idnits 2.17.1 draft-keyupate-i2rs-bgp-usecases-01.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 : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'MUST not' in this paragraph: o Avoid Unwanted Route Announcements. Filter prefixes that MUST not be routed [RFC5735], [RFC5156]. Filter prefixes that are not allocated by Internet Routing Registries. == 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 (February 12, 2014) is 3724 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC2629' is defined on line 526, but no explicit reference was found in the text == Unused Reference: 'RFC3392' is defined on line 529, but no explicit reference was found in the text == Unused Reference: 'RFC3552' is defined on line 532, but no explicit reference was found in the text == Unused Reference: 'RFC4271' is defined on line 536, but no explicit reference was found in the text == Unused Reference: 'RFC2858' is defined on line 575, but no explicit reference was found in the text == Outdated reference: A later version (-15) exists of draft-ietf-i2rs-architecture-00 ** Obsolete normative reference: RFC 2629 (Obsoleted by RFC 7749) ** Obsolete normative reference: RFC 3392 (Obsoleted by RFC 5492) == Outdated reference: A later version (-07) exists of draft-ietf-grow-ops-reqs-for-bgp-error-handling-05 == Outdated reference: A later version (-04) exists of draft-white-grow-overlapping-routes-01 -- Obsolete informational reference (is this intentional?): RFC 2858 (Obsoleted by RFC 4760) -- Obsolete informational reference (is this intentional?): RFC 5156 (Obsoleted by RFC 6890) -- Obsolete informational reference (is this intentional?): RFC 5575 (Obsoleted by RFC 8955) -- Obsolete informational reference (is this intentional?): RFC 5735 (Obsoleted by RFC 6890) Summary: 3 errors (**), 0 flaws (~~), 11 warnings (==), 5 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Patel 3 Internet-Draft R. Fernando 4 Intended status: Informational Cisco Systems 5 Expires: August 16, 2014 H. Gredler 6 Juniper Networks 7 S. Amante 8 Level 3 Communications, Inc. 9 R. White 10 Ericsson 11 S. Hares 12 Hickory Hill Consulting 13 February 12, 2014 15 Use Cases for an Interface to BGP Protocol 16 draft-keyupate-i2rs-bgp-usecases-01.txt 18 Abstract 20 A network routing protocol like BGP is typically configured and 21 analyzed through some form of Command Line Interface (CLI) or 22 NETCONF. These interactions to control BGP and diagnose its 23 operation encompass: configuration of protocol parameters, display of 24 protocol data, setting of certain protocol state and debugging of the 25 protocol. 27 Interface to the Routing System's (I2RS) Programmatic interfaces, as 28 defined in [draft-ietf-i2rs-architecture], provides an alternate way 29 to control and diagnose the operation of the BGP protocol. I2RS may 30 be used for the configuration, manipulation, analyzing or collecting 31 the protocol data. This document describes set of use cases for 32 which I2RS can be used for BGP protocol. It is intended to provide a 33 base for the solution draft describing a set of interfaces to the BGP 34 protocol. 36 Status of This Memo 38 This Internet-Draft is submitted in full conformance with the 39 provisions of BCP 78 and BCP 79. 41 Internet-Drafts are working documents of the Internet Engineering 42 Task Force (IETF). Note that other groups may also distribute 43 working documents as Internet-Drafts. The list of current Internet- 44 Drafts is at http://datatracker.ietf.org/drafts/current/. 46 Internet-Drafts are draft documents valid for a maximum of six months 47 and may be updated, replaced, or obsoleted by other documents at any 48 time. It is inappropriate to use Internet-Drafts as reference 49 material or to cite them other than as "work in progress." 51 This Internet-Draft will expire on August 16, 2014. 53 Copyright Notice 55 Copyright (c) 2014 IETF Trust and the persons identified as the 56 document authors. All rights reserved. 58 This document is subject to BCP 78 and the IETF Trust's Legal 59 Provisions Relating to IETF Documents 60 (http://trustee.ietf.org/license-info) in effect on the date of 61 publication of this document. Please review these documents 62 carefully, as they describe your rights and restrictions with respect 63 to this document. Code Components extracted from this document must 64 include Simplified BSD License text as described in Section 4.e of 65 the Trust Legal Provisions and are provided without warranty as 66 described in the Simplified BSD License. 68 This document may contain material from IETF Documents or IETF 69 Contributions published or made publicly available before November 70 10, 2008. The person(s) controlling the copyright in some of this 71 material may not have granted the IETF Trust the right to allow 72 modifications of such material outside the IETF Standards Process. 