idnits 2.17.1 draft-ietf-rtgwg-lfa-manageability-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 12, 2014) is 3724 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'I-D.litkowski-rtgwg-lfa-rsvpte-cooperation' is defined on line 928, but no explicit reference was found in the text == Unused Reference: 'I-D.psarkar-isis-node-admin-tag' is defined on line 934, but no explicit reference was found in the text == Unused Reference: 'RFC3630' is defined on line 948, but no explicit reference was found in the text == Unused Reference: 'RFC3906' is defined on line 952, but no explicit reference was found in the text == Unused Reference: 'RFC4090' is defined on line 956, but no explicit reference was found in the text == Unused Reference: 'RFC5305' is defined on line 960, but no explicit reference was found in the text == Unused Reference: 'RFC5714' is defined on line 963, but no explicit reference was found in the text == Unused Reference: 'RFC5715' is defined on line 966, but no explicit reference was found in the text ** Obsolete normative reference: RFC 4205 (Obsoleted by RFC 5307) == Outdated reference: A later version (-11) exists of draft-ietf-rtgwg-remote-lfa-04 == Outdated reference: A later version (-03) exists of draft-psarkar-isis-node-admin-tag-00 == Outdated reference: A later version (-05) exists of draft-psarkar-rtgwg-rlfa-node-protection-03 Summary: 1 error (**), 0 flaws (~~), 12 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Routing Area Working Group S. Litkowski 3 Internet-Draft B. Decraene 4 Intended status: Standards Track Orange 5 Expires: August 16, 2014 C. Filsfils 6 K. Raza 7 Cisco Systems 8 M. Horneffer 9 Deutsche Telekom 10 P. Sarkar 11 Juniper Networks 12 February 12, 2014 14 Operational management of Loop Free Alternates 15 draft-ietf-rtgwg-lfa-manageability-03 17 Abstract 19 Loop Free Alternates (LFA), as defined in RFC 5286 is an IP Fast 20 ReRoute (IP FRR) mechanism enabling traffic protection for IP traffic 21 (and MPLS LDP traffic by extension). Following first deployment 22 experiences, this document provides operational feedback on LFA, 23 highlights some limitations, and proposes a set of refinements to 24 address those limitations. It also proposes required management 25 specifications. 27 This proposal is also applicable to remote LFA solution. 29 Requirements Language 31 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 32 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 33 document are to be interpreted as described in [RFC2119]. 35 Status of this Memo 37 This Internet-Draft is submitted in full conformance with the 38 provisions of BCP 78 and BCP 79. 40 Internet-Drafts are working documents of the Internet Engineering 41 Task Force (IETF). Note that other groups may also distribute 42 working documents as Internet-Drafts. The list of current Internet- 43 Drafts is at http://datatracker.ietf.org/drafts/current/. 45 Internet-Drafts are draft documents valid for a maximum of six months 46 and may be updated, replaced, or obsoleted by other documents at any 47 time. It is inappropriate to use Internet-Drafts as reference 48 material or to cite them other than as "work in progress." 49 This Internet-Draft will expire on August 16, 2014. 51 Copyright Notice 53 Copyright (c) 2014 IETF Trust and the persons identified as the 54 document authors. All rights reserved. 56 This document is subject to BCP 78 and the IETF Trust's Legal 57 Provisions Relating to IETF Documents 58 (http://trustee.ietf.org/license-info) in effect on the date of 59 publication of this document. Please review these documents 60 carefully, as they describe your rights and restrictions with respect 61 to this document. Code Components extracted from this document must 62 include Simplified BSD License text as described in Section 4.e of 63 the Trust Legal Provisions and are provided without warranty as 64 described in the Simplified BSD License. 66 Table of Contents 68 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 69 2. Operational issues with default LFA tie breakers . . . . . . . 4 70 2.1. Case 1: Edge router protecting core failures . . . . . . . 5 71 2.2. Case 2: Edge router choosen to protect core failures 72 while core LFA exists . . . . . . . . . . . . . . . . . . 6 73 2.3. Case 3: suboptimal core alternate choice . . . . . . . . . 7 74 2.4. Case 4: ISIS overload bit on LFA computing node . . . . . 8 75 3. Need for coverage monitoring . . . . . . . . . . . . . . . . . 8 76 4. Need for LFA activation granularity . . . . . . . . . . . . . 9 77 5. Configuration requirements . . . . . . . . . . . . . . . . . . 9 78 5.1. LFA enabling/disabling scope . . . . . . . . . . . . . . . 9 79 5.2. Policy based LFA selection . . . . . . . . . . . . . . . . 10 80 5.2.1. Connected vs remote alternates . . . . . . . . . . . . 