idnits 2.17.1 draft-ietf-lsvr-applicability-08.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 (March 25, 2021) is 1121 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Outdated reference: A later version (-29) exists of draft-ietf-lsvr-bgp-spf-12 == Outdated reference: A later version (-17) exists of draft-acee-idr-lldp-peer-discovery-08 == Outdated reference: A later version (-18) exists of draft-ietf-idr-bgp-ls-segment-routing-ext-16 == Outdated reference: A later version (-18) exists of draft-ietf-lsr-dynamic-flooding-08 == Outdated reference: A later version (-12) exists of draft-ietf-lsvr-l3dl-07 -- Obsolete informational reference (is this intentional?): RFC 5549 (Obsoleted by RFC 8950) -- Obsolete informational reference (is this intentional?): RFC 7752 (Obsoleted by RFC 9552) Summary: 0 errors (**), 0 flaws (~~), 6 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 LSVR K. Patel 3 Internet-Draft Arrcus, Inc. 4 Intended status: Informational A. Lindem 5 Expires: September 26, 2021 Cisco Systems 6 S. Zandi 7 G. Dawra 8 Linkedin 9 March 25, 2021 11 Usage and Applicability of Link State Vector Routing in Data Centers 12 draft-ietf-lsvr-applicability-08 14 Abstract 16 This document discusses the usage and applicability of Link State 17 Vector Routing (LSVR) extensions in data center networks utilizing 18 CLOS or Fat-Tree topologies. The document is intended to provide a 19 simplified guide for the deployment of LSVR extensions. 21 Status of This Memo 23 This Internet-Draft is submitted in full conformance with the 24 provisions of BCP 78 and BCP 79. 26 Internet-Drafts are working documents of the Internet Engineering 27 Task Force (IETF). Note that other groups may also distribute 28 working documents as Internet-Drafts. The list of current Internet- 29 Drafts is at https://datatracker.ietf.org/drafts/current/. 31 Internet-Drafts are draft documents valid for a maximum of six months 32 and may be updated, replaced, or obsoleted by other documents at any 33 time. It is inappropriate to use Internet-Drafts as reference 34 material or to cite them other than as "work in progress." 36 This Internet-Draft will expire on September 26, 2021. 38 Copyright Notice 40 Copyright (c) 2021 IETF Trust and the persons identified as the 41 document authors. All rights reserved. 43 This document is subject to BCP 78 and the IETF Trust's Legal 44 Provisions Relating to IETF Documents 45 (https://trustee.ietf.org/license-info) in effect on the date of 46 publication of this document. Please review these documents 47 carefully, as they describe your rights and restrictions with respect 48 to this document. Code Components extracted from this document must 49 include Simplified BSD License text as described in Section 4.e of 50 the Trust Legal Provisions and are provided without warranty as 51 described in the Simplified BSD License. 53 Table of Contents 55 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 56 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 3 57 3. Recommended Reading . . . . . . . . . . . . . . . . . . . . . 3 58 4. Common Deployment Scenario . . . . . . . . . . . . . . . . . 3 59 5. Justification for BGP SPF Extension . . . . . . . . . . . . . 4 60 6. LSVR Applicability to CLOS Networks . . . . . . . . . . . . . 5 61 6.1. Usage of BGP-LS SPF SAFI . . . . . . . . . . . . . . . . 5 62 6.1.1. Relationship to Other BGP AFI/SAFI Tuples . . . . . . 6 63 6.2. Peering Models . . . . . . . . . . . . . . . . . . . . . 6 64 6.2.1. Sparse Peering Model . . . . . . . . . . . . . . . . 6 65 6.2.2. Bi-Connected Graph Heuristic . . . . . . . . . . . . 7 66 6.3. BGP Spine/Leaf Topology Policy . . . . . . . . . . . . . 7 67 6.4. BGP Peer Discovery Requirements . . . . . . . . . . . . . 8 68 6.5. BGP Peer Discovery . . . . . . . . . . . . . . . . . . . 9 69 6.5.1. BGP Peer Discovery Alternatives . . . . . . . . . . . 9 70 6.5.2. BGP IPv6 Simplified Peering . . . . . . . . . . . . . 9 71 6.5.3. BGP-LS SPF Topology Visibility for Management . . . . 10 72 6.5.4. Data Center Interconnect (DCI) Applicability . . . . 10 73 7. Non-CLOS/FAT Tree Topology Applicability . . . . . . . . . . 10 74 8. Non-Transit Node Capability . . . . . . . . . . . . . . . . . 10 75 9. BGP Policy Applicability . . . . . . . . . . . . . . . . . . 