73 Without obtaining an adequate license from the person(s) controlling 74 the copyright in such materials, this document may not be modified 75 outside the IETF Standards Process, and derivative works of it may 76 not be created outside the IETF Standards Process, except to format 77 it for publication as an RFC or to translate it into languages other 78 than English. 80 Table of Contents 82 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 83 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 84 2. BGP Protocol Operation . . . . . . . . . . . . . . . . . . . 4 85 2.1. BGP Error Handling for Internal BGP Sessions . . . . . . 4 86 3. BGP Route Manipulation . . . . . . . . . . . . . . . . . . . 4 87 3.1. Customized Best Path Selection Criteria . . . . . . . . . 5 88 3.2. Flowspec Routes . . . . . . . . . . . . . . . . . . . . . 5 89 3.3. Route Filter Routes for Legacy Routers . . . . . . . . . 5 90 3.4. Optimized Exit Control . . . . . . . . . . . . . . . . . 6 91 4. BGP Events . . . . . . . . . . . . . . . . . . . . . . . . . 6 92 4.1. Notification of Routing Events . . . . . . . . . . . . . 7 93 4.2. Tracing Dropped BGP Routes . . . . . . . . . . . . . . . 8 94 4.3. BGP Protocol Statistics . . . . . . . . . . . . . . . . . 8 95 5. Central membership computation for MPLS based VPNs . . . . . 9 96 6. Marking Overlapping Traffic Engineering Routes for Removal . 10 97 7. Security Considerations . . . . . . . . . . . . . . . . . . . 11 98 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 11 99 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 11 100 9.1. Normative References . . . . . . . . . . . . . . . . . . 11 101 9.2. Informative References . . . . . . . . . . . . . . . . . 12 102 Appendix A. BGP Configuration . . . . . . . . . . . . . . . . . 13 103 A.1. BGP Protocol Configuration . . . . . . . . . . . . . . . 14 104 A.2. BGP Policy Configuration . . . . . . . . . . . . . . . . 14 105 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 16 107 1. Introduction 109 Typically, a network routing protocol like BGP is configured and 110 results of its operation are analyzed through some form of Command 111 Line Interface (CLI) or NETCONF. These interactions to control BGP 112 and diagnose its operation encompass: configuration of protocol 113 parameters, display of protocol data, setting of certain protocol 114 state and debugging of the protocol. 116 The I2RS Framework document [I-D.ietf-i2rs-architecture] describes a 117 mechanism to control network protocols like BGP using a set of 118 programmatic interfaces. These programmatic interfaces allow one to 119 control the BGP protocol by analyzing its operational state and 120 routing protocol data, plus manipulating BGP's configuration to 121 achieve various goals. The I2RS is not intended to replace any 122 existing configuration mechanisms, (i.e.: Command Line Interface or 123 NETCONF). Instead, I2RS is intended to augment those existing 124 mechanisms by defining a standardized set of programmatic interfaces 125 to enable easier configuration, interrogation and analysis of the BGP 126 protocol. 128 This document describes set of use cases for which I2RS's 129 programmatic interfaces can be used to control and analyze the 130 operation of BGP. The use cases described in this document cover the 131 following aspects of BGP: protocol parameter configuration, protocol 132 route manipulation and tracking of protocol events. The goal is to 133 inform the community's understanding of where the I2RS BGP extensions 134 fit within the overall I2RS architecture. It is intended to provide 135 a basis for the solutions draft describing the set of Interfaces to 136 the BGP protocol. 138 1.1. Requirements Language 140 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 141 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 142 document are to be interpreted as described in RFC 2119 [RFC2119]. 144 2. BGP Protocol Operation 146 It is increasingly common for services facilitated via BGP to be 147 subject to severe, widespread disruptions (outages), primarily due to 148 the destructive teardown of BGP sessions as a result of receiving 149 malformed BGP attributes. The document Operational Requirements for 150 Enhanced Error Handling Behaviour in BGP-4 151 [I-D.ietf-grow-ops-reqs-for-bgp-error-handling] outlines requirements 152 to try to minimize the scope of the impact attributed to such errors. 153 Unfortunately, more fine-grained BGP error handling solutions, which 154 would result in little to no impact on the operation of BGP protocol, 155 remain elusive. 157 2.1. BGP Error Handling for Internal BGP Sessions 159 It is possible that I2RS could enable enhanced error handling 160 techniques for Internal BGP sessions. At a minimum, I2RS-capable BGP 161 routers could signal an event such as "Malformed Attribute Received" 162 toward an I2RS controller(s). I2RS controller(s) may already have a 163 real-time view of BGP routes, and corresponding BGP attributes, or 164 may dynamically interrogate BGP routers in the network to identify 165 the present propagation scope of the BGP route(s) that are affected. 