calculation, when a new 354 vertex V is added to the SPT, its SRLG_set(N, V) is set to the union 355 of SRLG sets associated with its parents, and the SRLG sets 356 associated with the links from V's parents to V. The union of the set 357 of SRLG associated with a candidate alternate next-hop and the 358 SRLG_set(N, D) for the neighbor reached via that candidate next-hop 359 is used to determine SRLG protection. 361 The following sections provide information required for calculation 362 of LFAs. Sections Section 3.1 through Section 3.4 define different 363 types of LFA conditions. Section 3.5 describes constrains imposed by 364 the IS-IS overload and OSPF stub router functionality. Section 3.6 365 defines the summarized algorithm for LFA calculation using the 366 definitions in the previous sections. 368 3.1 Basic Loop-free Condition 370 Alternate next hops used by implementations following this 371 specification MUST conform to at least the loop-freeness condition 372 stated above in Equation 1. This condition guarantees that 373 forwarding traffic to an LFA will not result in a loop after a link 374 failure. 376 Further conditions may be applied when determining link-protecting 377 and/or node-protecting alternate next-hops as described in Sections 378 Section 3.2 and Section 3.3. 380 3.2 Node-Protecting Alternate Next-Hops 382 For an alternate next-hop N to protect against node failure of a 383 primary neighbor E for destination D, N must be loop-free with 384 respect to both E and D. In other words, N's path to D must not go 385 through E. This is the case if Equation 3 is true, where N is the 386 neighbor providing a loop-free alternate. 388 Distance_opt(N, D) < Distance_opt(N, E) + Distance_opt(E, D) 390 Equation 3: Criteria for a Node-Protecting Loop-Free Alternate 392 If Distance_opt(N,D) = Distance_opt(N, E) + Distance_opt(E, D), it is 393 possible that N has equal-cost paths and one of those could provide 394 protection against E's node failure. However, it is equally possible 395 that one of N's paths goes through E, and the calculating router has 396 no way to influence N's decision to use it. Therefore, it SHOULD be 397 assumed that an alternate next-hop does not offer node protection if 398 Equation 3 is not met. 400 3.3 Broadcast and NBMA Links 402 Verification of the link-protection property of a next hop in the 403 case of a broadcast link is more elaborate than for a point-to-point 404 link. This is because a broadcast link is represented as a pseudo- 405 node with zero-cost links connecting it to other nodes. 407 Because failure of an interface attached to a broadcast segment may 408 mean loss of connectivity of the whole segment, the condition 409 described for broadcast link protection is pessimistic and requires 410 that the alternate is loop-free with regard to the pseudo-node. 411 Consider the example in Figure 3. 413 +-----+ 15 414 | S |-------- 415 +-----+ | 416 | 5 | 417 | | 418 | 0 | 419 /----\ 0 5 +-----+ 420 | PN |-----| N | 421 \----/ +-----+ 422 | 0 | 423 | | 8 424 | 5 | 425 +-----+ 5 +-----+ 426 | E |----| D | 427 +-----+ +-----+ 429 Figure 3: Loop-Free Alternate that is Link-Protecting 431 In Figure 3, N offers a loop-free alternate which is link-protecting. 432 If the primary next-hop uses a broadcast link, then an alternate 433 SHOULD be loop-free with respect to that link's pseudo-node to 434 provide link protection. This requirement is described in Equation 4 435 below. 437 D_opt(N, D) < D_opt(N, pseudo) + D_opt(pseudo, D) 439 Equation 4: Loop-Free Link-Protecting Criterion for Broadcast Links 441 Because the shortest path from the pseudo-node goes through E, if a 442 loop-free alternate from a neighbor N is node-protecting, the 443 alternate will also be link-protecting unless the router S can only 444 reach the alternate neighbor N via the same pseudo-node. This can 445 occur because S will direct traffic away from the shortest path to 446 use an alternate. 