11 76 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11 77 11. Security Considerations . . . . . . . . . . . . . . . . . . . 11 78 12. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 11 79 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 80 13.1. Normative References . . . . . . . . . . . . . . . . . . 11 81 13.2. Informative References . . . . . . . . . . . . . . . . . 12 82 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 84 1. Introduction 86 This document complements [I-D.ietf-lsvr-bgp-spf] by discussing the 87 applicability of the technology in a simple and fairly common 88 deployment scenario, which is described in Section 4. 90 After describing the deployment scenario, Section 5 will describe the 91 reasons for BGP modifications for such deployments. 93 Once the control plane routing protocol requirements are described, 94 Section 6 will cover the LSVR protocol enhancements to BGP to meet 95 these requirements and their applicability to Data Center CLOS 96 networks. 98 2. Requirements Language 100 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 101 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 102 "OPTIONAL" in this document are to be interpreted as described in BCP 103 14 [RFC2119] [RFC8174] when, and only when, they appear in all 104 capitals, as shown here. 106 3. Recommended Reading 108 This document assumes knowledge of existing data center networks and 109 data center network topologies [CLOS]. This document also assumes 110 knowledge of data center routing protocols like BGP [RFC4271], BGP- 111 SPF [I-D.ietf-lsvr-bgp-spf], OSPF [RFC2328], as well as, data center 112 OAM protocols like LLDP [RFC4957] and BFD [RFC5580]. 114 4. Common Deployment Scenario 116 Within a Data Center, servers are commonly interconnected the CLOS 117 topology [CLOS]. The CLOS topology is fully non-blocking and the 118 topology is realized using Equal Cost Multi-Path (ECMP). In a CLOS 119 topology, the minimum number of parallel paths between two servers is 120 determined by the width of a tier-1 stage as shown in the figure 1. 122 The following example illustrates multi-stage CLOS topology. 124 Tier-1 125 +-----+ 126 |NODE | 127 +->| 12 |--+ 128 | +-----+ | 129 Tier-2 | | Tier-2 130 +-----+ | +-----+ | +-----+ 131 +------------>|NODE |--+->|NODE |--+--|NODE |-------------+ 132 | +-----| 9 |--+ | 10 | +--| 11 |-----+ | 133 | | +-----+ +-----+ +-----+ | | 134 | | | | 135 | | +-----+ +-----+ +-----+ | | 136 | +-----+---->|NODE |--+ |NODE | +--|NODE |-----+-----+ | 137 | | | +---| 6 |--+->| 7 |--+--| 8 |---+ | | | 138 | | | | +-----+ | +-----+ | +-----+ | | | | 139 | | | | | | | | | | 140 +-----+ +-----+ | +-----+ | +-----+ +-----+ 141 |NODE | |NODE | Tier-3 +->|NODE |--+ Tier-3 |NODE | |NODE | 142 | 1 | | 2 | | 3 | | 4 | | 5 | 143 +-----+ +-----+ +-----+ +-----+ +-----+ 144 | | | | | | | | 145 A O B O <- Servers -> Z O O O 147 Figure 1: Illustration of the basic CLOS 149 5. Justification for BGP SPF Extension 151 In order to simplify layer-3 routing and operations [RFC7938], many 152 data centers use BGP as a routing protocol to create both an underlay 153 and overlay network for their CLOS Topologies. However, BGP is a 154 path-vector routing protocol. Since it does not create a fabric 155 topology, it uses hop-by-hop EBGP peering to facilitate hop-by-hop 156 routing to create the underlay network and to resolve any overlay 157 next hops. The hop-by-hop BGP peering paradigm imposes several 158 restrictions within a CLOS. It severely prohibits a deployment of 159 Route Reflectors/Route Controllers as the EBGP sessions are congruent 160 with the data path. The BGP best-path algorithm is prefix-based and 161 it prevents announcements of prefixes to other BGP speakers until the 162 best-path decision process has been performed for the prefix at each 163 intermediate hop. These restrictions significantly delay the overall 164 convergence of the underlay network within a CLOS network. 166 The LSVR SPF modifications allow BGP to overcome these limitations. 167 Furthermore, using the BGP-LS NLRI format [RFC7752] allows the LSVR 168 data to be advertised for nodes, links, and prefixes in the BGP 169 routing domain and used for SPF computations. 