166 Finally, the I2RS controller(s) could then signal back to BGP routers 167 to apply a filter that would block propagation of the BGP attribute 168 or BGP route, as necessary, in order to temporarily aid in 169 consistency of BGP routing information across the entire network 170 until a permanent fix can be developed and deployed within BGP 171 routers. 173 I2RS would enable the global visibility and global control over the 174 operational state of BGP, within a given Autonomous System, that is 175 necessary to facilitate the learning of, rapid response to and more 176 fine-grained isolation/scoping of BGP protocol events that currently 177 cause a destructive tear-down of BGP sessions that lead to widespread 178 disruptions of services. 180 3. BGP Route Manipulation 182 Multiprotocol BGP [RFC4760] provides support to carry routing 183 information for different BGP address families. Route manipulation 184 is heavily done across these different address families for different 185 reasons. BGP IPv4 and IPv6 address families use BGP Communities 186 [RFC1997] and other IBGP and EBGP attributes to manipulate BGP routes 187 for Traffic Engineering purpose. BGP VPN adddress families use 188 Extended Communities [RFC4360] to filter unwanted BGP routes. BGP 189 Flowspec address family [RFC5575] is used to install Flow based 190 filters to filter unwanted data traffic. The following sub-sections 191 describe the use of IRS towards BGP Route Manipulation for different 192 BGP address families. 194 3.1. Customized Best Path Selection Criteria 196 The BGP customized Bestpath facilitates custom bestpath computations 197 within a BGP speaking network. It is usually used within an IBGP 198 network. Customized bestpaths use special extended communities known 199 as cost communities. Cost communities carry enough information; 200 Point of Insertion (POI) and the cost value to signal where in BGP 201 bestpath the customize checks need to be done. Both, the traffic 202 engineering as well as backdoor (SHAM) links use customized bestpath 203 computation. 205 With I2RS, it would be possible for an I2RS controller to push routes 206 with custom cost communities on the BGP routers for Traffic 207 Engineering purpose. I2RS controller now can act as a central entity 208 keeping track of all Traffic engineering data that get applied to BGP 209 routes within an IBGP network. 211 3.2. Flowspec Routes 213 The BGP flowspec address family is used to disseminate the traffic 214 flow specification to the BGP Autonomous System Border Routers 215 (ASBRs) and Provider Edge (PE) routers. Both, the BGP ASBRs and the 216 PEs would translate the received BGP traffic flow specification into 217 an Access Control List (ACL) and install it in router's forwarding 218 path. Using such ACLs routers can now classify, shape, rate limit, 219 filter, or redirect traffic flows. 221 With I2RS, it would be possible for an I2RS controller to push 222 traffic flow specifications to the BGP ASBRs and the PE routers. 223 I2RS controller can act as a central entity tracking all the traffic 224 flow specifications that are installed within an IBGP network. I2RS 225 controller could also prioritize and control the announcement of 226 traffic flow specifications according to various ASRBs and PE 227 router's capacity. BGP ASBRs and PE routers MAY forward traffic flow 228 specifications received from EBGP speakers to I2RS Agents. This 229 would allow I2RS agents to centrally manage and track any externally 230 received traffic flow specifications. 232 3.3. Route Filter Routes for Legacy Routers 234 The BGP Route Filter address family is used to disseminate the Route 235 Target filter information between VPN BGP speakers. This information 236 is then used to build a route distribution graph that helps in 237 limiting the propagation of VPN NLRI within a VPN network. However, 238 it requires that all the BGP VPN routers are upgraded to support this 239 functionality. Otherwise, the graph information is incomplete when a 240 VPN network consists of legacy routers that participates in VPN but 241 does not implement the BGP route filter address family. 243 With I2RS, it would be possible for an I2RS controller to push router 244 filter information to BGP RR routers on behalf of all legacy routers 245 that participates in VPN but does not support or implement the BGP 246 route filter address family. I2RS controller can act as a central 247 entity tracking all the configured Route Filters for legacy routers 248 and push them on appropriate RRs who in turn would push it to ASBRs 249 and PE routers. In this way, I2RS agents help build an optimal route 250 distribution graph that would assist in filtering of VPN NLRIs in a 251 VPN network. 253 3.4. Optimized Exit Control 255 Optimized Exit Control is used to provide route optimization and load 256 distribution for multiple network connections between networks. 