idnits 2.17.1 draft-ietf-lsvr-bgp-spf-06.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 seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (September 30, 2019) is 1669 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) == Missing Reference: 'RFC2328' is mentioned on line 966, but not defined == Missing Reference: 'RFC5286' is mentioned on line 1000, but not defined == Missing Reference: 'RFC4456' is mentioned on line 970, but not defined == Missing Reference: 'RFC4915' is mentioned on line 995, but not defined == Missing Reference: 'RFC5549' is mentioned on line 1005, but not defined ** Obsolete undefined reference: RFC 5549 (Obsoleted by RFC 8950) == Missing Reference: 'RFC4790' is mentioned on line 990, but not defined == Missing Reference: 'RFC5880' is mentioned on line 1010, but not defined == Missing Reference: 'I-D.ietf-lsvr-applicability' is mentioned on line 960, but not defined == Missing Reference: 'RFC4760' is mentioned on line 985, but not defined == Missing Reference: 'RFC4750' is mentioned on line 980, but not defined == Missing Reference: 'RFC4724' is mentioned on line 975, but not defined ** Obsolete normative reference: RFC 7752 (Obsoleted by RFC 9552) ** Downref: Normative reference to an Informational RFC: RFC 7938 Summary: 3 errors (**), 0 flaws (~~), 13 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Patel 3 Internet-Draft Arrcus, Inc. 4 Intended status: Standards Track A. Lindem 5 Expires: April 2, 2020 Cisco Systems 6 S. Zandi 7 Linkedin 8 W. Henderickx 9 Nokia 10 September 30, 2019 12 Shortest Path Routing Extensions for BGP Protocol 13 draft-ietf-lsvr-bgp-spf-06 15 Abstract 17 Many Massively Scaled Data Centers (MSDCs) have converged on 18 simplified layer 3 routing. Furthermore, requirements for 19 operational simplicity have lead many of these MSDCs to converge on 20 BGP as their single routing protocol for both their fabric routing 21 and their Data Center Interconnect (DCI) routing. This document 22 describes a solution which leverages BGP Link-State distribution and 23 the Shortest Path First (SPF) algorithm similar to Internal Gateway 24 Protocols (IGPs) such as OSPF. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at https://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on April 2, 2020. 43 Copyright Notice 45 Copyright (c) 2019 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (https://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 This document may contain material from IETF Documents or IETF 59 Contributions published or made publicly available before November 60 10, 2008. The person(s) controlling the copyright in some of this 61 material may not have granted the IETF Trust the right to allow 62 modifications of such material outside the IETF Standards Process. 63 Without obtaining an adequate license from the person(s) controlling 64 the copyright in such materials, this document may not be modified 65 outside the IETF Standards Process, and derivative works of it may 66 not be created outside the IETF Standards Process, except to format 67 it for publication as an RFC or to translate it into languages other 68 than English. 70 Table of Contents 72 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 73 1.1. BGP Shortest Path First (SPF) Motivation . . . . . . . . 4 74 1.2. Requirements Language . . . . . . . . . . . . . . . . . . 5 75 2. BGP Peering Models . . . . . . . . . . . . . . . . . . . . . 5 76 2.1. BGP Single-Hop Peering on Network Node Connections . . . 5 77 2.2. BGP Peering Between Directly Connected Network Nodes . . 6 78 2.3. BGP Peering in Route-Reflector or Controller Topology . . 6 79 3. BGP-LS Shortest Path Routing (SPF) SAFI . . . . . . . . . . . 6 80 4. Extensions to BGP-LS . . . . . . . . . . . . . . . . . . . . 7 81 4.1. Node NLRI Usage . . . . . . . . . . . . . . . . . . . . . 7 82 4.1.1. Node NLRI Attribute SPF Capability TLV . . . . . . . 7 83 4.1.2. BGP-LS Node NLRI Attribute SPF Status TLV . . . . . . 8 84 4.2. Link NLRI Usage . . . . . . . . . . . . . . . . . . . . . 9 85 4.2.1. BGP-LS Link NLRI Attribute Prefix-Length TLVs . . . . 9 86 4.2.2. BGP-LS Link NLRI Attribute SPF Status TLV . . . . . . 10 87 4.3. Prefix NLRI Usage . . . . . . . . . . . . . . . . . . . . 10 88 4.3.1. BGP-LS Prefix NLRI Attribute SPF Status TLV . . . . . 11 89 4.4. BGP-LS Attribute Sequence-Number TLV . . . . . . . . . . 11 90 5. Decision Process with SPF Algorithm . . . . . . . . . . . . . 12 91 5.1. Phase-1 BGP NLRI Selection . . . . . . . . . . . . . . . 13 92 5.2. Dual Stack Support . . . . . . . . . . . . . . . . . . . 14 93 5.3. SPF Calculation based on BGP-LS NLRI . . . . . . . . . . 14 94 5.4. NEXT_HOP Manipulation . . . . . . . . . . . . . . . . . . 17 95 5.5. IPv4/IPv6 Unicast Address Family Interaction . . . . . . 17 96 5.6. NLRI Advertisement and Convergence . . . . . . . . . . . 17 97 5.6.1. Link/Prefix Failure Convergence . . . . . . . . . . . 17 98 5.6.2. Node Failure Convergence . . . . . . . . . . . . . . 18 99 5.7. Error Handling . . . . . . . . . . . . . . . . . . . . . 18 100 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 19 101 7. Security Considerations . . . . . . . . . . . . . . . . . . . 19 102 8. Management Considerations . . . . . . . . . . . . . . . . . . 19 103 8.1. Configuration . . . . . . . . . . . . . . . . . . . . . . 19 104 8.2. Operational Data . . . . . . . . . . . . . . . . . . . . 19 105 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 20 106 10. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 20 107 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 20 108 11.1. Normative References . . . . . . . . . . . . . . . . . . 20 109 11.2. Information References . . . . . . . . . . . . . . . . . 21 110 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 23 112 1. Introduction 114 Many Massively Scaled Data Centers (MSDCs) have converged on 115 simplified layer 3 routing. Furthermore, requirements for 116 operational simplicity have lead many of these MSDCs to converge on 117 BGP [RFC4271] as their single routing protocol for both their fabric 118 routing and their Data Center Interconnect (DCI) routing. 