448 To obtain link protection, it is necessary both that the path from 449 the selected alternate next-hop does not traverse the link of 450 interest and that the link used from S to reach that alternate next- 451 hop is not the link of interest. The latter can only occur with non- 452 point-to-point links. Therefore, if the primary next-hop is across a 453 broadcast or NBMA interface, it is necessary to consider link 454 protection during the alternate selection. Similar consideration of 455 the link from S to the selected alternate next-hop as well as the 456 path from the selected alternate next-hop is also necessary for SRLG 457 protection. 459 3.4 ECMP and Alternates 461 With equal-cost multi-path, a prefix may have multiple primary next- 462 hops that are used to forward traffic. When a particular primary 463 next-hop fails, alternate next-hops should be used to preserve the 464 traffic. These alternate next-hops may themselves also be primary 465 next-hops, but need not be. Other primary next-hops are not 466 guaranteed to provide protection against the failure scenarios of 467 concern. 469 20 L1 L3 3 470 [N]-----[ S ]--------[E3] 471 | | | 472 | 5 | L2 | 473 20 | | | 474 | --------- | 2 475 | 5 | | 5 | 476 | [E1] [E2]------| 477 | | | 478 | 10 | 10 | 479 |---[A] [B] 480 | | 481 2 |--[D]--| 2 483 Figure 4: ECMP where Primary Next-Hops Provide Limited Protection 485 In Figure 4 S has three primary next-hops to reach D; these are L2 to 486 E1, L2 to E2 and L3 to E3. The primary next-hop L2 to E1 can obtain 487 link and node protection from L3 to E3, which is one of the other 488 primary next-hops; L2 to E1 cannot obtain link protection from the 489 other primary next-hop L2 to E2. Similarly, the primary next-hop L2 490 to E2 can only get node protection from L2 to E1 and can only get 491 link protection from L3 to E3. The third primary next-hop L3 to E3 492 can obtain link and node protection from L2 to E1 and from L2 to E2. 493 It is possible for both the primary next-hop L2 to E2 and the primary 494 next-hop L2 to E1 to obtain an alternate next-hop that provides both 495 link and node protection by using L1. 497 Alternate next-hops are determined for each primary next-hop 498 separately. As with alternate selection in the non-ECMP case, these 499 alternate next-hops should maximize the coverage of the failure 500 cases. 502 3.5 Interactions with ISIS Overload, RFC 3137 and Costed Out Links 504 As described in [RFC3137], there are cases where it is desirable not 505 to have a router used as a transit node. For those cases, it is also 506 desirable not to have the router used on an alternate path. 508 For computing an alternate, a router MUST NOT use an alternate next- 509 hop that is along a link whose cost or reverse cost is LSInfinity 510 (for OSPF) or the maximum cost (for ISIS) or which has the overload 511 bit set (for ISIS). For a broadcast link, the reverse cost 512 associated with a potential alternate next-hop is the cost towards 513 the pseudo-node advertised by the next-hop router. For point-to- 514 point links, if a specific link from the next-hop router cannot be 515 associated with a particular link, then the reverse cost considered 516 is that of the minimum cost link from the next-hop router back to S. 518 In the case of OSPF, if all links from router S to a neighbor N_i 519 have a reverse cost of LSInfinity, then router S MUST NOT use N_i as 520 an alternate. 522 Similarly in the case of ISIS, if N_i has the overload bit set, then 523 S MUST NOT consider using N_i as an alternate. 525 This preserves the desired behavior of diverting traffic away from a 526 router which is following [RFC3137] and it also preserves the desired 527 behavior when an operator sets the cost of a link to LSInfinity for 528 maintenance which is not permitting traffic across that link unless 529 there is no other path. 531 If a link or router which is costed out was the only possible 532 alternate to protect traffic from a particular router S to a 533 particular destination, then there should be no alternate provided 534 for protection. 536 3.6 Selection Procedure 538 A router supporting this specification SHOULD attempt to select at 539 least one loop-free alternate next-hop for each primary next-hop used 540 for a given prefix. A router MAY decide to not use an available 541 loop-free alternate next-hop. A reason for such a decision might be 542 that the loop-free alternate next-hop does not provide protection for 543 the failure scenario of interest. 