257 Network operators can monitor IP traffic flows and then could define 258 policies and rules based on traffic class performance, link bandwidth 259 monetary costs, link load distribution, traffic types, link failures, 260 etc. 262 With I2RS, it would be possible for an I2RS controller to manipulate 263 BGP routes and its parameters that influence BGP bestpath decisions. 264 I2RS controller could act as a central entity that would monitor and 265 manipulate BGP routes based on central network based policies. Such 266 routes would then be injected by a I2RS controller into the network 267 so as to get the load distribution for multiple network connections. 269 4. BGP Events 271 Given the extremely large number of BGP Routes in networks, it is 272 critical to have scalable mechanisms that can be used to monitor for 273 events affecting routing state and, consequently, reachability. In 274 addition, similar tools are needed in order to monitor BGP protocol 275 statistics, which help operators and developers better understand 276 scalability of software and hardware that BGP utilizes. 278 I2RS could provide a publish-subscribe capability to applications to: 280 o request monitoring of BGP routes and related events; and, 282 o subscribe to the I2RS controller to receive events related to BGP 283 routes or other protocol-related events of interest. 285 4.1. Notification of Routing Events 287 There are certain IP prefixes, for example those that are arbitrarily 288 classified by a given network operator as "high visibility" by its 289 end-users, for which immediate notification of changes in their state 290 are extremely useful to know about. Upon notification of such 291 events, a Network Operations Center (NOC) could respond to customer 292 inquiries in a more timely fashion; alternatively, the NOC may decide 293 to perform Traffic Engineering to restore service, etc. 295 Currently, the only way to learn of such events is for a BGP 296 monitoring system to establish a BGP session with a multitude of BGP 297 routers in an AS. Then, the BGP monitoring system needs to look 298 through all BGP UPDATE's in order to identify those events that are 299 of interest to it. Note, this doesn't account for the fact that 300 there are several applications that might be simultaneously 301 interested in learning of events to a given IP prefix nor the fact 302 that some applications may want to dynamically insert or remove "IP 303 prefixes of interest", depending on the needs of their constituent 304 applications. 306 With I2RS, it is conceivable that applications could tell an I2RS 307 controller, through a North-Bound API, their "IP prefixes" (or, 308 AS_PATH's, BGP communities, etc.) that are of interest. For example, 309 a NOC application may be interested in changes to high visibility 310 content or service-provider Web sites; alternatively, a security 311 application may be interested in events associated with a different 312 set of IP prefixes. The I2RS controller would then consolidate the 313 list of IP prefixes, and associated characteristics, to be monitored 314 and program BGP routers in an AS to observe this subset of routes for 315 changes. Some examples of changes in routing state might include: 317 o an IP prefix being announced or withdrawn 319 o an IP prefix being suppressed, due to route flap dampening 321 o an alternative best-path being chosen for a given IP prefix 323 When the requisite events for a BGP Route are observed by a BGP 324 router, it would notify I2RS agents. 326 The I2RS agents would have a publish/subscribe mechanism whereby 327 various sets of applications may subscribe to events of interest. 328 The I2RS controller would then publish these events so applications 329 would immediately receive them and take the appropriate domain- 330 specific action necessary. 332 4.2. Tracing Dropped BGP Routes 334 It is extremely useful to operators to be able to rapidly identify 335 instances where a BGP route is not being propagated within an 336 Autonomous System. At a minimum, this could result in sub-optimal 337 performance when attempting to reach such destinations. 339 There are two instances when this scenario will occur. First, when a 340 Service Provider is using "Soft Reconfiguration Inbound", it allows 341 their ASBR routers to receive a copy of a BGP route, but show that 342 route was not permitted into the Adj-RIB-In most likely as a result 343 of the inbound BGP policy not permitting that IP prefix. Thus, this 344 BGP route is not even eligible for BGP Path Selection. The second 345 instance is where the BGP route is permitted by the inbound BGP 346 policy into the Adj-RIB-In, but due to BGP Path Selection (i.e.: 347 lower LOCAL_PREF, longer AS_PATH length, etc.) was not chosen as the 348 best path and, subsequently, this particular BGP route is not 349 forwarded on to other internal BGP speakers in the AS. In both 350 instances, the BGP route is only visible within the ASBR on which 351 that BGP route was first learned. Needless to say, in large Service 352 Provider networks with a numerous interconnects to a single customer 353 it can be very time-consuming to discover where such a BGP route is 354 learned before ultimately determining why the route was blocked or 355 not preferred. 