119 Requirements and procedures for using BGP are described in [RFC7938]. 120 This document describes an alternative solution which leverages BGP- 121 LS [RFC7752] and the Shortest Path First algorithm similar to 122 Internal Gateway Protocols (IGPs) such as OSPF [RFC2328]. 124 [RFC4271] defines the Decision Process that is used to select routes 125 for subsequent advertisement by applying the policies in the local 126 Policy Information Base (PIB) to the routes stored in its Adj-RIBs- 127 In. The output of the Decision Process is the set of routes that are 128 announced by a BGP speaker to its peers. These selected routes are 129 stored by a BGP speaker in the speaker's Adj-RIBs-Out according to 130 policy. 132 [RFC7752] describes a mechanism by which link-state and TE 133 information can be collected from networks and shared with external 134 components using BGP. This is achieved by defining NLRI advertised 135 within the BGP-LS/BGP-LS-SPF AFI/SAFI. The BGP-LS extensions defined 136 in [RFC7752] makes use of the Decision Process defined in [RFC4271]. 138 This document augments [RFC7752] by replacing its use of the existing 139 Decision Process. Rather than reusing the BGP-LS SAFI, the BGP-LS- 140 SPF SAFI is introduced to insure backward compatibility. The Phase 1 141 and 2 decision functions of the Decision Process are replaced with 142 the Shortest Path First (SPF) algorithm also known as the Dijkstra 143 algorithm. The Phase 3 decision function is also simplified since it 144 is no longer dependent on the previous phases. This solution avails 145 the benefits of both BGP and SPF-based IGPs. These include TCP based 146 flow-control, no periodic link-state refresh, and completely 147 incremental NLRI advertisement. These advantages can reduce the 148 overhead in MSDCs where there is a high degree of Equal Cost Multi- 149 Path (ECMPs) and the topology is very stable. Additionally, using an 150 SPF-based computation can support fast convergence and the 151 computation of Loop-Free Alternatives (LFAs) [RFC5286] in the event 152 of link failures. Furthermore, a BGP based solution lends itself to 153 multiple peering models including those incorporating route- 154 reflectors [RFC4456] or controllers. 156 Support for Multiple Topology Routing (MTR) as described in [RFC4915] 157 is an area for further study dependent on deployment requirements. 159 1.1. BGP Shortest Path First (SPF) Motivation 161 Given that [RFC7938] already describes how BGP could be used as the 162 sole routing protocol in an MSDC, one might question the motivation 163 for defining an alternate BGP deployment model when a mature solution 164 exists. For both alternatives, BGP offers the operational benefits 165 of a single routing protocol. However, BGP SPF offers some unique 166 advantages above and beyond standard BGP distance-vector routing. 168 A primary advantage is that all BGP speakers in the BGP SPF routing 169 domain will have a complete view of the topology. This will allow 170 support for ECMP, IP fast-reroute (e.g., Loop-Free Alternatives), 171 Shared Risk Link Groups (SRLGs), and other routing enhancements 172 without advertisement of addition BGP paths or other extensions. In 173 short, the advantages of an IGP such as OSPF [RFC2328] are availed in 174 BGP. 176 With the simplified BGP decision process as defined in Section 5.1, 177 NLRI changes can be disseminated throughout the BGP routing domain 178 much more rapidly (equivalent to IGPs with the proper 179 implementation). 181 Another primary advantage is a potential reduction in NLRI 182 advertisement. With standard BGP distance-vector routing, a single 183 link failure may impact 100s or 1000s prefixes and result in the 184 withdrawal or re-advertisement of the attendant NLRI. With BGP SPF, 185 only the BGP speakers corresponding to the link NLRI need withdraw 186 the corresponding BGP-LS Link NLRI. This advantage will contribute 187 to both faster convergence and better scaling. 189 With controller and route-reflector peering models, BGP SPF 190 advertisement and distributed computation require a minimal number of 191 sessions and copies of the NLRI since only the latest version of the 192 NLRI from the originator is required. Given that verification of the 193 adjacencies is done outside of BGP (see Section 2), each BGP speaker 194 will only need as many sessions and copies of the NLRI as required 195 for redundancy (e.g., one for the SPF computation and another for 196 backup). Functions such as Optimized Route Reflection (ORR) are 197 supported without extension by virtue of the primary advantages. 198 Additionally, a controller could inject topology that is learned 199 outside the BGP routing domain. 201 Given that controllers are already consuming BGP-LS NLRI [RFC7752], 202 reusing for the BGP-LS SPF leverages the existing controller 203 implementations. 205 Another potential advantage of BGP SPF is that both IPv6 and IPv4 can 206 be supported in the same address family using the same topology. 207 Although not described in this version of the document, multi- 208 topology extensions can be used to support separate IPv4, IPv6, 209 unicast, and multicast topologies while sharing the same NLRI. 211 Finally, the BGP SPF topology can be used as an underlay for other 212 BGP address families (using the existing model) and realize all the 213 above advantages. A simplified peering model using IPv6 link-local 214 addresses as next-hops can be deployed similar to [RFC5549]. 216 1.2. Requirements Language 218 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 219 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 220 "OPTIONAL" in this document are to be interpreted as described in BCP 221 14 [RFC2119] [RFC8174] when, and only when, they appear in all 222 capitals, as shown here. 224 2. BGP Peering Models 226 Depending on the requirements, scaling, and capabilities of the BGP 227 speakers, various peering models are supported. The only requirement 228 is that all BGP speakers in the BGP SPF routing domain receive link- 229 state NLRI on a timely basis, run an SPF calculation, and update 230 their data plane appropriately. The content of the Link NLRI is 231 described in Section 4.2. 