11 81 5.2.2. Mandatory criteria . . . . . . . . . . . . . . . . . . 11 82 5.2.3. Enhanced criteria . . . . . . . . . . . . . . . . . . 12 83 5.2.4. Retrieving alternate path attributes . . . . . . . . . 12 84 5.2.5. ECMP LFAs . . . . . . . . . . . . . . . . . . . . . . 14 85 5.2.6. SRLG . . . . . . . . . . . . . . . . . . . . . . . . . 15 86 5.2.7. Link coloring . . . . . . . . . . . . . . . . . . . . 16 87 5.2.8. Bandwidth . . . . . . . . . . . . . . . . . . . . . . 17 88 5.2.9. Alternate preference . . . . . . . . . . . . . . . . . 18 89 6. Operational aspects . . . . . . . . . . . . . . . . . . . . . 19 90 6.1. ISIS overload bit on LFA computing node . . . . . . . . . 19 91 6.2. Manual triggering of FRR . . . . . . . . . . . . . . . . . 19 92 6.3. Required local information . . . . . . . . . . . . . . . . 20 93 6.4. Coverage monitoring . . . . . . . . . . . . . . . . . . . 20 94 6.5. LFA and network planning . . . . . . . . . . . . . . . . . 21 95 7. Security Considerations . . . . . . . . . . . . . . . . . . . 21 96 8. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 21 97 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 22 98 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 22 99 11. References . . . . . . . . . . . . . . . . . . . . . . . . . . 22 100 11.1. Normative References . . . . . . . . . . . . . . . . . . . 22 101 11.2. Informative References . . . . . . . . . . . . . . . . . . 22 102 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 23 104 1. Introduction 106 Following the first deployments of Loop Free Alternates (LFA), this 107 document provides feedback to the community about the management of 108 LFA. 110 Section 2 provides real uses cases illustrating some limitations 111 and suboptimal behavior. 113 Section 4 proposes requirements for activation granularity and 114 policy based selection of the alternate. 116 Section 5 express requirements for the operational management of 117 LFA. 119 2. Operational issues with default LFA tie breakers 121 [RFC5286] introduces the notion of tie breakers when selecting the 122 LFA among multiple candidate alternate next-hops. When multiple LFA 123 exist, RFC 5286 has favored the selection of the LFA providing the 124 best coverage of the failure cases. While this is indeed a goal, 125 this is one among multiple and in some deployment this lead to the 126 selection of a suboptimal LFA. The following sections details real 127 use cases of such limitations. 129 Note that the use case of per-prefix LFA is assumed throughout this 130 analysis. 132 2.1. Case 1: Edge router protecting core failures 134 R1 --------- R2 ---------- R3 --------- R4 135 | 1 100 1 | 136 | | 137 | 100 | 100 138 | | 139 | 1 100 1 | 140 R5 --------- R6 ---------- R7 --------- R8 -- R9 - PE1 141 | | | | 142 | 5k | 5k | 5k | 5k 143 | | | | 144 +--- n*PEx ---+ +---- PE2 ----+ 145 | 146 | 147 PEy 149 Figure 1 151 Rx routers are core routers using n*10G links. PEs are connected 152 using links with lower bandwidth. PEx are a set of PEs connected to 153 R5 and R6. 155 In figure 1, let us consider the traffic flowing from PE1 to PEx. 156 The nominal path is R9-R8-R7-R6-PEx. Let us consider the failure of 157 link R7-R8. For R8, R4 is not an LFA and the only available LFA is 158 PE2. 160 When the core link R8-R7 fails, R8 switches all traffic destined to 161 all the PEx towards the edge node PE2. Hence an edge node and edge 162 links are used to protect the failure of a core link. Typically, 163 edge links have less capacity than core links and congestion may 164 occur on PE2 links. Note that although PE2 was not directly affected 165 by the failure, its links become congested and its traffic will 166 suffer from the congestion. 168 In summary, in case of failure, the impact on customer traffic is: 170 o From PE2 point of view : 172 * without LFA: no impact 174 * with LFA: traffic is partially dropped (but possibly 175 prioritized by a QoS mechanism). It must be highlighted that 176 in such situation, traffic not affected by the failure may be 177 affected by the congestion. 179 o From R8 point of view: 181 * without LFA: traffic is totally dropped until convergence 182 occurs. 184 * with LFA: traffic is partially dropped (but possibly 185 prioritized by a QoS mechanism). 187 Besides the congestion aspects of using an Edge router as an 188 alternate to protect a core failure, a service provider may consider 189 this as a bad routing design and would like to prevent it. 191 2.2. Case 2: Edge router choosen to protect core failures while core 192 LFA exists 194 R1 --------- R2 ------------ R3 --------- R4 195 | 1 100 | 1 | 196 | | | 197 | 100 | 30 | 30 198 | | | 199 | 1 50 50 | 10 | 200 R5 -------- R6 ---- R10 ---- R7 -------- R8 --- R9 - PE1 201 | | \ | 202 | 5000 | 5000 \ 5000 | 5000 203 | | \ | 204 +--- n*PEx --+ +----- PE2 ----+ 205 | 206 | 207 PEy 209 Figure 2 211 Rx routers are core routers meshed with n*10G links. PEs are meshed 212 using links with lower bandwidth. 