357 With I2RS, it would be possible for an I2RS controller to rapidly 358 gather information from across a large set of BGP routers in the 359 network to determine at what ASBR's the BGP route is being learned. 360 Next, the I2RS controller could interrogate those routers BGP 361 policies to determine the root cause of why the route was either not 362 learned or not preferred in BGP. Finally, if necessary, the I2RS 363 controller(s) could amend BGP policies and push them out to BGP 364 routers to permit the BGP route or make it a preferred route 365 according to the BGP path selection algorithm. 367 4.3. BGP Protocol Statistics 369 There are a variety of statistics related to the operation of BGP 370 that are invaluable to network operators. These statistics generally 371 help operators, and developers, understand the present state and 372 future scalability of BGP. 374 One statistic that is invaluable to operators is the current number 375 of BGP routes learned through an eBGP session. Operators then apply 376 a command against each eBGP session to limit the maximum number of 377 BGP routes that may be learned through that eBGP session before a 378 warning message is triggered and/or the eBGP session is torn down 379 completely. This configuration capability is often referred to as a 380 "max-prefix limit". This command must be routinely audited and, if 381 necessary, adjusted in order to not trigger a false warning or 382 teardown due to the natural organic growth in BGP routes learned from 383 a given BGP neighbor. 385 I2RS agents could provide an invaluable capability to help audit and 386 re-program the "max-prefix limit" on a periodic basis, which is 387 generally once per day. Specifically, the first task would be for an 388 I2RS controller to validate that there is a "max-prefix limit" 389 applied to every eBGP session. (If there is not, that should either 390 trigger a red alarm to the NOC to manually fix this condition or for 391 the I2RS controller to automatically apply a "max-prefix limit" that 392 would alleviate this hazardous condition). Assuming there is a "max- 393 prefix limit" already in place, the I2RS controller would 394 simultaneously retrieve, from each BGP router, the current number of 395 BGP routes learned through a BGP session and value used for the "max- 396 prefix limit" on that same BGP session. These two values could then 397 be handed off to an application that determines if adjustments in the 398 "max-prefix limit" value are required for each BGP session. The 399 application would then notify the I2RS controller of the subset of 400 eBGP sessions and their associated change in "max-prefix limit" 401 value, whereby the I2RS controller would then adjust the BGP protocol 402 configuration on each requisite BGP router in the network. Finally, 403 it should be noted that the above is just one method whereby "max- 404 prefix limit" values are adjusted. It's similarly possible that the 405 BGP routers may, through the I2RS, pull the "max-prefix limit" values 406 for each eBGP neighbor they have on-board on a periodic basis and 407 validate their accuracy. 409 The above is just one use case related to BGP protocol statistics. 410 There are wealth of other BGP protocol statistics or state 411 information that would be invaluable to have programmatic visibility 412 into that operators do not have today. 414 5. Central membership computation for MPLS based VPNs 416 MPLS based VPNs use route target extended communities to express 417 membership information. Every PE router holds incoming BGP NLRI and 418 processes them to determine membership and then import the NLRI into 419 the appropriate MPLS/VPN routing tables. This consumes resources, 420 both memory and compute on each of the PE devices. 422 An alternative approach is to monitor routing updates on every PE 423 from the attached CEs and then compute membership in a central 424 manner. Once computed the routes are pushed to the VPN RIBs of the 425 participating PEs. 427 This centralization of membership control has a few advantages. 429 o The membership mechanism (route-targets) need not be configured in 430 each of the PEs and can be expressed once centrally. 432 o No resources in the PEs need to be spent to categorize routes into 433 the VRF tables that they belong and to filter out unwanted state. 435 o Doing it centrally means the availability of almost unlimited 436 compute capacity to compute membership and hence can be done in a 437 scaleable manner. 439 o More sophisticated routing policies and filters can be applied 440 during the central import/export process than can be expressed and 441 performed using the traditional route target mechanism. 