545 The alternate selection should maximize the coverage of the failure 546 cases. 548 When calculating alternate next-hops, the calculating router S 549 applies the following rules. 551 1. S SHOULD select a loop-free node-protecting alternate next-hop, 552 if one is available. If no loop-free node-protecting alternate 553 is available, then S MAY select a loop-free link-protecting 554 alternate. 556 2. If S has a choice between a loop-free link-protecting node- 557 protecting alternate and a loop-free node-protecting alternate 558 which is not link-protecting, S SHOULD select a loop-free node- 559 protecting alternate which is also link-protecting. This can 560 occur as explained in Section 3.3. 562 3. If S has multiple primary next-hops, then S SHOULD select as a 563 loop-free alternate either one of the other primary next-hops or 564 a loop-free node-protecting alternate if available. If no loop- 565 free node-protecting alternate is available and no other primary 566 next-hop can provide link-protection, then S SHOULD select a 567 loop-free link-protecting alternate. 569 4. Implementations SHOULD support a mode where other primary next- 570 hops satisfying the basic loop-free condition and providing at 571 least a minimal level of protection are preferred over any non- 572 primary alternates. This mode is provided to allow the 573 administrator to preserve traffic patterns based on regular ECMP 574 behavior. 576 Following the above rules maximizes the level of protection and use 577 of primary (ECMP) next-hops. 579 Each next-hop is associated with a set of non-mutually-exclusive 580 characteristics based on whether it is used as a primary next-hop to 581 a particular destination D, and the type of protection it can provide 582 relative to a specific primary next-hop E: 584 Primary Path - The next-hop is used by S as primary. 586 Loop-Free Node-Protecting Alternate - This next-hop satisfies 587 Equation 1 and Equation 3. The path avoids S, S's primary 588 neighbor E, and the link from S to E. 590 Loop-Free Link-Protecting Alternate - This next-hop satisfies 591 Equation 1 but not Equation 3. If the primary next-hop uses a 592 broadcast link, then this next-hop satisfies Equation 4. 594 An alternate path may also provide none, some or complete SRLG 595 protection as well as node and link or link protection. For 596 instance, a link may belong to two SRLGs G1 and G2. The alternate 597 path might avoid other links in G1 but not G2, in which case the 598 alternate would only provide partial SRLG protection. 600 Below is an algorithm that can be used to calculate loop-free 601 alternate next-hops. The algorithm is given for informational 602 purposes and implementations are free to use any other algorithm as 603 long as it satisfies the rules described above. 605 The following procedure describes how to select an alternate next- 606 hop. The procedure is described to determine alternate next-hops to 607 use to reach each router in the topology. Prefixes that are 608 advertised by a single router can use the alternate next-hop computed 609 for the router to which they are attached. The same procedure can be 610 used to reach a prefix that is advertised by more than one router 611 when the logical topological transformation described in Section 6.1 612 is used. 614 S is the computing router and has candidate next-hops H_1 through 615 H_k. N_i and N_j are used to refer to neighbors of S. For a next-hop 616 to be a candidate, the next-hop must be associated with a bi- 617 directional link, as is determined by the IGP. For a particular 618 destination router D, let S have already computed D_opt(S, D), and 619 for each neighbor N_i, D_opt(N_i, D), D_opt(N_i, S), and D_opt(N_i, 620 N_j), the distance from N_i to each other neighbor N_j, and the set 621 of SRLGs traversed by the path D_opt(N_i, D). S SHOULD follow the 622 below procedure for every primary next-hop selected to reach D. This 623 set of primary next-hops is represented P_1 to P_p. This procedure 624 finds the alternate next-hop(s) for P_i. 626 First, initialize the alternate