214 In the figure 2, let us consider the traffic coming from PE1 to PEx. 215 Nominal path is R9-R8-R7-R10-R6-PEx. Let us consider the failure of 216 the link R7-R8. For R8, R4 is a link-protecting LFA and PE2 is a 217 node-protecting LFA. PE2 is chosen as best LFA due to its better 218 protection type. Just like in case 1, this may lead to congestion on 219 PE2 links upon LFA activation. 221 2.3. Case 3: suboptimal core alternate choice 223 +--- PE3 --+ 224 / \ 225 1000 / \ 1000 226 / \ 227 +----- R1 ---------------- R2 ----+ 228 | | 500 | | 229 | 10 | | | 10 230 | | | | 231 R5 | 10 | 10 R7 232 | | | | 233 | 10 | | | 10 234 | | 500 | | 235 +---- R3 ---------------- R4 -----+ 236 \ / 237 1000 \ / 1000 238 \ / 239 +--- PE1 ---+ 241 Figure 3 243 Rx routers are core routers. R1-R2 and R3-R4 links are 1G links. 244 All others inter Rx links are 10G links. 246 In the figure above, let us consider the failure of link R1-R3. For 247 destination PE3, R3 has two possible alternates: 249 o R4, which is node-protecting 251 o R5, which is link-protecting 253 R4 is chosen as best LFA due to its better protection type. However, 254 it may not be desirable to use R4 for bandwidth capacity reason. A 255 service provider may prefer to use high bandwidth links as prefered 256 LFA. In this example, prefering shortest path over protection type 257 may achieve the expected behavior, but in cases where metric are not 258 reflecting bandwidth, it would not work and some other criteria would 259 need to be involved when selecting the best LFA. 261 2.4. Case 4: ISIS overload bit on LFA computing node 263 P1 P2 264 | \ / | 265 50 | 50 \/ 50 | 50 266 | /\ | 267 PE1-+ +-- PE2 268 \ / 269 45 \ / 45 270 -PE3-+ 271 (OL set) 273 Figure 4 275 In the figure above, PE3 has its overload bit set (permanently, for 276 design reason) and wants to protect traffic using LFA for destination 277 PE2. 279 On PE3, the loopfree condition is not satisified : 100 !< 45 + 45. 280 PE1 is thus not considered as an LFA. However thanks to the overload 281 bit set on PE3, we know that PE1 is loopfree so PE1 is an LFA to 282 reach PE2. 284 In case of overload condition set on a node, LFA behavior must be 285 clarified. 287 3. Need for coverage monitoring 289 As per [RFC6571], LFA coverage highly depends on the used network 290 topology. Even if remote LFA ([I-D.ietf-rtgwg-remote-lfa]) extends 291 significantly the coverage of the basic LFA specification, there is 292 still some cases where protection would not be available. As network 293 topologies are constantly evolving (network extension, capacity 294 addings, latency optimization ...), the protection coverage may 295 change. Fast reroute functionality may be critical for some services 296 supported by the network, a service provider must constantly know 297 what protection coverage is currently available on the network. 298 Moreover, predicting the protection coverage in case of network 299 topology change is mandatory. 301 Today network simulation tool associated with whatif scenarios 302 functionnality are often used by service providers for the overall 303 network design (capacity, path optimization ...). Section 6.5, 304 Section 6.4 and Section 6.3 of this document propose to add LFA 305 informations into such tool and within routers, so a service provider 306 may be able : 308 o to evaluate protection coverage after a topology change. 310 o to adjust the topology change to cover the primary need (e.g. 311 latency optimization or bandwidth increase) as well as LFA 312 protection. 314 o monitor constantly the LFA coverage in the live network and being 315 alerted. 317 4. Need for LFA activation granularity 319 As all FRR mechanism, LFA installs backup paths in Forwarding 320 Information Base (FIB). Depending of the hardware used by a service 321 provider, FIB ressource may be critical. Activating LFA, by default, 322 on all available components (IGP topologies, interface, address 323 families ...) may lead to waste of FIB ressource as generally in a 324 network only few destinations should be protected (e.g. loopback 325 addresses supporting MPLS services) compared to the amount of 326 destinations in RIB. 328 Moreover a service provider may implement multiple different FRR 329 mechanism in its networks for different usages (MRT, TE FRR), 330 computing LFAs for prefixes or interfaces that are already protected 331 by another mechanism is useless. 333 Section 5 of this document propose some implementation guidelines. 335 5. Configuration requirements 337 Controlling best alternate and LFA activation granularity is a 338 requirement for Service Providers. This section defines 339 configuration requirements for LFA. 341 5.1. LFA enabling/disabling scope 343 The granularity of LFA activation should be controlled (as alternate 344 nexthop consume memory in forwarding plane). 