443 o Routes can be selectively pushed only to the participating PE's 444 further reducing the memory load on the individual routers in the 445 network. This further obviates for a distributed mechanisms such 446 as rt constraints to reduce unnecessary path state in the routers. 448 Note that centrally computation of membership can be applied to other 449 scenarios as well such as VPLS, MVPNs, MAC VPNs and others. 450 Depending on the scenario, what gets monitored from the CE might 451 vary. Central computation will especially help VPLS where multi- 452 homing and load balancing using distributed techniques has 453 particularly been a challenge. 455 Also note that one of the biggest promises of central route 456 computation is simplification and reduction of computation and memory 457 load on all devices in the network. This use case is just one 458 example that illustrates these benefits of central computation very 459 well. 461 Summary of I2RS Capabilities and Interactions: 463 o The ability to read the loc-RIB-In BGP table that gets all the 464 routes that the CE has provided to a PE router. 466 o The ability to install destination based routes in the local RIB 467 of the PE devices. This must include the ability to supply the 468 destination prefix (NLRI), a table identifier, a route preference, 469 a route metric, a next-hop tunnel through which traffic would be 470 carried 472 6. Marking Overlapping Traffic Engineering Routes for Removal 474 It is often the case that routes are advertised not to provide 475 reachability (in the strict sense), but rather to provide optimal 476 reachability, or to engineer the path traffic takes to a particular 477 destination. While this can improve the efficiency of a network's 478 operation, it can also increase the amount of state carried in the 479 control plane beyond the point where the additional state has any 480 real effect on traffic flow. Removing Overlapping Routes 481 [I-D.white-grow-overlapping-routes] provides a mechanism designed to 482 remove these traffic engineering routes once they are beyond the 483 point of actually impacting traffic flows in the network. 485 Summary of I2RS Capabilities and Interactions: 487 o The ability to read the loc-RIB-in BGP table to discover 488 overlapping routes, and determine which may be safely marked for 489 removal. 491 o The ability to modify filtering rules and initiate a re- 492 computation of the local BGP table through those policies to cause 493 specific routes to be marked for removal at the outbound eBGP 494 edge. 496 7. Security Considerations 498 The BGP use cases described in this document assumes use of I2RS 499 programmatic interfaces described in the I2RS framework mentioned in 500 [I-D.ietf-i2rs-architecture]. This document does not change the 501 underlying security issues inherent in the existing in 502 [I-D.ietf-i2rs-architecture]. 504 8. Acknowledgements 506 The authors would like to thank Ed Crabbe, Joel Halpern, Wes George, 507 Carlos Pignataro, Jon Mitchell and Bill Atwood for their comments and 508 suggestions. 510 9. References 512 9.1. Normative References 514 [I-D.ietf-i2rs-architecture] 515 Atlas, A., Halpern, J., Hares, S., Ward, D., and T. 516 Nadeau, "An Architecture for the Interface to the Routing 517 System", draft-ietf-i2rs-architecture-00 (work in 518 progress), August 2013. 520 [RFC1997] Chandrasekeran, R., Traina, P., and T. Li, "BGP 521 Communities Attribute", RFC 1997, August 1996. 523 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 524 Requirement Levels", BCP 14, RFC 2119, March 1997. 526 [RFC2629] Rose, M., "Writing I-Ds and RFCs using XML", RFC 2629, 527 June 1999. 529 [RFC3392] Chandra, R. and J. Scudder, "Capabilities Advertisement 530 with BGP-4", RFC 3392, November 2002. 532 [RFC3552] Rescorla, E. and B. Korver, "Guidelines for Writing RFC 533 Text on Security Considerations", BCP 72, RFC 3552, July 534 2003. 536 [RFC4271] Rekhter, Y., Li, T., and S. Hares, "A Border Gateway 537 Protocol 4 (BGP-4)", RFC 4271, January 2006. 539 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 540 Communities Attribute", RFC 4360, February 2006. 542 [RFC4760] Bates, T., Chandra, R., Katz, D., and Y. Rekhter, 543 "Multiprotocol Extensions for BGP-4", RFC 4760, January 544 2007. 546 9.2. Informative References 548 [I-D.ietf-grow-ops-reqs-for-bgp-error-handling] 549 Shakir, R., "Operational Requirements for Enhanced Error 550 Handling Behaviour in BGP-4", draft-ietf-grow-ops-reqs- 551 for-bgp-error-handling-05 (work in progress), July 2012. 553 [I-D.ietf-i2rs-architecture] 554 Atlas, A., Halpern, J., Hares, S., Ward, D., and T. 555 Nadeau, "An Architecture for the Interface to the Routing 556 System", draft-ietf-i2rs-architecture-00 (work in 557 progress), August 2013. 559 [I-D.mcpherson-irr-routing-policy-considerations] 560 McPherson, D., Amante, S., Osterweil, E., and L. Blunk, 561 "IRR & Routing Policy Configuration Considerations", 562 draft-mcpherson-irr-routing-policy-considerations-01 (work 563 in progress), September 2012. 