233 2.1. BGP Single-Hop Peering on Network Node Connections 235 The simplest peering model is the one described in section 5.2.1 of 236 [RFC7938]. In this model, EBGP single-hop sessions are established 237 over direct point-to-point links interconnecting the SPF domain 238 nodes. For the purposes of BGP SPF, Link NLRI is only advertised if 239 a single-hop BGP session has been established and the Link-State/SPF 240 address family capability has been exchanged [RFC4790] on the 241 corresponding session. If the session goes down, the corresponding 242 Link NLRI will be withdrawn. Topologically, this would be equivalent 243 to the peering model in [RFC7938] where there is a BGP session on 244 every link in the data center switch fabric. 246 2.2. BGP Peering Between Directly Connected Network Nodes 248 In this model, BGP speakers peer with all directly connected network 249 nodes but the sessions may be multi-hop and the direct connection 250 discovery and liveliness detection for those connections are 251 independent of the BGP protocol. How this is accomplished is outside 252 the scope of this document. Consequently, there will be a single 253 session even if there are multiple direct connections between BGP 254 speakers. For the purposes of BGP SPF, Link NLRI is advertised as 255 long as a BGP session has been established, the Link-State/SPF 256 address family capability has been exchanged [RFC4790] and the 257 corresponding link is considered is up and considered operational. 258 This is much like the previous peering model only peering is on a 259 single loopback address and the switch fabric links can be 260 unnumbered. However, there will be the same unnumber of sessions as 261 with the previous peering model unless there are parrallel links 262 between switches in the fabric. 264 2.3. BGP Peering in Route-Reflector or Controller Topology 266 In this model, BGP speakers peer solely with one or more Route 267 Reflectors [RFC4456] or controllers. As in the previous model, 268 direct connection discovery and liveliness detection for those 269 connections are done outside the BGP protocol. More specifically, 270 the Liveliness detection is done using BFD protocol described in 271 [RFC5880]. For the purposes of BGP SPF, Link NLRI is advertised as 272 long as the corresponding link is up and considered operational. 274 This peering model, known as sparse peering, allows for many fewer 275 BGP sessions and, consequently, instances of the same NLRI received 276 from multiple peers. It is discussed in greater detail in 277 [I-D.ietf-lsvr-applicability]. 279 3. BGP-LS Shortest Path Routing (SPF) SAFI 281 In order to replace the Phase 1 and 2 decision functions of the 282 existing Decision Process with an SPF-based Decision Process and 283 streamline the Phase 3 decision functions in a backward compatible 284 manner, this draft introduces the BGP-LS-SFP SAFI for BGP-LS SPF 285 operation. The BGP-LS-SPF (AF 16388 / SAFI TBD1) [RFC4790] is 286 allocated by IANA as specified in the Section 6. A BGP speaker using 287 the BGP-LS SPF extensions described herein MUST exchange the AFI/SAFI 288 using Multiprotocol Extensions Capability Code [RFC4760] with other 289 BGP speakers in the SPF routing domain. 291 4. Extensions to BGP-LS 293 [RFC7752] describes a mechanism by which link-state and TE 294 information can be collected from networks and shared with external 295 components using BGP protocol. It describes both the definition of 296 BGP-LS NLRI that describes links, nodes, and prefixes comprising IGP 297 link-state information and the definition of a BGP path attribute 298 (BGP-LS attribute) that carries link, node, and prefix properties and 299 attributes, such as the link and prefix metric or auxiliary Router- 300 IDs of nodes, etc. 302 The BGP protocol will be used in the Protocol-ID field specified in 303 table 1 of [I-D.ietf-idr-bgpls-segment-routing-epe]. The local and 304 remote node descriptors for all NLRI will be the BGP Router-ID (TLV 305 516) and either the AS Number (TLV 512) [RFC7752] or the BGP 306 Confederation Member (TLV 517) [RFC8402]. However, if the BGP 307 Router-ID is known to be unique within the BGP Routing domain, it can 308 be used as the sole descriptor. 310 4.1. Node NLRI Usage 312 The BGP Node NLRI will be advertised unconditionally by all routers 313 in the BGP SPF routing domain. 315 4.1.1. Node NLRI Attribute SPF Capability TLV 317 The SPF capability is a new Node Attribute TLV that will be added to 318 those defined in table 7 of [RFC7752]. The new attribute TLV will 319 only be applicable when BGP is specified in the Node NLRI Protocol ID 320 field. The TBD TLV type will be defined by IANA. The new Node 321 Attribute TLV will contain a single-octet SPF algorithm as defined in 322 [RFC8402]. 324 0 1 2 3 325 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 326 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 327 | Type | Length | 328 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 329 | SPF Algorithm | 330 +-+-+-+-+-+-+-+-+ 332 The SPF Algorithm may take the following values: 334 0 - Normal Shortest Path First (SPF) algorithm based on link 335 metric. This is the standard shortest path algorithm as 336 computed by the IGP protocol. Consistent with the deployed 337 practice for link-state protocols, Algorithm 0 permits any 338 node to overwrite the SPF path with a different path based on 339 its local policy. 340 1 - Strict Shortest Path First (SPF) algorithm based on link 341 metric. The algorithm is identical to Algorithm 0 but Algorithm 342 1 requires that all nodes along the path will honor the SPF 343 routing decision. Local policy at the node claiming support for 344 Algorithm 1 MUST NOT alter the SPF paths computed by Algorithm 1. 346 Note that usage of Strict Shortest Path First (SPF) algorithm is 347 defined in the IGP algorithm registry but usage is restricted to 348 [I-D.ietf-idr-bgpls-segment-routing-epe]. Hence, its usage for BGP- 349 LS SPF is out of scope. 