346 An implementation of LFA SHOULD allow its activation with the 347 following criteria: 349 o Per address-family : ipv4 unicast, ipv6 unicast, LDP IPv4 unicast, 350 LDP IPv6 unicast ... 352 o Per routing context : VRF, virtual/logical router, global routing 353 table, ... 355 o Per interface 357 o Per protocol instance, topology, area 359 o Per prefixes: prefix protection SHOULD have a better priority 360 compared to interface protection. This means that if a specific 361 prefix must be protected due to a configuration request, LFA must 362 be computed and installed for this prefix even if the primary 363 outgoing interface is not configured for protection. 365 5.2. Policy based LFA selection 367 When multiple alternates exist, LFA selection algorithm is based on 368 tie breakers. Current tie breakers do not provide sufficient control 369 on how the best alternate is chosen. This document proposes an 370 enhanced tie breaker allowing service providers to manage all 371 specific cases: 373 1. An implementation of LFA SHOULD support policy-based decision for 374 determining the best LFA. 376 2. Policy based decision SHOULD be based on multiple criterions, 377 with each criteria having a level of preference. 379 3. If the defined policy does not permit to determine a unique best 380 LFA, an implementation SHOULD pick only one based on its own 381 decision, as a default behavior. An implementation SHOULD also 382 support election of multiple LFAs, for loadbalancing purposes. 384 4. Policy SHOULD be applicable to a protected interface or to a 385 specific set of destinations. In case of application on the 386 protected interface, all destinations primarily routed on this 387 interface SHOULD use the interface policy. 389 5. It is an implementation choice to reevaluate policy dynamically 390 or not (in case of policy change). If a dynamic approach is 391 chosen, the implementation SHOULD recompute the best LFAs and 392 reinstall them in FIB, without service disruption. If a non- 393 dynamic approach is chosen, the policy would be taken into 394 account upon the next IGP event. In this case, the 395 implementation SHOULD support a command to manually force the 396 recomputation/reinstallation of LFAs. 398 5.2.1. Connected vs remote alternates 400 In addition to direct LFAs, tunnels (e.g. IP, LDP or RSVP-TE) to 401 distant routers may be used to complement LFA coverage (tunnel tail 402 used as virtual neighbor). When a router has multiple alternate 403 candidates for a specific destination, it may have connected 404 alternates and remote alternates reachable via a tunnel. Connected 405 alternates may not always provide an optimal routing path and it may 406 be preferable to select a remote alternate over a connected 407 alternate. The usage of tunnels to extend LFA coverage is described 408 in [I-D.ietf-rtgwg-remote-lfa]. 410 In figure 1, there is no core alternate for R8 to reach PEs located 411 behind R6, so R8 is using PE2 as alternate, which may generate 412 congestion when FRR is activated. Instead, we could have a remote 413 core alternate for R8 to protect PEs destinations. For example, a 414 tunnel from R8 to R3 would ensure LFA protection without using an 415 edge router to protect a core router. 417 When selecting the best alternate, the selection algorithm MUST 418 consider all available alternates (connected or tunnel). Especially, 419 computation of PQ set ([I-D.ietf-rtgwg-remote-lfa]) SHOULD be 420 performed before best alternate selection. 422 5.2.2. Mandatory criteria 424 An implementation of LFA MUST support the following criteria: 426 o Non candidate link: A link marked as "non candidate" will never be 427 used as LFA. 429 o A primary nexthop being protected by another primary nexthop of 430 the same prefix (ECMP case). 432 o Type of protection provided by the alternate: link protection, 433 node protection. In case of node protection preference, an 434 implementation SHOULD support fallback to link protection if node 435 protection is not available. 437 o Shortest path: lowest IGP metric used to reach the destination. 439 o SRLG (as defined in [RFC5286] Section 3, see also Section 5.2.6 440 for more details). 442 5.2.3. Enhanced criteria 444 An implementation of LFA SHOULD support the following enhanced 445 criteria: 447 o Downstreamness of an alternate : preference of a downstream path 448 over a non downstream path SHOULD be configurable. 450 o Link coloring with : include, exclude and preference based system 451 (see Section 5.2.7). 453 o Link Bandwidth (see Section 5.2.8). 455 o Alternate preference (see Section 5.2.9). 