565 [I-D.white-grow-overlapping-routes] 566 White, R., Retana, A., and S. Hares, "Filtering of 567 Overlapping Routes", draft-white-grow-overlapping- 568 routes-01 (work in progress), February 2013. 570 [RFC2622] Alaettinoglu, C., Villamizar, C., Gerich, E., Kessens, D., 571 Meyer, D., Bates, T., Karrenberg, D., and M. Terpstra, 572 "Routing Policy Specification Language (RPSL)", RFC 2622, 573 June 1999. 575 [RFC2858] Bates, T., Rekhter, Y., Chandra, R., and D. Katz, 576 "Multiprotocol Extensions for BGP-4", RFC 2858, June 2000. 578 [RFC5156] Blanchet, M., "Special-Use IPv6 Addresses", RFC 5156, 579 April 2008. 581 [RFC5575] Marques, P., Sheth, N., Raszuk, R., Greene, B., Mauch, J., 582 and D. McPherson, "Dissemination of Flow Specification 583 Rules", RFC 5575, August 2009. 585 [RFC5735] Cotton, M. and L. Vegoda, "Special Use IPv4 Addresses", 586 RFC 5735, January 2010. 588 Appendix A. BGP Configuration 590 The configuration of BGP is arduous to establish and maintain, 591 particularly on networks whose services have a requirement for 592 complex routing policies. This need is magnified by the need to 593 routinely perform changes to large numbers of BGP routers to, for 594 example: add or remove customer's BGP sessions, announce or withdraw 595 (customer) IP prefixes in BGP, modify BGP policies to effect changes 596 in Traffic Engineering, audit BGP routers to ensure they have 597 consistent and appropriate BGP policies, and others. 599 There are three categories of BGP configuration: 601 1. Local BGP routing protocol configuration: local Autonomous System 602 Number (ASN), BGP path selection properties of the router, 603 injection of (aggregate) routes into BGP, etc. 605 2. Local BGP policies: policies designed to filter and/or manipulate 606 BGP attributes associated with BGP routes learned through BGP 607 sessions. These policies typically live in the global 608 configuration of a BGP router, but are applied on a per-BGP 609 neighbor basis (or, group of BGP neighbors); and, 611 3. BGP neighbor sessions: remote ASN, remote IP address, address 612 families, BGP policies to applied to routes, max-prefix limits, 613 etc. 615 The sum total of BGP configuration on a BGP router is typically the 616 largest quantify of configuration on Service Provider's BGP routers, 617 by a fairly large margin. When that is combined with the large set 618 of routine configuration changes, mentioned above, it should be 619 fairly clear that systematic reading, configuration and control of 620 BGP routers through a mechanism like I2RS would greatly benefit all 621 operators of BGP routers. 623 While it may not be possible to provide programmatic APIs for 624 esoteric vendor-specific policy configuration, it is possible to 625 provide such API's for BGP protocol specific configuration and the 626 more commonly used BGP routing policies. 628 A.1. BGP Protocol Configuration 630 Ability to enable and disable new address families within a BGP 631 protocol for a network of BGP speaking routers is a challenge. The 632 challenge is mainly in keeping track of BGP speaker's feature 633 capabilities and then configuration of new address families on a 634 multiple BGP speakers within a given network. With the necessary 635 information, I2RS agents allow a network operator to push 636 configuration information for enabling and disabling of new address 637 families on a partial or entire set of BGP speakers within a given 638 network. This would assist in building BGP overlay networks as 639 needed. 641 For VPN address families, the main challenge lies in the complex VPN 642 configuration required to setup the control plane for Customer VPNs. 643 The configuration involves creating a Virtual Routing and Forwarding 644 instance (VRF), a Route Distinguisher (RD) that ensures each customer 645 prefixes remains unique across VPNs, and Route Targets (RT) that help 646 ensure that the Customer prefixes are segregated appropriately so 647 that they do not cross the VPN boundaries. I2RS would allow a 648 network operator to push such configuration from a central location 649 where a global VPN provisioning information could be stored. This 650 helps avoid manual configuration of a VPN on multiple routers. 651 Instead the configuration is controlled and pushed though a central 652 I2RS controller using a programmatic set of APIs on targeted set of 653 BGP speakers. 655 Use of I2RS agents to announce protocol configuration information 656 would simplify and automate configuration of BGP protocol in IBGP 657 deployments where the protocol based policies are seldom used. To 658 facilitate such a centralized configuration model, BGP speakers could 659 be extended to use programmatic APIs to announce their feature 660 capabilities as part of protocol initialization to the centralize 661 I2RS agents. This would assist I2RS agents to auto-discover BGP 662 protocol capabilities of various BGP speakers in a given network. 