171 6. LSVR Applicability to CLOS Networks 173 With the BGP SPF extensions [I-D.ietf-lsvr-bgp-spf], the BGP best- 174 path computation and route computation are replaced with OSPF-like 175 algorithms [RFC2328] both to determine whether an BGP-LS SPF NLRI has 176 changed and needs to be re-advertised and to compute the BGP routes. 177 These modifications will significantly improve convergence of the 178 underlay while affording the operational benefits of a single routing 179 protocol [RFC7938]. 181 Data center controllers typically require visibility to the BGP 182 topology to compute traffic-engineered paths. These controllers 183 learn the topology and other relevant information via the BGP-LS 184 address family [RFC7752] which is totally independent of the underlay 185 address families (usually IPv4/IPv6 unicast). Furthermore, in 186 traditional BGP underlays, all the BGP routers will need to advertise 187 their BGP-LS information independently. With the BGP SPF extensions, 188 controllers can learn the topology using the same BGP advertisements 189 used to compute the underlay routes. Furthermore, these data center 190 controllers can avail the convergence advantages of the BGP SPF 191 extensions. The placement of controllers can be outside of the 192 forwarding path or within the forwarding path. 194 Alternatively, as each and every router in the BGP SPF domain will 195 have a complete view of the topology, the operator can also choose to 196 configure BGP sessions in hop-by-hop peering model described in 197 [RFC7938] along with BFD [RFC5580]. In doing so, while the hop-by- 198 hop peering model lacks the inherent benefits of the controller-based 199 model, BGP updates need not be serialized by BGP best-path algorithm 200 in either of these models. This helps overall network convergence. 202 6.1. Usage of BGP-LS SPF SAFI 204 The BGP SPF extensions [I-D.ietf-lsvr-bgp-spf] define a new BGP-LS 205 SPF SAFI for announcement of BGP SPF link-state. The NLRI format and 206 its associated attributes follow the format of BGP-LS for node, link, 207 and prefix announcements. Whether the peering model within a CLOS 208 follows hop-by-hop peering described in [RFC7938] or any controller- 209 based or route-reflector peering, an operator can exchange BGP SPF 210 SAFI routes over the BGP peering by simply configuring BGP SPF SAFI 211 between the necessary BGP speakers. 213 The BGP-LS SPF SAFI can also co-exist with BGP IP Unicast SAFI which 214 could exchange overlapping IP routes. The routes received by these 215 SAFIs are evaluated, stored, and announced independently according to 216 the rules of [RFC4760]. The tie-breaking of route installation is a 217 matter of the local policies and preferences of the network operator. 219 Finally, as the BGP SPF peering is done following the procedures 220 described in [RFC4271], all the existing transport security 221 mechanisms including [RFC5925] are available for the BGP-LS SPF SAFI. 223 6.1.1. Relationship to Other BGP AFI/SAFI Tuples 225 Normally, the BGP-LS AFI/SAFI is used solely to compute the underlay 226 and is given preference over other AFI/SAFIs. Other BGP SAFIs, e.g., 227 IPv6/IPv6 Unicast VPN would use the BGP-SPF computed routes for next 228 hop resolution. However, if BGP-LS NLRI is also being advertised for 229 controller consumption, there is no need to replicate the Node, Link, 230 and Prefix NLRI in BGP-NLRI. Rather, additional NLRI attributes can 231 be advertised in the BGP-LS SPF AFI/SAFI as required (e.g., BGP-LS TE 232 metric extensions [RFC8571] and BGP-LS segment routing extensions 233 [I-D.ietf-idr-bgp-ls-segment-routing-ext]). 235 6.2. Peering Models 237 As previously stated, BGP SPF can be deployed using the existing 238 peering model where there is a single-hop BGP session on each and 239 every link in the data center fabric [RFC7938]. This provides for 240 both the advertisement of routes and the determination of link and 241 neighboring switch availability. With BGP SPF, the underlay will 242 converge faster due to changes to the decision process that will 243 allow NLRI changes to be advertised faster after detecting a change. 