457 5.2.4. Retrieving alternate path attributes 459 The policy to select the best alternate evaluate multiple criterions 460 (e.g. metric, SRLG, link colors ...) which first need to be computed 461 for each alternate.. In order to compare the different alternate 462 path, a router must retrieve the attributes of each alternate path. 463 The alternate path is composed of two distinct parts : PLR to 464 alternate and alternate to destination. 466 5.2.4.1. Connected alternate 468 For alternate path using a connected alternate : 470 o attributes from PLR to alternate path are retrieved from the 471 interface connected to the alternate. 473 o attributes from alternate to destination path are retrieved from 474 SPF rooted at the alternate. As the alternate is a connected 475 alternate, the SPF has already been computed to find the 476 alternate, so there is no need of additional computation. 478 5.2.4.2. Remote alternate 480 For alternate path using a remote alternate (tunnel) : 482 o attributes from the PLR to alternate path are retrieved using the 483 PLR's primary SPF if P space is used or using the neighbor's SPF 484 if extended P space is used, combined with the attributes of the 485 link(s) to reach that neighbor. In both cases, no additional SPF 486 is required. 488 o attributes from alternate to destination path are retrieved from 489 SPF rooted at the remote alternate. An additional forward SPF is 490 required for each remote alternate as indicated in 491 [I-D.psarkar-rtgwg-rlfa-node-protection] section 3.2.. 493 The number of remote alternates may be very high, simulations shown 494 that hundred's of PQs may exist for a single interface being 495 protected. Running a forward SPF for every PQ-node in the network is 496 not scalable. 498 To handle this situation, it is needed to limit the number of remote 499 alternates to be evaluated to a finite number before collecting 500 alternate path attributes and running the policy evaluation. 501 [I-D.psarkar-rtgwg-rlfa-node-protection] Section 2.3.3 provides a way 502 to reduce the number of PQ to be evaluated. 504 Link Remote Remote 505 alternate alternate alternate 506 ------------- ------------------ ------------- 507 Alternates | LFA | | rLFA (PQs) | | Static | 508 sources | | | | | tunnels | 509 ------------- ------------------ ------------- 510 | | | 511 | | | 512 | ---------------------- | 513 | | Prune some PQs | | 514 | | (sorting strategy) | | 515 | ---------------------- | 516 | | | 517 | | | 518 ------------------------------------------------ 519 | Collect alternate attributes | 520 ------------------------------------------------ 521 | 522 | 523 ------------------------- 524 | Evaluate policy | 525 ------------------------- 526 | 527 | 528 Best alternates 530 5.2.5. ECMP LFAs 532 10 533 PE2 - PE3 534 | | 535 50 | 5 | 50 536 P1----P2 537 \\ // 538 50 \\ // 50 539 PE1 541 Figure 5 543 Links between P1 and PE1 are L1 and L2, links between P2 and PE1 are 544 L3 and L4 546 In the figure above, primary path from PE1 to PE2 is through P1 using 547 ECMP on two parallel links L1 and L2. In case of standard ECMP 548 behavior, if L1 is failing, postconvergence nexthop would become L2 549 and there would be no longer ECMP. If LFA is activated, as stated in 550 [RFC5286] Section 3.4., "alternate next-hops may themselves also be 551 primary next-hops, but need not be" and "alternate next-hops should 552 maximize the coverage of the failure cases". In this scenario there 553 is no alternate providing node protection, LFA will so prefer L2 as 554 alternate to protect L1 which makes sense compared to postconvergence 555 behavior. 557 Considering a different scenario using figure 5, where L1 and L2 are 558 configured as a layer 3 bundle using a local feature, as well as 559 L3/L4 being a second layer 3 bundle. Layer 3 bundles are configured 560 as if a link in the bundle is failing, the traffic must be rerouted 561 out of the bundle. Layer 3 bundles are generally introduced to 562 increase bandwidth between nodes. In nominal situation, ECMP is 563 still available from PE1 to PE2, but if L1 is failing, 564 postconvergence nexthop would become ECMP on L3 and L4. In this 565 case, LFA behavior SHOULD be adapted in order to reflect the 566 bandwidth requirement. 568 We would expect the following FIB entry on PE1 : 570 On PE1 : PE2 +--> ECMP -> L1 571 | | 572 | +----> L2 573 | 574 +--> LFA(ECMP) -> L3 575 | 576 +---------> L4 578 If L1 or L2 is failing, traffic must be switched on the LFA ECMP 579 bundle rather than using the other primary nexthop. 581 As mentioned in [RFC5286] Section 3.4., protecting a link within an 582 ECMP by another primary nexthop is not a MUST. Moreover, we already 583 presented in this document, that maximizing the coverage of the 584 failure case may not be the right approach and policy based choice of 585 alternate may be preferred. 