351 When computing the SPF for a given BGP routing domain, only BGP nodes 352 advertising the SPF capability attribute will be included the 353 Shortest Path Tree (SPT). 355 4.1.2. BGP-LS Node NLRI Attribute SPF Status TLV 357 A BGP-LS Attribute TLV to BGP-LS Node NLRI is defined to indicate the 358 status of the node with respect to the BGP SPF calculation. This 359 will be used to rapidly take a node out of service or to indicate the 360 node is not to be used for transit (i.e., non-local) traffic. If the 361 SPF Status TLV is not included with the Node NLRI, the node is 362 considered to be up and is available for transit traffic. 364 0 1 2 3 365 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 366 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 367 | TBD Type | Length | 368 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 369 | SPF Status | 370 +-+-+-+-+-+-+-+-+ 372 BGP Status Values: 0 - Reserved 373 1 - Node Unreachable with respect to BGP SPF 374 2 - Node does not support transit with respect 375 to BGP SPF 376 3-254 - Undefined 377 255 - Reserved 379 4.2. Link NLRI Usage 381 The criteria for advertisement of Link NLRI are discussed in 382 Section 2. 384 Link NLRI is advertised with local and remote node descriptors as 385 described above and unique link identifiers dependent on the 386 addressing. For IPv4 links, the links local IPv4 (TLV 259) and 387 remote IPv4 (TLV 260) addresses will be used. For IPv6 links, the 388 local IPv6 (TLV 261) and remote IPv6 (TLV 262) addresses will be 389 used. For unnumbered links, the link local/remote identifiers (TLV 390 258) will be used. For links supporting having both IPv4 and IPv6 391 addresses, both sets of descriptors may be included in the same Link 392 NLRI. The link identifiers are described in table 5 of [RFC7752]. 394 The link IGP metric attribute TLV (TLV 1095) as well as any others 395 required for non-SPF purposes SHOULD be advertised. Algorithms such 396 as setting the metric inversely to the link speed as done in the OSPF 397 MIB [RFC4750] MAY be supported. However, this is beyond the scope of 398 this document. 400 4.2.1. BGP-LS Link NLRI Attribute Prefix-Length TLVs 402 Two BGP-LS Attribute TLVs to BGP-LS Link NLRI are defined to 403 advertise the prefix length associated with the IPv4 and IPv6 link 404 prefixes. The prefix length is used for the optional installation of 405 prefixes corresponding to Link NLRI as defined in Section 5.3. 407 0 1 2 3 408 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 409 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 410 | TBD IPv4 or IPv6 Type | Length | 411 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 412 | Prefix-Length | 413 +-+-+-+-+-+-+-+-+ 415 Prefix-length - A one-octet length restricted to 1-32 for IPv4 416 Link NLIR endpoint prefixes and 1-128 for IPv6 417 Link NLRI endpoint prefixes. 419 4.2.2. BGP-LS Link NLRI Attribute SPF Status TLV 421 A BGP-LS Attribute TLV to BGP-LS Link NLRI is defined to indicate the 422 status of the link with respect to the BGP SPF calculation. This 423 will be used to expedite convergence for link failures as discussed 424 in Section 5.6.1. If the SPF Status TLV is not included with the 425 Link NLRI, the link is considered up and available. 427 0 1 2 3 428 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 429 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 430 | TBD Type | Length | 431 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 432 | SPF Status | 433 +-+-+-+-+-+-+-+-+ 435 BGP Status Values: 0 - Reserved 436 1 - Link Unreachable with respect to BGP SPF 437 2-254 - Undefined 438 255 - Reserved 440 4.3. Prefix NLRI Usage 442 Prefix NLRI is advertised with a local node descriptor as described 443 above and the prefix and length used as the descriptors (TLV 265) as 444 described in [RFC7752]. The prefix metric attribute TLV (TLV 1155) 445 as well as any others required for non-SPF purposes SHOULD be 446 advertised. For loopback prefixes, the metric should be 0. For non- 447 loopback prefixes, the setting of the metric is a local matter and 448 beyond the scope of this document. 450 4.3.1. BGP-LS Prefix NLRI Attribute SPF Status TLV 452 A BGP-LS Attribute TLV to BGP-LS Prefix NLRI is defined to indicate 453 the status of the prefix with respect to the BGP SPF calculation. 454 This will be used to expedite convergence for prefix unreachability 455 as discussed in Section 5.6.1. If the SPF Status TLV is not included 456 with the Prefix NLRI, the prefix is considered reachable. 458 0 1 2 3 459 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 460 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 461 | TBD Type | Length | 462 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 463 | SPF Status | 464 +-+-+-+-+-+-+-+-+ 466 BGP Status Values: 0 - Reserved 467 1 - Prefix down with respect to SPF 468 2-254 - Undefined 469 255 - Reserved 471 4.4. BGP-LS Attribute Sequence-Number TLV 473 A new BGP-LS Attribute TLV to BGP-LS NLRI types is defined to assure 474 the most recent version of a given NLRI is used in the SPF 475 computation. The TBD TLV type will be defined by IANA. The new BGP- 476 LS Attribute TLV will contain an 8-octet sequence number. The usage 477 of the Sequence Number TLV is described in Section 5.1. 479 0 1 2 3 480 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 481 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 482 | Type | Length | 483 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 484 | Sequence Number (High-Order 32 Bits) | 485 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 486 | Sequence Number (Low-Order 32 Bits) | 487 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 489 Sequence Number 491 The 64-bit strictly increasing sequence number is incremented for 492 every version of BGP-LS NLRI originated. BGP speakers implementing 493 this specification MUST use available mechanisms to preserve the 494 sequence number's strictly increasing property for the deployed life 495 of the BGP speaker (including cold restarts). One mechanism for 496 accomplishing this would be to use the high-order 32 bits of the 497 sequence number as a wrap/boot count that is incremented anytime the 498 BGP router loses its sequence number state or the low-order 32 bits 499 wrap. 501 When incrementing the sequence number for each self-originated NLRI, 502 the sequence number should be treated as an unsigned 64-bit value. 