information for P_i as follows: 628 P_i.alt_next_hops = {} 629 P_i.alt_type = NONE P_i.alt_link-protect = FALSE 630 P_i.alt_node-protect = FALSE 631 P_i.alt_srlg-protect = {} 633 For each candidate next-hop H_h, 635 1. Initialize variables as follows: 637 cand_type = NONE 638 cand_link-protect = FALSE 639 cand_node-protect = FALSE 640 cand_srlg-protect = {} 642 2. If H_h is P_i, skip it and continue to the next candidate next- 643 hop. 645 3. If H_h.link is administratively allowed to be used as an 646 alternate, 648 and the cost of H_h.link less than the maximum, 649 and the reverse cost of H_h is less than the maximum, 650 and H_h.neighbor is not overloaded (for ISIS), 651 and H_h.link is bi-directional, 653 then H_h can be considered as an alternate. Otherwise, skip it 654 and continue to the next candidate next-hop. 656 4. If D_opt( H_h.neighbor, D) >= D_opt( H_h.neighbor, S) + D_opt(S, 657 D), then H_h is not loop-free. Skip it and continue to the next 658 candidate next-hop. 660 5. cand_type = LOOP-FREE. 662 6. If H_h is a primary next-hop, set cand_type to PRIMARY. 664 7. If H_h.link is not P_i.link, set can_link-protect to TRUE. 666 8. If D_opt(H_h.neighbor, D) < D_opt(H_h.neighbor, P_i.neighbor) + 667 D_opt(P_i.neighbor, D), set cand_node-protect to TRUE. 669 9. If the router considers SRLGs, then set the cand_srlg-protect to 670 the set of SRLGs traversed on the path from S via P_i to 671 P_i.neighbor. Remove the set of SRLGs to which H_h belongs from 672 cand_srlg-protect. Remove from cand_srlg-protect the set of 673 SRLGs traversed on the path from H_h.neighbor to D. Now 674 cand_srlg-protect holds the set of SRLGs to which P_i belongs 675 and that are not traversed on the path from S via H_h to D. 677 10. If cand_type is PRIMARY, the router prefers other primary next- 678 hops for use as the alternate, and the P_i.alt_type is not 679 PRIMARY, goto Step 18. 681 11. If cand_node-protect is TRUE and P_i.alt_node-protect is FALSE, 682 goto Paragraph 18. 684 12. If cand_link-protect is TRUE and P_i.alt_link-protect is FALSE, 685 goto Step 18. 687 13. If cand_srlg-protect has a better set of SRLGs than 688 P_i.alt_srlg-protect, goto Step 18. 690 14. If cand_srlg-protect is different from P_i.alt_srlg-protect, 691 then select between H_h and P_i.alt_next_hops based upon 692 distance, IP addresses, or any router-local tie-breaker. If H_h 693 is preferred, then goto to Step 18. Otherwise, skip H_h and 694 continue to the next candidate next-hop. 696 15. Based upon the alternate types, the alternate distances, IP 697 addresses, or other tie-breakers, decide if H_h is preferred to 698 P_i.alt_next_hops. If so, goto Step 18. 700 16. Decide if P_i.alt_next_hops is preferred to H_h. If so, then 701 skip H_h and continue to the next candidate next-hop. 703 17. Add H_h into P_i.alt_next_hops. Set P_i.alt_type to the better 704 type of H_h.alt_type and P_i.alt_type. Continue to the next 705 candidate next-hop. 707 18. Replace the P_i alternate next-hop set with H_h as follows: 709 P_i.alt_next_hops = {H_h} 710 P_i.alt_type = cand_type 711 P_i.alt_link-protect = cand_link-protect 712 P_i.alt_node-protect = cand_node-protect 713 P_i.alt_srlg-protect = cand_srlg-protect 715 Continue to the next candidate next-hop. 717 4. Using an Alternate 719 If an alternate next-hop is available, the router redirects traffic 720 to the alternate next-hop in case of a primary next-hop failure as 721 follows. 723 When a next-hop failure is detected via a local interface failure or 724 other failure detection mechanisms (see [FRAMEWORK]), the router 725 SHOULD: 727 1. Remove the primary next-hop associated with the failure. 729 2. Install the loop-free alternate calculated for the failed next- 730 hop if it is not already installed (e.g. the alternate is also a 731 primary next-hop). 733 Note that the router MAY remove other next-hops if it believes (via 734 SRLG analysis) that they may have been affected by the same failure, 735 even if it is not visible at the time of failure detection. 737 The alternate next-hop MUST be used only for traffic types which are 738 routed according to the shortest path. Multicast traffic is 739 specifically out of scope for this specification. 