587 An implementation SHOULD permit to prefer a primary nexthop by 588 another primary nexthop with the possibility to deactivate this 589 criteria. An implementation SHOULD permit to use an ECMP bundle as a 590 LFA. 592 5.2.6. SRLG 594 [RFC5286] Section 3. proposes to reuse GMPLS IGP extensions to encode 595 SRLGs ([RFC4205] and [RFC4203]). The section is also describing the 596 algorithm to compute SRLG protection. 598 When SRLG protection is computed, and implementation SHOULD permit to 599 : 601 o Exclude alternates violating SRLG. 603 o Maintain a preference system between alternates based on number of 604 SRLG violations : more violations = less preference. 606 When applying SRLG criteria, the SRLG violation check SHOULD be 607 performed on source to alternate as well as alternate to destination 608 paths. In the case of remote LFA, PQ to destination path attributes 609 would be retrieved from SPT rooted at PQ. 611 5.2.7. Link coloring 613 Link coloring is a powerful system to control the choice of 614 alternates. Protecting interfaces are tagged with colors. Protected 615 interfaces are configured to include some colors with a preference 616 level, and exclude others. 618 Link color information SHOULD be signalled in the IGP. How 619 signalling is done is out of scope of the document but it may be 620 useful to reuse existing admin-groups from traffic-engineering 621 extensions. 623 PE2 624 | +---- P4 625 | / 626 PE1 ---- P1 --------- P2 627 | 10Gb 628 1Gb | 629 | 630 P3 632 Figure 5 634 Example : P1 router is connected to three P routers and two PEs. 636 P1 is configured to protect the P1-P4 link. We assume that given the 637 topology, all neighbors are candidate LFA. We would like to enforce 638 a policy in the network where only a core router may protect against 639 the failure of a core link, and where high capacity links are 640 prefered. 642 In this example, we can use the proposed link coloring by: 644 o Marking PEs links with color RED 646 o Marking 10Gb CORE link with color BLUE 648 o Marking 1Gb CORE link with color YELLOW 650 o Configured the protected interface P1->P4 with : 652 * Include BLUE, preference 200 654 * Include YELLOW, preference 100 656 * Exclude RED 658 Using this, PE links will never be used to protect against P1-P4 link 659 failure and 10Gb link will be be preferred. 661 The main advantage of this solution is that it can easily be 662 duplicated on other interfaces and other nodes without change. A 663 Service Provider has only to define the color system (associate color 664 with a significance), as it is done already for TE affinities or BGP 665 communities. 667 An implementation of link coloring: 669 o SHOULD support multiple include and exclude colors on a single 670 protected interface. 672 o SHOULD provide a level of preference between included colors. 674 o SHOULD support multiple colors configuration on a single 675 protecting interface. 677 5.2.8. Bandwidth 679 As mentionned in previous sections, not taking into account bandwidth 680 of an alternate could lead to congestion during FRR activation. We 681 propose to base the bandwidth criteria on the link speed information 682 for the following reason : 684 o if a router S has a set of X destinations primarly forwarded to N, 685 using per prefix LFA may lead to have a subset of X protected by a 686 neighbor N1, another subset by N2, another subset by Nx ... 688 o S is not aware about traffic flows to each destination and is not 689 able to evaluate how much traffic will be sent to N1,N2, ... Nx 690 in case of FRR activation. 692 Based on this, it is not useful to gather available bandwidth on 693 alternate paths, as the router does not know how much bandwidth it 694 requires for protection. The proposed link speed approach provides a 695 good approximation with a small cost as information is easily 696 available. 698 The bandwidth criteria of the policy framework SHOULD work in two 699 ways : 701 o PRUNE : exclude a LFA if link speed to reach it is lower than the 702 link speed of the primary nexthop interface. 704 o PREFER : prefer a LFA based on his bandwidth to reach it compared 705 to the link speed of the primary nexthop interface. 707 5.2.9. Alternate preference 709 Rather than tagging interface on each node (using link color) to 710 identify alternate node type (as example), it would be helpful if 711 routers could be identified in the IGP. This would permit a grouped 712 processing on multiple nodes. As an implementation need to exclude 713 some specific alternates (see Section 5.2.3), an implementation : 715 o SHOULD be able to give a preference to specific alternate. 717 o SHOULD be able to give a preference to a group of alternate. 719 o SHOULD be able to exclude a group of alternate. 721 A specific alternate may be identified by its interface, IP address 722 or router ID and group of alternates may be identified by a marker 723 (tag). 