503 If the lower-order 32-bit value wraps, the higher-order 32-bit value 504 should be incremented and saved in non-volatile storage. If by some 505 chance the BGP Speaker is deployed long enough that there is a 506 possibility that the 64-bit sequence number may wrap or a BGP Speaker 507 completely loses its sequence number state (e.g., the BGP speaker 508 hardware is replaced or experiences a cold-start), the phase 1 509 decision function (see Section 5.1) rules will insure convergence, 510 albeit, not immediately. 512 5. Decision Process with SPF Algorithm 514 The Decision Process described in [RFC4271] takes place in three 515 distinct phases. The Phase 1 decision function of the Decision 516 Process is responsible for calculating the degree of preference for 517 each route received from a BGP speaker's peer. The Phase 2 decision 518 function is invoked on completion of the Phase 1 decision function 519 and is responsible for choosing the best route out of all those 520 available for each distinct destination, and for installing each 521 chosen route into the Loc-RIB. The combination of the Phase 1 and 2 522 decision functions is characterized as a Path Vector algorithm. 524 The SPF based Decision process replaces the BGP best-path Decision 525 process described in [RFC4271]. This process starts with selecting 526 only those Node NLRI whose SPF capability TLV matches with the local 527 BGP speaker's SPF capability TLV value. Since Link-State NLRI always 528 contains the local descriptor [RFC7752], it will only be originated 529 by a single BGP speaker in the BGP routing domain. These selected 530 Node NLRI and their Link/Prefix NLRI are used to build a directed 531 graph during the SPF computation. The best paths for BGP prefixes 532 are installed as a result of the SPF process. 534 When BGP-LS-SPF NLRI is received, all that is required is to 535 determine whether it is the best-path by examining the Node-ID and 536 sequence number as described in Section 5.1. If the received best- 537 path NLRI had changed, it will be advertised to other BGP-LS-SPF 538 peers. If the attributes have changed (other than the sequence 539 number), a BGP SPF calculation will be scheduled. However, a changed 540 NLRI MAY be advertised to other peers almost immediately and 541 propagation of changes can approach IGP convergence times. To 542 accomplish this, the MinRouteAdvertisementIntervalTimer and 543 MinASOriginationIntervalTimer [RFC4271] are not applicable to the 544 BGP-LS-SPF SAFI. Rather, SPF calculations SHOULD be triggered and 545 dampened consistent with the SPF backoff algorithm specified in 546 [RFC8405]. 548 The Phase 3 decision function of the Decision Process [RFC4271] is 549 also simplified since under normal SPF operation, a BGP speaker would 550 advertise the NLRI selected for the SPF to all BGP peers with the 551 BGP-LS/BGP-LS-SPF AFI/SAFI. Application of policy would not be 552 prevented however its usage to best-path process would be limited as 553 the SPF relies solely on link metrics. 555 5.1. Phase-1 BGP NLRI Selection 557 The rules for NLRI selection are greatly simplified from [RFC4271]. 559 1. If the NLRI is received from the BGP speaker originating the NLRI 560 (as determined by the comparing BGP Router ID in the NLRI Node 561 identifiers with the BGP speaker Router ID), then it is preferred 562 over the same NLRI from non-originators. This rule will assure 563 that stale NLRI is updated even if a BGP-LS router loses its 564 sequence number state due to a cold-start. 566 2. If the Sequence-Number TLV is present in the BGP-LS Attribute, 567 then the NLRI with the most recent, i.e., highest sequence number 568 is selected. BGP-LS NLRI with a Sequence-Number TLV will be 569 considered more recent than NLRI without a BGP-LS Attribute or a 570 BGP-LS Attribute that doesn't include the Sequence-Number TLV. 572 3. The final tie-breaker is the NLRI from the BGP Speaker with the 573 numerically largest BGP Router ID. 575 When a BGP speaker completely loses its sequence number state, i.e., 576 due to a cold start, or in the unlikely possibility that that 577 sequence number wraps, the BGP routing domain will still converge. 578 This is due to the fact that BGP speakers adjacent to the router will 579 always accept self-originated NLRI from the associated speaker as 580 more recent (rule # 1). When BGP speaker reestablishes a connection 581 with its peers, any existing session will be taken down and stale 582 NLRI will be replaced by the new NLRI and stale NLRI will be 583 discarded independent of whether or not BGP graceful restart is 584 deployed, [RFC4724]. The adjacent BGP speaker will update their NLRI 585 advertisements in turn until the BGP routing domain has converged. 587 The modified SPF Decision Process performs an SPF calculation rooted 588 at the BGP speaker using the metrics from Link and Prefix NLRI 589 Attribute TLVs [RFC7752]. As a result, any attributes that would 590 influence the Decision process defined in [RFC4271] like ORIGIN, 591 MULTI_EXIT_DISC, and LOCAL_PREF attributes are ignored by the SPF 592 algorithm. Furthermore, the NEXT_HOP attribute value is preserved 593 but otherwise ignored during the SPF or best-path. 595 5.2. Dual Stack Support 597 The SPF-based decision process operates on Node, Link, and Prefix 598 NLRIs that support both IPv4 and IPv6 addresses. Whether to run a 599 single SPF instance or multiple SPF instances for separate AFs is a 600 matter of a local implementation. Normally, IPv4 next-hops are 601 calculated for IPv4 prefixes and IPv6 next-hops are calculated for 602 IPv6 prefixes. However, an interesting use-case is deployment of 603 [RFC5549] where IPv6 next-hops are calculated for both IPv4 and IPv6 604 prefixes. As stated in Section 1, support for Multiple Topology 605 Routing (MTR) is an area for future study. 