741 4.1 Terminating Use of Alternate 743 NOTE: this section may be replaced with a reference to [ULOOP]. 745 A router MUST limit the amount of time an alternate next-hop is used 746 after the primary next-hop has become unavailable. This ensures that 747 the router will start using the new primary next-hops. It ensures 748 that all possible transient conditions are removed and the network 749 converges according to the deployed routing protocol. 751 It is desirable to avoid micro-forwarding loops involving S. An 752 example illustrating the problem is given in Figure 5. If the link 753 from S to E fails, S will use N1 as an alternate and S will compute 754 N2 as the new primary next-hop to reach D. If S starts using N2 as 755 soon as S can compute and install its new primary, it is probable 756 that N2 will not have yet installed its new primary next-hop. This 757 would cause traffic to loop and be dropped until N2 has installed the 758 new topology. This can be avoided by S delaying its installation and 759 leaving traffic on the alternate next-hop. 761 +-----+ 762 | N2 |-------- | 763 +-----+ 1 | \|/ 764 | | 765 | +-----+ @@> +-----+ 766 | | S |---------| N1 | 767 10 | +-----+ 10 +-----+ 768 | | | 769 | 1 | | | 770 | | \|/ 10 | 771 | +-----+ | | 772 | | E | | \|/ 773 | +-----+ | 774 | | | 775 | 1 | | | 776 | | \|/ | 777 | +-----+ | 778 |----| D |-------------- 779 +-----+ 781 Figure 5: Example where Continued Use of Alternate is Desirable 783 This is an example of a case where the new primary is not a loop-free 784 alternate before the failure and therefore may have been forwarding 785 traffic through S. This will occur when the path via a previously 786 upstream node is shorter than the the path via a loop-free alternate 787 neighbor. In these cases, it is useful to give sufficient time to 788 ensure that the new primary neighbor and other nodes on the new 789 primary path have switched to the new route. 791 If the newly selected primary was loop-free before the failure, then 792 it is safe to switch to that new primary immediately; the new primary 793 wasn't dependent on the failure and therefore its path will not have 794 changed. 796 Given that there is an alternate providing appropriate protection and 797 while the assumption of a single failure holds, it is safe to delay 798 the installation of the new primaries; this will not create 799 forwarding loops because the alternate's path to the destination is 800 known to not go via S or the failed element and will therefore not be 801 affected by the failure. 803 An implementation SHOULD continue to use the alternate next-hops for 804 packet forwarding even after the new routing information is available 805 based on the new network topology. The use of the alternate next- 806 hops for packet forwarding SHOULD terminate: 808 a. if the new primary next-hop was loop-free prior to the topology 809 change, or 811 b. if a configured hold-down, which represents a worst-case bound on 812 the length of the network convergence transition, has expired, or 814 c. if notification of an unrelated topological change in the network 815 is received. 817 5. Requirements on LDP Mode 819 Since LDP traffic will follow the path specified by the IGP, it is 820 also possible for the LDP traffic to follow the loop-free alternates 821 indicated by the IGP. To do so, it is necessary for LDP to have the 822 appropriate labels available for the alternate so that the 823 appropriate out-segments can be installed in the forwarding plane 824 before the failure occurs. 826 This means that a Label Switched Router (LSR) running LDP must 827 distribute its labels for the FECs it can provide to all its 828 neighbors, regardless of whether or not they are upstream. 829 Additionally, LDP must be acting in liberal label retention mode so 830 that the labels which correspond to neighbors that aren't currently 831 the primary neighbor are stored. Similarly, LDP should be in 832 downstream unsolicited mode, so that the labels for the FEC are 833 distributed other than along the SPT. 835 If these requirements are met, then LDP can use the loop-free 836 alternates without requiring any targeted sessions or signaling 837 extensions for this purpose. 