725 Consider the following network: 727 PE3 728 | 729 | 730 PE2 731 | +---- P4 732 | / 733 PE1 ---- P1 -------- P2 734 | 10Gb 735 1Gb | 736 | 737 P3 739 Figure 6 741 In the example above, each node is configured with a specific tag 742 flooded through the IGP. 744 o PE1,PE3: 200 (non candidate). 746 o PE2: 100 (edge/core). 748 o P1,P2,P3: 50 (core). 750 A simple policy could be configured on P1 to choose the best 751 alternate for P1->P4 based on router function/role as follows : 753 o criteria 1 -> alternate preference: exclude tag 100 and 200. 755 o criteria 2 -> bandwidth. 757 6. Operational aspects 759 6.1. ISIS overload bit on LFA computing node 761 In [RFC5286], Section 3.5, the setting of the overload bit condition 762 in LFA computation is only taken into account for the case where a 763 neighbor has the overload bit set. 765 In addition to RFC 5286 inequality 1 Loop-Free Criterion 766 (Distance_opt(N, D) < Distance_opt(N, S) + Distance_opt(S, D)), the 767 IS-IS overload bit of the LFA calculating neighbor (S) SHOULD be 768 taken into account. Indeed, if it has the overload bit set, no 769 neighbor will loop back to traffic to itself. 771 6.2. Manual triggering of FRR 773 Service providers often perform manual link shutdown (using router 774 CLI) to perform some network changes/tests. A manual link shutdown 775 may be done at multiple level : physical interface, logical 776 interface, IGP interface, BFD session ... Especially testing or 777 troubleshooting FRR requires to perform the manual shutdown on the 778 remote end of the link as generally a local shutdown would not 779 trigger FRR. 781 To enhance such situation, an implementation SHOULD support 782 triggering/activating LFA Fast Reroute for a given link when a manual 783 shutdown is done on a component that currently supports FRR 784 activation. 786 For example : 788 o if an implementation supports FRR activation upon BFD session down 789 event, this implementation SHOULD support FRR activation when a 790 manual shutdown is done on the BFD session. But if an 791 implementation does not support FRR activation on BFD session 792 down, there is no need for this implementation to support FRR 793 activation on manual shutdown of BFD session. 795 o if an implementation supports FRR activation on physical link down 796 event (e.g. Rx laser Off detection, or error threshold raised 797 ...), this implementation SHOULD support FRR activation when a 798 manual shutdown at physical interface is done. But if an 799 implementation does not support FRR activation on physical link 800 down event, there is no need for this implementation to support 801 FRR activation on manual physical link shutdown. 803 6.3. Required local information 805 LFA introduction requires some enhancement in standard routing 806 information provided by implementations. Moreover, due to the non 807 100% coverage, coverage informations is also required. 809 Hence an implementation : 811 o MUST be able to display, for every prefixes, the primary nexthop 812 as well as the alternate nexthop information. 814 o MUST provide coverage information per activation domain of LFA 815 (area, level, topology, instance, virtual router, address family 816 ...). 818 o MUST provide number of protected prefixes as well as non protected 819 prefixes globally. 821 o SHOULD provide number of protected prefixes as well as non 822 protected prefixes per link. 824 o MAY provide number of protected prefixes as well as non protected 825 prefixes per priority if implementation supports prefix-priority 826 insertion in RIB/FIB. 828 o SHOULD provide a reason for chosing an alternate (policy and 829 criteria) and for excluding an alternate. 831 o SHOULD provide the list of non protected prefixes and the reason 832 why they are not protected (no protection required or no alternate 833 available). 835 6.4. Coverage monitoring 837 It is pretty easy to evaluate the coverage of a network in a nominal 838 situation, but topology changes may change the coverage. In some 839 situations, the network may no longer be able to provide the required 840 level of protection. Hence, it becomes very important for service 841 providers to get alerted about changes of coverage. 843 An implementation SHOULD : 845 o provide an alert system if total coverage (for a node) is below a 846 defined threshold or comes back to a normal situation. 848 o provide an alert system if coverage of a specific link is below a 849 defined threshold or comes back to a normal situation. 851 An implementation MAY : 853 o provide an alert system if a specific destination is not protected 854 anymore or when protection comes back up for this destination 856 Although the procedures for providing alerts are beyond the scope of 857 this document, we recommend that implementations consider standard 858 and well used mechanisms like syslog or SNMP traps. 