245 6.2.1. Sparse Peering Model 247 Alternately, BFD [RFC5580] can be used to swiftly determine the 248 availability of links and the BGP peering model can be significantly 249 sparser than the data center fabric. BGP SPF sessions only need to 250 be established with enough peers to provide a bi-connected graph. If 251 IEBGP is used, then the BGP routers at tier N-1 will act as route- 252 reflectors for the routers at tier N. 254 The obvious usage of sparse peering is to avoid parallel BGP sessions 255 on links between the same two switches in the data center fabric. 256 However, this use case is not very useful since parallel layer-3 257 links between the same two BGP routers are rare in CLOS or Fat-Tree 258 topologies. Additionally, when there are multiple links, they are 259 often aggregated at the link layer rather than the IP layer. Two 260 more interesting scenarios are described below. 262 In current data center topologies, there is often a very dense mesh 263 of links between levels, e.g., leaf and spine, providing 32-way, 264 64-way, or more Equal-Cost Multi-Path (ECMP) paths. In these 265 topologies, it is desirable not to have a BGP session on every link 266 and techniques such as the one described in Section 6.2.2 can be used 267 establish sessions on some subset of northbound links. For example, 268 in a Spine-Leaf topology, each leaf switch would only peer with a 269 subset of the spines dependent on the flooding redundancy required to 270 be reasonably certain that every node within the BGP-LS SPF routing 271 domain has the complete topology (refer to Section 6.2.1). 273 Alternately, controller-based data center topologies are envisioned 274 where BGP speakers within the data center only establish BGP sessions 275 with two or more controllers. In these topologies, fabric nodes 276 below the first tier (using [RFC7938] hierarchy) will establish BGP 277 multi-hop sessions with the controllers. For the multi-hop sessions, 278 determining the route to the controllers without depending on BGP 279 would need to be through some other means beyond the scope of this 280 document. However, the BGP discovery mechanisms described in 281 Section 6.5 would be one possibility. 283 6.2.2. Bi-Connected Graph Heuristic 285 With this heuristic, discovery of BGP peers is assumed, e.g., as 286 described in Section 6.5. Additionally, it assumed that the 287 direction of the peering can be ascertained. In the context of a 288 data center fabric, direction is either northbound (toward the 289 spine), southbound (toward the Top-Of-Rack (TOR) switches) or east- 290 west (same level in hierarchy. The determination of the direction is 291 beyond the scope of this document. However, it would be reasonable 292 to assume a technique where the TOR switches can be identified and 293 the number of hops to the TOR is used to determine the direction. 295 In this heuristic, BGP speakers allow passive session establishment 296 for southbound BGP sessions. For northbound sessions, BGP speakers 297 will attempt to maintain two northbound BGP sessions with different 298 switches (in data center fabrics there is normally a single layer-3 299 connection anyway). For east-west sessions, passive BGP session 300 establishment is allowed. However, BGP speaker will never actively 301 establish an east-west BGP session unless it cannot establish two 302 northbound BGP sessions. 304 6.3. BGP Spine/Leaf Topology Policy 306 One of the advantages of using BGP SPF as the underlay protocol is 307 that BGP policy can be applied at any level. For example, depending 308 upon the topology, it may be possible to aggregate prefix 309 advertisements using existing BGP policy. In Spine/Leaf topologies, 310 it is not necessary to advertise BGP-LS NLRI received by leaves 311 northbound to the spine nodes at the level above. If a common AS is 312 used for the spine nodes, this can easily be accomplished with EBGP 313 and a simple policy to filter advertisements from the leaves to the 314 spine if the first AS in the AS path is the spine AS. 316 In the figure below, the leaves would not advertise any NLRI with AS 317 64512 as the first AS in the AS path. 319 +--------+ +--------+ +--------+ 320 AS 64512 | | | | | | 321 for Spine | Spine 1+----+ Spine 2+- ......... -+ Spine N| 322 Nodes at | | | | | | 323 this Level +-+-+-+-++ ++-+-+-+-+ +-+-+-+-++ 324 +------+ | | | | | | | | | | | 325 | +-----|-|-|------+ | | | | | | | 326 | | +--|-|-|--------+-|-|-----------------+ | | | 327 | | | | | | +---+ | | | | | 328 | | | | | | | +--|-|-------------------+ | | 329 | | | | | | | | | | +------+ +----+ 330 | | | | | | | | | +--------------|----------+ | 331 | | | | | | | | +-------------+ | | | 332 | | | | | +----|--|----------------|--|--------+ | | 333 | | | | +------|--|--------------+ | | | | | 334 | | | +------+ | | | | | | | | 335 ++--+--++ +-+-+--++ ++-+--+-+ ++-+--+-+ 336 | Leaf 1|~~~~~~| Leaf 2| ........ | Leaf X| | Leaf Y| 337 +-------+ +-------+ +-------+ +-------+ 339 Figure 2: Spine/Leaf Topology Policy 341 6.4. BGP Peer Discovery Requirements 343 The most basic requirement is to be able to discover the address of a 344 single-hop peer without pre-configuration. This is being 345 accomplished today with using IPv6 Router Advertisements (RA) 346 [RFC4861] and assuming that a BGP sessions is desired with any 347 discovered peer. Beyond the basic requirement, it is useful to have 348 to following information relating to the BGP session: 350 o Autonomous System (AS) and BGP Identifier of a potential peer. 351 The latter can be used for debugging and to decrease the 352 likelihood of BGP session establishment collisions. 354 o Security capabilities supported and for cryptographic 355 authentication, the security capabilities and possibly a key-chain 356 [RFC8177] to be used. 358 o Session Policy Identifier - A group number or name used to 359 associate common session parameters with the peer. For example, 360 in a data center, BGP sessions with a Top of Rack (ToR) device 361 could have parameters than BGP sessions between leaf and spine. 363 In a data center fabric, it is often useful to know whether a peer is 364 southbound (towards the servers) or northbound (towards the spine or 365 super-spine), e.g., Section 6.2.2. A potential requirement would be 366 to determine this dynamically. One mechanism, without specifying all 367 the details, might be for the ToR switches to be identified when 368 installed and for the others switches in the fabric to determine 369 their level based on the distance from the closest ToR switch. 371 If there are multiple links between BGP speakers or the links between 372 BGP speakers are unnumbered, it is also useful to be able to 373 establish multi-hop sessions using the loopback addresses. This will 374 often require the discovery protocol to install route(s) toward the 375 potential peer loopback addresses prior to BGP session establishment. 377 Finally, a simple BGP discovery protocol could also be used to 378 establish a multi-hop session with one or more controllers by 379 advertising connectivity to one or more controllers. However, once 380 the multi-hop session actually traverses multiple nodes, it is 381 bordering a distance-vector routing protocol and possibly this is not 382 a good requirement for the discovery protocol. 384 6.5. BGP Peer Discovery 386 6.5.1. BGP Peer Discovery Alternatives 388 While BGP peer discovery is not part of [I-D.ietf-lsvr-bgp-spf], 389 there are, at least, three proposals for BGP peer discovery. At 390 least one of these mechanisms will be adopted and will be applicable 391 to deployments other than the data center. It is strongly 392 RECOMMENDED that the accepted mechanism be used in conjunction with 393 BGP SPF in data centers. The BGP discovery mechanism should 394 discovery both peer addresses and endpoints for BFD discovery. 395 Additionally, it would be great if there were a heuristic for 396 determining whether the peer is at a tier above or below the 397 discovering BGP speaker (refer to Section 6.2.2). 399 The BGP discovery mechanisms under consideration are 400 [I-D.acee-idr-lldp-peer-discovery], 401 [I-D.xu-idr-neighbor-autodiscovery], and [I-D.ietf-lsvr-l3dl]. 403 6.5.2. BGP IPv6 Simplified Peering 405 In order to conserve IPv4 address space and simplify operations, BGP- 406 LS SPF routers in CLOS/Fat-Tree deployments can use IPv6 addresses as 407 peer address. For IPv4 address families, IPv6 peering as specified 408 in [RFC5549] can be deployed to avoid configuring IPv4 addresses on 409 BGP-LS SPF router interfaces. When this is done, dynamic discovery 410 mechanisms, as described in Section 6.5, can used to learn the global 411 or link-local IPv6 peer addresses and IPv4 addresses need not be 412 configured on these interfaces. If IPv6 link-local peering is used, 413 then configuration of IPv6 global addresses is also not required and 414 these IPv6 link-local addresses must then be advertised in the BGP-LS 415 Link Descriptor IPv6 Address TLV (262) [RFC7752]. 