663 I2RS agents in turn would use the information towards enabling/ 664 disabling of BGP specific features on BGP speakers. 666 A.2. BGP Policy Configuration 668 Filtering of BGP routes is strongly recommended to control the 669 announcements of BGP prefixes across the internet. Most providers 670 make extensive use of BGP prefix filtering policies at the edge of 671 their networks. The reasons for filtering BGP prefixes are: 673 o Avoid Unwanted Route Announcements. Filter prefixes that MUST not 674 be routed [RFC5735], [RFC5156]. Filter prefixes that are not 675 allocated by Internet Routing Registries. 677 o Facilitate Route Summarization. Filter prefixes beyond certain 678 agreed prefix mask length between providers. Route Summarization 679 helps control BGP RIB and FIB table size. 681 o Defensive Security. Filter prefixes from Stub customer ASes that 682 are not owned by the customers. Filter customer prefixes 683 announced by other providers. This helps avoid prefix hijacking. 685 A set of standards-based schemas to enable configuration of Local BGP 686 policies and BGP neighbor sessions was realized through the Routing 687 Policy Specification Language (RSPL) [RFC2622]. The RPSL defined a 688 standards-based schemas, or 'objects' as it called them, that 689 defined: 691 o binding of IP prefixes to (one or more) Origin AS, (route 692 objects); 694 o collections of routes (route-set objects); 696 o collections of Autonomous Systems (as-set objects); and, 698 o routing policy of an Autonomous System to/from its adjacent 699 neighbor AS'es, (aut-num objects) 701 Each ASN is responsible for creation, modification and deletion of 702 its RPSL objects in an Internet Routing Registry (IRR). IRR's are 703 typically operated by Regional Internet Registries (RIR's) and a few 704 dozen larger ISP's and independent organizations. The IRR's provide 705 a well-known location for all organizations attached to the Internet 706 to retrieve or update RPSL objects. 708 While still widely and actively used by Internet Service Providers, 709 the prevailing belief is that the data contained in the IRR's is 710 inaccurate, primarily due to a lack of deployed authorization method 711 with respect to the creation of modification of RPSL objects. It 712 should be noted that this criticism is not directed at the previously 713 defined RPSL schemas, but rather at the data contained in RPSL 714 schemas by end-users of the IRR system. Please refer to the IRR And 715 Routing Policy Configuration Considerations 716 [I-D.mcpherson-irr-routing-policy-considerations] document for a more 717 thorough discussion of the history and present state of the IRR's. 719 Currently, RPSL schemas are exchanged between non-routing systems 720 (servers) used within the IRR system. In addition, open-source and 721 proprietary applications create or modify RPSL schemas, as necessary, 722 to signal the announcement (or, withdrawal) of an IP prefix from an 723 ASN or the creation (or, teardown) of a neighbor relationship between 724 two adjacent ASN's. Most importantly, these RPSL schemas are 725 consumed by similar applications to automatically build routing 726 policies, (i.e.: lists of IP prefixes, corresponding Origin ASN's and 727 /or AS_PATH's), that then get translated to device-specific syntax 728 (i.e.: CLI) before being pushed into individual BGP routers to effect 729 routing policy on the network. It is common for Internet Service 730 Providers to perform updates to these routing policies across their 731 entire network on a daily basis. 733 With I2RS it would be desirable to change the last step in the above 734 process so that BGP policies derived from RPSL schemas, and other 735 information sources, are translated into standards-based schemas that 736 are then pushed, or pulled, into individual BGP routers. More 737 generally, I2RS agents could use API's to gather information required 738 to build various types of BGP routing policies plus the corresponding 739 set of Autonomous System Border Routers (ASBR's) where such policies 740 need to be applied in the network and, finally, making those changes 741 to individual network elements so those BGP policies take effect in 742 the network. In doing so, a network operator now has a centralized 743 way of building and making these policies take effect across the 744 network in a coordinated manner. 746 Authors' Addresses 748 Keyur Patel 749 Cisco Systems 750 170 W. Tasman Drive 751 San Jose, CA 95134 752 USA 754 Email: keyupate@cisco.com 756 Rex Fernando 757 Cisco Systems 758 170 W. Tasman Drive 759 San Jose, CA 95134 760 USA 762 Email: rex@cisco.com 763 Hannes Gredler 764 Juniper Networks 765 1194 N. Mathilda Ave 766 Sunnyvale, CA 94089 767 USA 769 Email: hannes@juniper.net 771 Shane Amante 772 Level 3 Communications, Inc. 773 1025 Eldorado Blvd 774 Broomfield, CO 80021 775 USA 777 Email: shane@level3.net 779 Russ White 780 Ericsson 782 Email: russw@riw.us 784 Susan Hares 785 Hickory Hill Consulting 787 Email: shares@ndzh.com