860 6.5. LFA and network planning 862 The operator may choose to run simulations in order to ensure full 863 coverage of a certain type for the whole network or a given subset of 864 the network. This is particularly likely if he operates the network 865 in the sense of the third backbone profiles described in [RFC6571], 866 that is, he seeks to design and engineer the network topology in a 867 way that a certain coverage is always achieved. Obviously a complete 868 and exact simulation of the IP FRR coverage can only be achieved, if 869 the behavior is deterministic and if the algorithm used is available 870 to the simulation tool. Thus, an implementation SHOULD: 872 o Behave deterministic in its selection LFA process. I.e. in the 873 same topology and with the same policy configuration, the 874 implementation MUST always choose the same alternate for a given 875 prefix. 877 o Document its behavior. The implementation SHOULD provide enough 878 documentation of its behavior that allows an implementer of a 879 simulation tool, to foresee the exact choice of the LFA 880 implementation for every prefix in a given topology. This SHOULD 881 take into account all possible policy configuration options. One 882 possible way to document this behavior is to disclose the 883 algorithm used to choose alternates. 885 7. Security Considerations 887 This document does not introduce any change in security consideration 888 compared to [RFC5286]. 890 8. Contributors 892 Significant contributions were made by Pierre Francois, Hannes 893 Gredler, Chris Bowers, Jeff Tantsura, Uma Chunduri and Mustapha 894 Aissaoui which the authors would like to acknowledge. 896 9. Acknowledgements 898 10. IANA Considerations 900 This document has no action for IANA. 902 11. References 904 11.1. Normative References 906 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 907 Requirement Levels", BCP 14, RFC 2119, March 1997. 909 [RFC4203] Kompella, K. and Y. Rekhter, "OSPF Extensions in Support 910 of Generalized Multi-Protocol Label Switching (GMPLS)", 911 RFC 4203, October 2005. 913 [RFC4205] Kompella, K. and Y. Rekhter, "Intermediate System to 914 Intermediate System (IS-IS) Extensions in Support of 915 Generalized Multi-Protocol Label Switching (GMPLS)", 916 RFC 4205, October 2005. 918 [RFC5286] Atlas, A. and A. Zinin, "Basic Specification for IP Fast 919 Reroute: Loop-Free Alternates", RFC 5286, September 2008. 921 11.2. Informative References 923 [I-D.ietf-rtgwg-remote-lfa] 924 Bryant, S., Filsfils, C., Previdi, S., Shand, M., and S. 925 Ning, "Remote LFA FRR", draft-ietf-rtgwg-remote-lfa-04 926 (work in progress), November 2013. 928 [I-D.litkowski-rtgwg-lfa-rsvpte-cooperation] 929 Litkowski, S., Decraene, B., Filsfils, C., and K. Raza, 930 "Interactions between LFA and RSVP-TE", 931 draft-litkowski-rtgwg-lfa-rsvpte-cooperation-02 (work in 932 progress), August 2013. 934 [I-D.psarkar-isis-node-admin-tag] 935 psarkar@juniper.net, p., Gredler, H., Hegde, S., 936 Raghuveer, H., Litkowski, S., and B. Decraene, 937 "Advertising Per-node Admin Tags in IS-IS", 938 draft-psarkar-isis-node-admin-tag-00 (work in progress), 939 October 2013. 941 [I-D.psarkar-rtgwg-rlfa-node-protection] 942 psarkar@juniper.net, p., Gredler, H., Hegde, S., 943 Raghuveer, H., cbowers@juniper.net, c., and S. Litkowski, 944 "Remote-LFA Node Protection and Manageability", 945 draft-psarkar-rtgwg-rlfa-node-protection-03 (work in 946 progress), December 2013. 948 [RFC3630] Katz, D., Kompella, K., and D. Yeung, "Traffic Engineering 949 (TE) Extensions to OSPF Version 2", RFC 3630, 950 September 2003. 952 [RFC3906] Shen, N. and H. Smit, "Calculating Interior Gateway 953 Protocol (IGP) Routes Over Traffic Engineering Tunnels", 954 RFC 3906, October 2004. 956 [RFC4090] Pan, P., Swallow, G., and A. Atlas, "Fast Reroute 957 Extensions to RSVP-TE for LSP Tunnels", RFC 4090, 958 May 2005. 960 [RFC5305] Li, T. and H. Smit, "IS-IS Extensions for Traffic 961 Engineering", RFC 5305, October 2008. 963 [RFC5714] Shand, M. and S. Bryant, "IP Fast Reroute Framework", 964 RFC 5714, January 2010. 966 [RFC5715] Shand, M. and S. Bryant, "A Framework for Loop-Free 967 Convergence", RFC 5715, January 2010. 969 [RFC6571] Filsfils, C., Francois, P., Shand, M., Decraene, B., 970 Uttaro, J., Leymann, N., and M. Horneffer, "Loop-Free 971 Alternate (LFA) Applicability in Service Provider (SP) 972 Networks", RFC 6571, June 2012. 974 Authors' Addresses 976 Stephane Litkowski 977 Orange 979 Email: stephane.litkowski@orange.com 980 Bruno Decraene 981 Orange 983 Email: bruno.decraene@orange.com 985 Clarence Filsfils 986 Cisco Systems 988 Email: cfilsfil@cisco.com 990 Kamran Raza 991 Cisco Systems 993 Email: skraza@cisco.com 995 Martin Horneffer 996 Deutsche Telekom 998 Email: Martin.Horneffer@telekom.de 1000 Pushpasis Sarkar 1001 Juniper Networks 1003 Email: psarkar@juniper.net