417 6.5.3. BGP-LS SPF Topology Visibility for Management 419 Irrespective of whether or not BGP-LS SPF is used for route 420 calculation, the BGP-LS SPF route advertisements can be used to 421 periodically construct the CLOS/FAT Tree topology. This is 422 especially useful in deployments where an IGP is not used and the 423 base BGP-LS routes [RFC7752] are not available. The resultant 424 topology visibility can then be used for troubleshooting and 425 consistency checking. This would normally be done on a central 426 controller or other management tool which could also be used for 427 fabric data path verification. The precise algorithms and 428 heuristics, as well as, the complete set of management applications 429 is beyond the scope of this document. 431 6.5.4. Data Center Interconnect (DCI) Applicability 433 Since BGP SPF is to be used for the routing underlay and DCI gateway 434 boxes typically have direct or very simple connectivity, BGP external 435 sessions would typically not include the BGP SPF SAFI. 437 7. Non-CLOS/FAT Tree Topology Applicability 439 The BGP SPF extensions [I-D.ietf-lsvr-bgp-spf] can be used in other 440 topologies and avail the inherent convergence improvements. 441 Additionally, sparse peering techniques may be utilized Section 6.2. 442 However, determining whether or to establish a BGP session is more 443 complex and the heuristic described in Section 6.2.2 cannot be used. 444 In such topologies, other techniques such as those described in 445 [I-D.ietf-lsr-dynamic-flooding] may be employed. One potential 446 deployment would be the underlay for a Service Provider (SP) backbone 447 where usage of a single protocol, i.e., BGP, is desired. 449 8. Non-Transit Node Capability 451 In certain scenarios, a BGP node wishes to participate in the BGP SPF 452 topology but never be used for transit traffic. These in include 453 situations where a server wants to make application services 454 available to clients homed at subnets throughout the BGP SPF domain 455 but does not ever want to be used as a router (i.e., carry transit 456 traffic). Another specific instance is where a controller is 457 resident on a server and direct connectivity to the controller is 458 required throughout the entire domain. This can readily be 459 accomplished using the BGP-LS Node NLRI Attribute SPF Status TLV as 460 described in [I-D.ietf-lsvr-bgp-spf]. 462 9. BGP Policy Applicability 464 Existing BGP policy including aggregation and prefix filtering may be 465 used in conjunction with the BGP-LS SPF SAFI. When aggregation 466 policy is used, BGP-LS SPF prefix NLRI will be originated for the 467 aggregate prefix and BGP-LS SPF prefix NLRI for components will be 468 filtered. Additionally, link and node NLRI may be filtered and the 469 abstracted by the prefix NLRI. 471 When BGP policy is used with the BGP-LS SPF SAFI, BGP speakers in the 472 BGP-LS SPF routing domain will not all have the same set of NLRI and 473 will compute a different BGP local routing table. Consequently, care 474 must be taken to assure routing is consistent and blackholes or 475 routing loops do not ensue. However, this is no different than if 476 tradition BGP routing using the IPv4 and IPv6 address families were 477 used. 479 10. IANA Considerations 481 No IANA updates are requested by this document. 483 11. Security Considerations 485 This document introduces no new security considerations above and 486 beyond those already specified in the [RFC4271] and 487 [I-D.ietf-lsvr-bgp-spf]. 489 12. Acknowledgements 491 The authors would like to thank Alvaro Retana, Yan Filyurin, and 492 Boris Hassanov for their review and comments. 494 13. References 496 13.1. Normative References 498 [I-D.ietf-lsvr-bgp-spf] 499 Patel, K., Lindem, A., Zandi, S., and W. Henderickx, "BGP 500 Link-State Shortest Path First (SPF) Routing", draft-ietf- 501 lsvr-bgp-spf-12 (work in progress), January 2021. 