839 6. Routing Aspects 840 6.1 Multi-Homed Prefixes 842 An SPF-like computation is run for each topology, which corresponds 843 to a particular OSPF area or ISIS level. The IGP needs to determine 844 loop-free alternates to multi-homed routes. Multi-homed routes occur 845 for routes obtained from outside the routing domain by multiple 846 routers, for subnets on links where the subnet is announced from 847 multiple ends of the link, and for routes advertised by multiple 848 routers to provide resiliency. 850 Figure 6 demonstrates such a topology. In this example, the shortest 851 path to reach the prefix p is via E. The prefix p will have the link 852 to E as its primary next-hop. If the alternate next-hop for the 853 prefix p is simply inherited from the router advertising it on the 854 shortest path to p, then the prefix p's alternate next-hop would be 855 the link to C. This would provide link protection, but not the node 856 protection that is possible via A. 858 5 +---+ 4 +---+ 5 +---+ 859 ------| S |------| A |-----| B | 860 | +---+ +---+ +---+ 861 | | | 862 | 5 | 5 | 863 | | | 864 +---+ 5 +---+ 5 7 +---+ 865 | C |---| E |------ p -------| F | 866 +---+ +---+ +---+ 868 Figure 6: Multi-homed prefix 870 To determine the best protection possible, the prefix p can be 871 treated in the SPF computations as a node with uni-directional links 872 to it from those routers that have advertised the prefix. Such a 873 node need never have its links explored, as it has no out-going 874 links. 876 If there exist multiple multi-homed prefixes that share the same 877 connectivity and the difference in metrics to those routers, then a 878 single node can be used to represent the set. For instance, if in 879 Figure 6 there were another prefix X that was connected to E with a 880 metric of 1 and to F with a metric of 3, then that prefix X could use 881 the same alternate next-hop as was computed for prefix p. 883 A router SHOULD compute the alternate next-hop for an IGP multi-homed 884 prefix by considering alternate paths via all routers that have 885 announced that prefix. 887 6.2 OSPF 889 OSPF introduces certain complications because it is possible for the 890 traffic path to exit an area and then re-enter that area. This can 891 occur whenever a router considers the same route from multiple areas. 892 There are several cases where issues such as this can occur. They 893 happen when another area permits a shorter path to connect two ABRs 894 than is available in the area where the LFA has been computed. To 895 clarify, an example topology is given in Appendix A. 897 a. Virtual Links: These allow paths to leave the backbone area and 898 traverse the transit area. The path provided via the transit 899 area can exit via any ABR. The path taken is not the shortest 900 path determined by doing an SPF in the backbone area. 902 b. Alternate ABR[RFC3509]: When an ABR is not connected to the 903 backbone, it considers the inter-area summaries from multiple 904 areas. The ABR A may determine to use area 2 but that path could 905 traverse another alternate ABR B that determines to use area 1. 906 This can lead to scenarios similar to that illustrated in 907 Figure 7. 909 c. ASBR Summaries: An ASBR may itself be an ABR and can be announced 910 into multiple areas. This presents other ABRs with a decision as 911 to which area to use. This is the example illustrated in 912 Figure 7. 914 d. AS External Prefixes: A prefix may be advertised by multiple 915 ASBRs in different areas and/or with multiple forwarding 916 addresses that are in different areas, which are connected via at 917 least one common ABR. This presents such ABRs with a decision as 918 to which area to use to reach the prefix. 