607 5.3. SPF Calculation based on BGP-LS NLRI 609 This section details the BGP-LS SPF local routing information base 610 (RIB) calculation. The router will use BGP-LS Node, Link, and Prefix 611 NLRI to populate the local RIB using the following algorithm. This 612 calculation yields the set of intra-area routes associated with the 613 BGP-LS domain. A router calculates the shortest-path tree using 614 itself as the root. Variations and optimizations of the algorithm 615 are valid as long as it yields the same set of routes. The algorithm 616 below supports Equal Cost Multi-Path (ECMP) routes. Weighted Unequal 617 Cost Multi-Path are out of scope. The organization of this section 618 owes heavily to section 16 of [RFC2328]. 620 The following abstract data structures are defined in order to 621 specify the algorithm. 623 o Local Route Information Base (RIB) - This is abstract contains 624 reachability information (i.e., next hops) for all prefixes (both 625 IPv4 and IPv6) as well as the Node NLRI reachability. 626 Implementations may choose to implement this as separate RIBs for 627 each address family and/or Node NLRI. 629 o Link State NLRI Database (LSNDB) - Database of BGP-LS NLRI that 630 facilitates access to all Node, Link, and Prefix NLRI as well as 631 all the Link and Prefix NLRI corresponding to a given Node NLRI. 632 Other optimization, such as, resolving bi-directional connectivity 633 associations between Link NLRI are possible but of scope of this 634 document. 636 o Candidate List - This is a list of candidate Node NLRI with the 637 lowest cost Node NLRI at the front of the list. It is typically 638 implemented as a heap but other concrete data structures have also 639 been used. 641 The algorithm is comprised of the steps below: 643 1. The current local RIB is invalidated. The local RIB is built 644 again from scratch. The existing routing entries are preserved 645 for comparision to determine changes that need to be installed in 646 the global RIB. 648 2. The computing router's Node NLRI is installed in the local RIB 649 with a cost of 0 and as as the sole entry in the candidate list. 651 3. The Node NLRI with the lowest cost is removed from the candidate 652 list for processing. If the BGP-LS Node attribute includes an 653 SPF Status TLV (Section 4.1.2) indicating the node is 654 unreachable, the Node NLRI is ignored and the next lowest cost 655 Node NLRI is selected from candidate list. The Node 656 corresponding to this NLRI will be referred to as the Current 657 Node. If the candidate list is empty, the SPF calculation has 658 completed and the algorithm proceeds to step 6. 660 4. All the Prefix NLRI with the same Node Identifiers as the Current 661 Node will be considered for installation. The cost for each 662 prefix is the metric advertised in the Prefix NLRI added to the 663 cost to reach the Current Node. 665 * If the BGP-LS Prefix attribute includes an SPF Status TLV 666 indicating the prefix is unreachable, the BGP-LS Prefix NLRI 667 is considered unreachable and the next BGP-LS Prefix NLRI is 668 examined. 670 * If the prefix is in the local RIB and the cost is greater than 671 the Current route's metric, the Prefix NLRI does not 672 contribute to the route and is ignored. 674 * If the prefix is in the local RIB and the cost is less than 675 the current route's metric, the Prefix is installed with the 676 Current Node's next-hops replacing the local RIB route's next- 677 hops and the metric being updated. 679 * If the prefix is in the local RIB and the cost is same as the 680 current route's metric, the Prefix is installed with the 681 Current Node's next-hops being merged with local RIB route's 682 next-hops. 684 5. All the Link NLRI with the same Node Identifiers as the Current 685 Node will be considered for installation. Each link will be 686 examined and will be referred to in the following text as the 687 Current Link. The cost of the Current Link is the advertised 688 metric in the Link NLRI added to the cost to reach the Current 689 Node. 691 * Optionally, the prefix(es) associated with the Current Link 692 are installed into the local RIB using the same rules as were 693 used for Prefix NLRI in the previous steps. 695 * If the current Node NLRI attributes includes the SPF status 696 TLV (Section 4.1.2) and the status indicates that the Node 697 doesn't support transit, the next link for the current node is 698 processed. 700 * The Current Link's endpoint Node NLRI is accessed (i.e., the 701 Node NLRI with the same Node identifiers as the Link 702 endpoint). If it exists, it will be referred to as the 703 Endpoint Node NLRI and the algorithm will proceed as follows: 705 + If the BGP-LS Link NLRI attribute includes an SPF Status 706 TLV indicating the link is down, the BGP-LS Link NLRI is 707 considered down and the next BGP-LS Link NLRI is examined. 709 + All the Link NLRI corresponding the Endpoint Node NLRI will 710 be searched for a back-link NLRI pointing to the current 711 node. Both the Node identifiers and the Link endpoint 712 identifiers in the Endpoint Node's Link NLRI must match for 713 a match. If there is no corresponding Link NLRI 714 corresponding to the Endpoint Node NLRI, the Endpoint Node 715 NLIR fails the bi-directional connectivity test and is not 716 processed further. 718 + If the Endpoint Node NLRI is not on the candidate list, it 719 is inserted based on the link cost and BGP Identifier (the 720 latter being used as a tie-breaker). 722 + If the Endpoint Node NLRI is already on the candidate list 723 with a lower cost, it need not be inserted again. 