503 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 504 Requirement Levels", BCP 14, RFC 2119, 505 DOI 10.17487/RFC2119, March 1997, 506 . 508 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 509 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 510 May 2017, . 512 13.2. Informative References 514 [CLOS] "A Study of Non-Blocking Switching Networks", The Bell 515 System Technical Journal, Vol. 32(2), DOI 516 10.1002/j.1538-7305.1953.tb01433.x, March 1953. 518 [I-D.acee-idr-lldp-peer-discovery] 519 Lindem, A., Patel, K., Zandi, S., Haas, J., and X. Xu, 520 "BGP Logical Link Discovery Protocol (LLDP) Peer 521 Discovery", draft-acee-idr-lldp-peer-discovery-08 (work in 522 progress), December 2020. 524 [I-D.ietf-idr-bgp-ls-segment-routing-ext] 525 Previdi, S., Talaulikar, K., Filsfils, C., Gredler, H., 526 and M. Chen, "BGP Link-State extensions for Segment 527 Routing", draft-ietf-idr-bgp-ls-segment-routing-ext-16 528 (work in progress), June 2019. 530 [I-D.ietf-lsr-dynamic-flooding] 531 Li, T., Psenak, P., Ginsberg, L., Chen, H., Przygienda, 532 T., Cooper, D., Jalil, L., Dontula, S., and G. Mishra, 533 "Dynamic Flooding on Dense Graphs", draft-ietf-lsr- 534 dynamic-flooding-08 (work in progress), December 2020. 536 [I-D.ietf-lsvr-l3dl] 537 Bush, R., Austein, R., and K. Patel, "Layer 3 Discovery 538 and Liveness", draft-ietf-lsvr-l3dl-07 (work in progress), 539 January 2021. 541 [I-D.xu-idr-neighbor-autodiscovery] 542 Xu, X., Talaulikar, K., Bi, K., Tantsura, J., and N. 543 Triantafillis, "BGP Neighbor Discovery", draft-xu-idr- 544 neighbor-autodiscovery-12 (work in progress), November 545 2019. 547 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, 548 DOI 10.17487/RFC2328, April 1998, 549 . 551 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 552 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 553 DOI 10.17487/RFC4271, January 2006, 554 . 556 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 557 "Multiprotocol Extensions for BGP-4", RFC 4760, 558 DOI 10.17487/RFC4760, January 2007, 559 . 561 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 562 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 563 DOI 10.17487/RFC4861, September 2007, 564 . 566 [RFC4957] Krishnan, S., Ed., Montavont, N., Njedjou, E., Veerepalli, 567 S., and A. Yegin, Ed., "Link-Layer Event Notifications for 568 Detecting Network Attachments", RFC 4957, 569 DOI 10.17487/RFC4957, August 2007, 570 . 572 [RFC5549] Le Faucheur, F. and E. Rosen, "Advertising IPv4 Network 573 Layer Reachability Information with an IPv6 Next Hop", 574 RFC 5549, DOI 10.17487/RFC5549, May 2009, 575 . 577 [RFC5580] Tschofenig, H., Ed., Adrangi, F., Jones, M., Lior, A., and 578 B. Aboba, "Carrying Location Objects in RADIUS and 579 Diameter", RFC 5580, DOI 10.17487/RFC5580, August 2009, 580 . 582 [RFC5925] Touch, J., Mankin, A., and R. Bonica, "The TCP 583 Authentication Option", RFC 5925, DOI 10.17487/RFC5925, 584 June 2010, . 586 [RFC7752] Gredler, H., Ed., Medved, J., Previdi, S., Farrel, A., and 587 S. Ray, "North-Bound Distribution of Link-State and 588 Traffic Engineering (TE) Information Using BGP", RFC 7752, 589 DOI 10.17487/RFC7752, March 2016, 590 . 592 [RFC7938] Lapukhov, P., Premji, A., and J. Mitchell, Ed., "Use of 593 BGP for Routing in Large-Scale Data Centers", RFC 7938, 594 DOI 10.17487/RFC7938, August 2016, 595 . 597 [RFC8177] Lindem, A., Ed., Qu, Y., Yeung, D., Chen, I., and J. 598 Zhang, "YANG Data Model for Key Chains", RFC 8177, 599 DOI 10.17487/RFC8177, June 2017, 600 . 602 [RFC8571] Ginsberg, L., Ed., Previdi, S., Wu, Q., Tantsura, J., and 603 C. Filsfils, "BGP - Link State (BGP-LS) Advertisement of 604 IGP Traffic Engineering Performance Metric Extensions", 605 RFC 8571, DOI 10.17487/RFC8571, March 2019, 606 . 608 Authors' Addresses 610 Keyur Patel 611 Arrcus, Inc. 612 2077 Gateway Pl 613 San Jose, CA 95110 614 USA 616 Email: keyur@arrcus.com 618 Acee Lindem 619 Cisco Systems 620 301 Midenhall Way 621 Cary, NC 95110 622 USA 624 Email: acee@cisco.com 626 Shawn Zandi 627 Linkedin 628 222 2nd Street 629 San Francisco, CA 94105 630 USA 632 Email: szandi@linkedin.com 634 Gaurav Dawra 635 Linkedin 636 222 2nd Street 637 San Francisco, CA 94105 638 USA 640 Email: gdawra@linkedin.com