920 6.2.1 OSPF External Routing 922 When a forwarding address is set in an OSPF AS-external LSA, all 923 routers in the network calculate their next-hops for the external 924 prefix by doing a lookup for the forwarding address in the routing 925 table, rather than using the next-hops calculated for the ASBR. In 926 this case, the alternate next-hops SHOULD be computed by selecting 927 among the alternate paths to the forwarding link(s) instead of among 928 alternate paths to the ASBR. 930 6.3 BGP Next-Hop Synchronization 932 Typically BGP prefixes are advertised with AS exit routers router-id, 933 and AS exit routers are reached by means of IGP routes. BGP resolves 934 its advertised next-hop to the immediate next-hop by potential 935 recursive lookups in the routing database. IP Fast-Reroute computes 936 the alternate next-hops to all IGP destinations, which include 937 alternate next-hops to the AS exit router's router-id. BGP simply 938 inherits the alternate next-hop from IGP. The BGP decision process 939 is unaltered; BGP continues to use the IGP optimal distance to find 940 the nearest exit router. MBGP routes do not need to copy the 941 alternate next hops. 943 It is possible to provide ASBR protection if BGP selected selected a 944 set of IGP next-hops and allowed the IGP to determine the primary and 945 alternate next-hops as if the BGP route were a multi-homed prefix. 946 This is for future study. 948 6.4 Multicast Considerations 950 Multicast traffic is out of scope for this specification of IP Fast- 951 Reroute. The alternate next-hops SHOULD not be used for multi-cast 952 RPF checks. 954 7. Security Considerations 956 The mechanism described in this document does not modify any routing 957 protocol messages, and hence no new threats related to packet 958 modifications or replay attacks are introduced. Traffic to certain 959 destinations can be temporarily routed via next-hop routers that 960 would not be used with the same topology change if this mechanism 961 wasn't employed. However, these next-hop routers can be used anyway 962 when a different topological change occurs, and hence this can't be 963 viewed as a new security threat. 965 8. IANA Considerations 967 This document requires no IANA considerations. 969 9. References 971 9.1 Normative References 973 [RFC1195] Callon, R., "Use of OSI IS-IS for routing in TCP/IP and 974 dual environments", RFC 1195, December 1990. 976 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, April 1998. 978 [RFC3036] Andersson, L., Doolan, P., Feldman, N., Fredette, A., and 979 B. Thomas, "LDP Specification", RFC 3036, January 2001. 981 9.2 Informative References 983 [FRAMEWORK] 984 Shand, M., "IP Fast Reroute Framework", 985 draft-ietf-rtgwg-ipfrr-framework-03.txt (work in 986 progress), June 2005. 988 [ISIS-SRLG] 989 Kompella, K. and Y. Rekhter, "IS-IS Extensions in Support 990 of Generalized MPLS", draft-ietf-isis-gmpls-extensions-19 991 (work in progress), October 2003. 993 [OSPF-SRLG] 994 Kompella, K. and Y. Rekhter, "OSPF Extensions in Support 995 of Generalized Multi-Protocol Label Switching", 996 draft-ietf-ccamp-ospf-gmpls-extensions-12 (work in 997 progress), October 2003. 999 [RFC2966] Li, T., Przygienda, T., and H. Smit, "Domain-wide Prefix 1000 Distribution with Two-Level IS-IS", RFC 2966, 1001 October 2000. 1003 [RFC3137] Retana, A., Nguyen, L., White, R., Zinin, A., and D. 1004 McPherson, "OSPF Stub Router Advertisement", RFC 3137, 1005 June 2001. 1007 [RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V., 1008 and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP 1009 Tunnels", RFC 3209, December 2001. 1011 [RFC3509] Zinin, A., Lindem, A., and D. Yeung, "Alternative 1012 Implementations of OSPF Area Border Routers", RFC 3509, 1013 April 2003. 1015 [ULOOP] Zinin, A., "Analysis and Minimization of Microloops in 1016 Link-state Routing Protocols", 1017 draft-zinin-microloop-analysis-01.txt (work in progress), 1018 May 2005. 1020 Authors' Addresses 1022 Alia K. Atlas (editor) 1023 Google Inc. 1024 1600 Amphitheatre Parkway 1025 Mountain View, CA 94043 1026 USA 1028 Email: akatlas@alum.mit.edu 1030 Alex Zinin (editor) 1031 Alcatel 1032 701 E Middlefield Rd. 1033 Mountain View, CA 94043 1034 USA 1036 Email: zinin@psg.com 1038 Raveendra Torvi 1039 Avici Systems, Inc. 1040 101 Billerica Avenue 1041 N. Billerica, MA 01862 1042 USA 1044 Phone: +1 978 964 2026 1045 Email: rtorvi@avici.com 1047 Gagan Choudhury 1048 AT&T 1049 200 Laurel Avenue, Room D5-3C21 1050 Middletown, NJ 07748 1051 USA 1053 Phone: +1 732 420-3721 1054 Email: gchoudhury@att.com 1056 Christian