725 + If the Endpoint Node NLRI is already on the candidate list 726 with a higher cost, it must be removed and reinserted with 727 a lower cost. 729 * Return to step 3 to process the next lowest cost Node NLRI on 730 the candidate list. 732 6. The local RIB is examined and changes (adds, deletes, 733 modifications) are installed into the global RIB. 735 5.4. NEXT_HOP Manipulation 737 A BGP speaker that supports SPF extensions MAY interact with peers 738 that don't support SPF extensions. If the BGP-LS address family is 739 advertised to a peer not supporting the SPF extensions described 740 herein, then the BGP speaker MUST conform to the NEXT_HOP rules 741 specified in [RFC4271] when announcing the Link-State address family 742 routes to those peers. 744 All BGP peers that support SPF extensions would locally compute the 745 Loc-RIB next-hops as a result of the SPF process. Consequently, the 746 NEXT_HOP attribute is always ignored on receipt. However, BGP 747 speakers SHOULD set the NEXT_HOP address according to the NEXT_HOP 748 attribute rules specified in [RFC4271]. 750 5.5. IPv4/IPv6 Unicast Address Family Interaction 752 While the BGP-LS SPF address family and the IPv4/IPv6 unicast address 753 families install routes into the same device routing tables, they 754 will operate independently much the same as OSPF and IS-IS would 755 operate today (i.e., "Ships-in-the-Night" mode). There will be no 756 implicit route redistribution between the BGP address families. 757 However, implementation specific redistribution mechanisms SHOULD be 758 made available with the restriction that redistribution of BGP-LS SPF 759 routes into the IPv4 address family applies only to IPv4 routes and 760 redistribution of BGP-LS SPF route into the IPv6 address family 761 applies only to IPv6 routes. 763 Given the fact that SPF algorithms are based on the assumption that 764 all routers in the routing domain calculate the precisely the same 765 SPF tree and install the same set of routes, it is RECOMMENDED that 766 BGP-LS SPF IPv4/IPv6 routes be given priority by default when 767 installed into their respective RIBs. In common implementations the 768 prioritization is governed by route preference or administrative 769 distance with lower being more preferred. 771 5.6. NLRI Advertisement and Convergence 773 5.6.1. Link/Prefix Failure Convergence 775 A local failure will prevent a link from being used in the SPF 776 calculation due to the IGP bi-directional connectivity requirement. 777 Consequently, local link failures should always be given priority 778 over updates (e.g., withdrawing all routes learned on a session) in 779 order to ensure the highest priority propagation and optimal 780 convergence. 782 An IGP such as OSPF [RFC2328] will stop using the link as soon as the 783 Router-LSA for one side of the link is received. With normal BGP 784 advertisement, the link would continue to be used until the last copy 785 of the BGP-LS Link NLRI is withdrawn. In order to avoid this delay, 786 the originator of the Link NLRI will advertise a more recent version 787 of the BGP-LS Link NLRI including the SPF Status TLV Section 4.2.2 788 indicating the link is down with respect to BGP SPF. After some 789 configurable period of time, e.g., 2-3 seconds, the BGP-LS Link NLRI 790 can be withdrawn with no consequence. If the link becomes available 791 in that period, the originator of the BGP-LS LINK NLRI will simply 792 advertise a more recent version of the BGP-LS Link NLRI without the 793 SPF Status TLV in the BGP-LS Link Attributes. 795 Similarily, when a prefix becomes unreachable, a more recent version 796 of the BGP-LS Prefix NLRI will be advertised with the SPF Status TLV 797 Section 4.3.1 indicating the prefix is unreachable in the BGP-LS 798 Prefix Attributes and the prefix will be considered unreachable with 799 respect to BGP SPF. After some configurable period of time, e.g., 800 2-3 seconds, the BGP-LS Prefix NLRI can be withdrawn with no 801 consequence. If the prefix becomes reachable in that period, the 802 originator of the BGP-LS Prefix NLRI will simply advertise a more 803 recent version of the BGP-LS Prefix NLRI without the SPF Status TLV 804 in the BGP-LS Prefix Attributes. 806 5.6.2. Node Failure Convergence 808 With BGP without graceful restart [RFC4724], all the NLRI advertised 809 by node are implicitly withdrawn when a session failure is detected. 810 If fast failure detection such as BFD is utilized and the node is on 811 the fastest converging path, the most recent versions of BGP-LS NLRI 812 may be withdrawn while these versions are in-flight on longer paths. 813 This will result the older version of the NLRI being used until the 814 new versions arrive and, potentially, unnecessary route flaps. 815 Therefore, BGP-LS SPF NLRI SHOULD always be retained before being 816 implicitly withdrawn for a brief configurable interval, e.g., 2-3 817 seconds. This will not delay convergence since the adjacent nodes 818 will detect the link failure and advertise a more recent NLRI 819 indicating the link is down with respect to BGP SPF Section 5.6.1 and 820 the BGP-SPF calculation will failure the bi-directional connectivity 821 check. 823 5.7. Error Handling 825 When a BGP speaker receives a BGP Update containing a malformed SPF 826 Capability TLV in the Node NLRI BGP-LS Attribute [RFC7752], it MUST 827 ignore the received TLV and the Node NLRI and not pass it to other 828 BGP peers as specified in [RFC7606]. When discarding a Node NLRI 829 with malformed TLV, a BGP speaker SHOULD log an error for further 830 analysis. 832 6. IANA Considerations 834 This document defines an AFI/SAFI for BGP-LS SPF operation and 835 requests IANA to assign the BGP-LS/BGP-LS-SPF (AFI 16388 / SAFI TBD1) 836 as described in [RFC4750]. 838 This document also defines five attribute TLVs for BGP-LS NLRI. We 839 request IANA to assign types for the SPF capability TLV, Sequence 840 Number TLV, IPv4 Link Prefix-Length TLV, IPv6 Link Prefix-Length TLV, 841 and SPF Status TLV from the "BGP-LS Node Descriptor, Link Descriptor, 842 Prefix Descriptor, and Attribute TLVs" Registry. 844 7. Security Considerations 846 This extension to BGP does not change the underlying security issues 847 inherent in the existing [RFC4271], [RFC4724], and [RFC7752]. 