Martin 1057 Verizon 1058 1880 Campus Commons Drive 1059 Reston, VA 20191 1060 USA 1061 Brent Imhoff 1062 LightCore 1063 14567 North Outer Forty Rd. 1064 Chesterfield, MO 63017 1065 USA 1067 Phone: +1 314 880 1851 1068 Email: brent@lightcore.net 1070 Don Fedyk 1071 Nortel Networks 1072 600 Technology Park 1073 Billerica, MA 01821 1074 USA 1076 Phone: +1 978 288 3041 1077 Email: dwfedyk@nortelnetworks.com 1079 Appendix A. OSPF Example Where LFA Based on Local Area Topology is 1080 Insufficient 1082 This appendix provides an example scenario where the local area 1083 topology does not suffice to determine that an LFA is available. As 1084 described in Section 6.2, one problem scenario is for ASBR summaries 1085 where the ASBR is available in two areas via intra-area routes and 1086 there is at least one ABR or alternate ABR that is in both areas. 1087 The following Figure 7 illustrates this case. 1089 5 1090 [ F ]-----------[ C ] 1091 | | 1092 | | 5 1093 20 | 5 | 1 1094 | [ N ]-----[ A ]*****[ F ] 1095 | | # * 1096 | 40 | # 50 * 2 1097 | | 5 # 2 * 1098 | [ S ]-----[ B ]*****[ G ] 1099 | | * 1100 | 5 | * 15 1101 | | * 1102 | [ E ] [ H ] 1103 | | * 1104 | 5 | * 10** 1105 | | * 1106 |---[ X ]-----[ASBR] 1107 5 1109 ---- Link in Area 1 1110 **** Link in Area 2 1111 #### Link in Backbone Area 0 1113 Figure 7: Topology with Multi-area ASBR Causing Area Transiting 1115 In Figure 7, the ASBR is also an ABR and is announced into both area 1116 1 and area 2. A and B are both ABRs that are also connected to the 1117 backbone area. S determines that N can provide a loop-free alternate 1118 to reach the ASBR. N's path goes via A. A also sees an intra-area 1119 route to ASBR via Area 2; the cost of the path in area 2 is 30, which 1120 is less than 35, the cost of the path in area 1. Therefore, A uses 1121 the path from area 2 and directs traffic to F. The path from F in 1122 area 2 goes to B. B is also an ABR and learns the ASBR from both 1123 areas 1 and area 2; B's path via area 1 is shorter (cost 20) than B's 1124 path via area 2 (cost 25). Therefore, B uses the path from area 1 1125 that connects to S. 1127 Intellectual Property Statement 1129 The IETF takes no position regarding the validity or scope of any 1130 Intellectual Property Rights or other rights that might be claimed to 1131 pertain to the implementation or use of the technology described in 1132 this document or the extent to which any license under such rights 1133 might or might not be available; nor does it represent that it has 1134 made any independent effort to identify any such rights. Information 1135 on the procedures with respect to rights in RFC documents can be 1136 found in BCP 78 and BCP 79. 1138 Copies of IPR disclosures made to the IETF Secretariat and any 1139 assurances of licenses to be made available, or the result of an 1140 attempt made to obtain a general license or permission for the use of 1141 such proprietary rights by implementers or users of this 1142 specification can be obtained from the IETF on-line IPR repository at 1143 http://www.ietf.org/ipr. 1145 The IETF invites any interested party to bring to its attention any 1146 copyrights, patents or patent applications, or other proprietary 1147 rights that may cover technology that may be required to implement 1148 this standard. Please address the information to the IETF at 1149 ietf-ipr@ietf.org. 1151 Disclaimer of Validity 1153 This document and the information contained herein are provided on an 1154 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 1155 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET 1156 ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, 1157 INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE 1158 INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 1159 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 1161 Copyright Statement 1163 Copyright (C) The Internet Society (2005). This document is subject 1164 to the rights, licenses and restrictions contained in BCP 78, and 1165 except as set forth therein, the authors retain all their rights. 1167 Acknowledgment 1169 Funding for the RFC Editor function is currently provided by the 1170 Internet Society.