849 8. Management Considerations 851 This section includes unique management considerations for the BGP-LS 852 SPF address family. 854 8.1. Configuration 856 In addition to configuration of the BGP-LS SPF address family, 857 implementations SHOULD support the configuratio of the 858 INITIAL_SPF_DELAY, SHORT_SPF_DELAY, LONG_SPF_DELAY, TIME_TO_LEARN, 859 and HOLDDOWN_INTERVAL as documented in [RFC8405]. 861 8.2. Operational Data 863 In order to troubleshoot SPF issues, implementations SHOULD support 864 an SPF log including entries for previous SPF computations, Each SPF 865 log entry would include the BGP-LS NLRI SPF triggering the SPF, SPF 866 scheduled time, SPF start time, SPF end time, and SPF type if 867 different types of SPF are supported. Since the size of the log will 868 be finite, implementations SHOULD also maintain counters for the 869 total number of SPF computations of each type and the total number of 870 SPF triggering events. Additionally, to troubleshoot SPF scheduling 871 and backoff [RFC8405], the current SPF backoff state, remaining time- 872 to-learn, remaining holddown, last trigger event time, last SPF time, 873 and next SPF time should be available. 875 9. Acknowledgements 877 The authors would like to thank Sue Hares, Jorge Rabadan, Boris 878 Hassanov, Dan Frost, and Fred Baker for their review and comments. 879 Thanks to Chaitanya Yadlapalli and Pushpais Sarkar for discussions on 880 preventing a BGP SPF Router from being used for non-local traffic 881 (i.e., transit traffic). 883 The authors extend special thanks to Eric Rosen for fruitful 884 discussions on BGP-LS SPF convergence as compared to IGPs. 886 10. Contributors 888 In addition to the authors listed on the front page, the following 889 co-authors have contributed to the document. 891 Derek Yeung 892 Arrcus, Inc. 893 derek@arrcus.com 895 Gunter Van De Velde 896 Nokia 897 gunter.van_de_velde@nokia.com 899 Abhay Roy 900 Cisco Systems 901 akr@cisco.com 903 Venu Venugopal 904 Cisco Systems 905 venuv@cisco.com 907 11. References 909 11.1. Normative References 911 [I-D.ietf-idr-bgpls-segment-routing-epe] 912 Previdi, S., Talaulikar, K., Filsfils, C., Patel, K., Ray, 913 S., and J. Dong, "BGP-LS extensions for Segment Routing 914 BGP Egress Peer Engineering", draft-ietf-idr-bgpls- 915 segment-routing-epe-19 (work in progress), May 2019. 917 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 918 Requirement Levels", BCP 14, RFC 2119, 919 DOI 10.17487/RFC2119, March 1997, 920 . 922 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 923 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 924 DOI 10.17487/RFC4271, January 2006, 925 . 927 [RFC7606] Chen, E., Ed., Scudder, J., Ed., Mohapatra, P., and K. 928 Patel, "Revised Error Handling for BGP UPDATE Messages", 929 RFC 7606, DOI 10.17487/RFC7606, August 2015, 930 . 932 [RFC7752] Gredler, H., Ed., Medved, J., Previdi, S., Farrel, A., and 933 S. Ray, "North-Bound Distribution of Link-State and 934 Traffic Engineering (TE) Information Using BGP", RFC 7752, 935 DOI 10.17487/RFC7752, March 2016, 936 . 938 [RFC7938] Lapukhov, P., Premji, A., and J. Mitchell, Ed., "Use of 939 BGP for Routing in Large-Scale Data Centers", RFC 7938, 940 DOI 10.17487/RFC7938, August 2016, 941 . 943 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 944 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 945 May 2017, . 947 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 948 Decraene, B., Litkowski, S., and R. Shakir, "Segment 949 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 950 July 2018, . 952 [RFC8405] Decraene, B., Litkowski, S., Gredler, H., Lindem, A., 953 Francois, P., and C. Bowers, "Shortest Path First (SPF) 954 Back-Off Delay Algorithm for Link-State IGPs", RFC 8405, 955 DOI 10.17487/RFC8405, June 2018, 956 . 958 11.2. Information References 960 [I-D.ietf-lsvr-applicability] 961 Patel, K., Lindem, A., Zandi, S., and G. Dawra, "Usage and 962 Applicability of Link State Vector Routing in Data 963 Centers", draft-ietf-lsvr-applicability-02 (work in 964 progress), May 2019. 966 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, 967 DOI 10.17487/RFC2328, April 1998, 968 . 970 [RFC4456] Bates, T., Chen, E., and R. Chandra, "BGP Route 971 Reflection: An Alternative to Full Mesh Internal BGP 972 (IBGP)", RFC 4456, DOI 10.17487/RFC4456, April 2006, 973 . 975 [RFC4724] Sangli, S., Chen, E., Fernando, R., Scudder, J., and Y. 976 Rekhter, "Graceful Restart Mechanism for BGP", RFC 4724, 977 DOI 10.17487/RFC4724, January 2007, 978 . 980 [RFC4750] Joyal, D., Ed., Galecki, P., Ed., Giacalone, S., Ed., 981 Coltun, R., and F. Baker, "OSPF Version 2 Management 982 Information Base", RFC 4750, DOI 10.17487/RFC4750, 983 December 2006, . 985 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 986 "Multiprotocol Extensions for BGP-4", RFC 4760, 987 DOI 10.17487/RFC4760, January 2007, 988 . 990 [RFC4790] Newman, C., Duerst, M., and A. Gulbrandsen, "Internet 991 Application Protocol Collation Registry", RFC 4790, 992 DOI 10.17487/RFC4790, March 2007, 993 . 995 [RFC4915] Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L., and P. 996 Pillay-Esnault, "Multi-Topology (MT) Routing in OSPF", 997 RFC 4915, DOI 10.17487/RFC4915, June 2007, 998 . 1000 [RFC5286] Atlas, A., Ed. and A. Zinin, Ed., "Basic Specification for 1001 IP Fast Reroute: Loop-Free Alternates", RFC 5286, 1002 DOI 10.17487/RFC5286, September 2008, 1003 . 1005 [RFC5549] Le Faucheur, F. and E. Rosen, "Advertising IPv4 Network 1006 Layer Reachability Information with an IPv6 Next Hop", 1007 RFC 5549, DOI 10.17487/RFC5549, May 2009, 1008 . 1010 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 1011 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 1012 . 1014 Authors' Addresses 1016 Keyur Patel 1017 Arrcus, Inc. 1019 Email: keyur@arrcus.com 1021 Acee Lindem 1022 Cisco Systems 1023 301 Midenhall Way 1024 Cary, NC 27513 1025 USA 1027 Email: acee@cisco.com 1029 Shawn Zandi 1030 Linkedin 1031 222 2nd Street 1032 San Francisco, CA 94105 1033 USA 1035 Email: szandi@linkedin.com 1037 Wim Henderickx 1038 Nokia 1039 Antwerp 1040 Belgium 1042 Email: wim.henderickx@nokia.com