idnits 2.17.1 draft-ietf-grow-bgp-gshut-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 : ---------------------------------------------------------------------------- ** 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 == 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 (August 14, 2014) is 3536 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC4360' is defined on line 417, but no explicit reference was found in the text == Unused Reference: 'BGPWKC' is defined on line 420, but no explicit reference was found in the text == Outdated reference: A later version (-15) exists of draft-ietf-idr-add-paths-09 Summary: 1 error (**), 0 flaws (~~), 5 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group Pierre Francois 3 Internet-Draft Institute IMDEA Networks 4 Intended status: Informational Bruno Decraene 5 Expires: February 15, 2015 Orange 6 Cristel Pelsser 7 Internet Initiative Japan 8 Keyur Patel 9 Clarence Filsfils 10 Cisco Systems 11 August 14, 2014 13 Graceful BGP session shutdown 14 draft-ietf-grow-bgp-gshut-06 16 Abstract 18 This draft describes operational procedures aimed at reducing the 19 amount of traffic lost during planned maintenances of routers or 20 links, involving the shutdown of BGP peering sessions. 22 Status of this Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at http://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on February 15, 2015. 39 Copyright Notice 41 Copyright (c) 2014 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (http://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 This document may contain material from IETF Documents or IETF 55 Contributions published or made publicly available before November 56 10, 2008. The person(s) controlling the copyright in some of this 57 material may not have granted the IETF Trust the right to allow 58 modifications of such material outside the IETF Standards Process. 59 Without obtaining an adequate license from the person(s) controlling 60 the copyright in such materials, this document may not be modified 61 outside the IETF Standards Process, and derivative works of it may 62 not be created outside the IETF Standards Process, except to format 63 it for publication as an RFC or to translate it into languages other 64 than English. 66 Table of Contents 68 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 69 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 70 3. Packet loss upon manual eBGP session shutdown . . . . . . . . 5 71 4. Practices to avoid packet losses . . . . . . . . . . . . . . . 5 72 4.1. Improving availability of alternate paths . . . . . . . . 5 73 4.2. Make before break convergence: g-shut . . . . . . . . . . 6 74 4.2.1. eBGP g-shut . . . . . . . . . . . . . . . . . . . . . 6 75 4.2.2. iBGP g-shut . . . . . . . . . . . . . . . . . . . . . 7 76 4.2.3. Router g-shut . . . . . . . . . . . . . . . . . . . . 7 77 5. Forwarding modes and transient forwarding loops during 78 convergence . . . . . . . . . . . . . . . . . . . . . . . . . 8 79 6. Link Up cases . . . . . . . . . . . . . . . . . . . . . . . . 8 80 6.1. Unreachability local to the ASBR . . . . . . . . . . . . . 8 81 6.2. iBGP convergence . . . . . . . . . . . . . . . . . . . . . 9 82 7. IANA assigned g-shut BGP community . . . . . . . . . . . . . . 9 83 8. Security Considerations . . . . . . . . . . . . . . . . . . . 10 84 9. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 10 85 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 10 86 10.1. Normative References . . . . . . . . . . . . . . . . . . . 10 87 10.2. Informative References . . . . . . . . . . . . . . . . . . 11 88 Appendix A. Alternative techniques with limited applicability . . 11 89 A.1. Multi Exit Discriminator tweaking . . . . . . . . . . . . 11 90 A.2. IGP distance Poisoning . . . . . . . . . . . . . . . . . . 11 91 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 12 93 1. Introduction 95 Routing changes in BGP can be caused by planned, maintenance 96 operations. This document discusses operational procedures to be 97 applied in order to reduce or eliminate losses of packets during the 98 maintenance. These losses come from the transient lack of 99 reachability during the BGP convergence following the shutdown of an 100 eBGP peering session between two Autonomous System Border Routers 101 (ASBR). 103 This document presents procedures for the cases where the forwarding 104 plane is impacted by the maintenance, hence when the use of Graceful 105 Restart does not apply. 107 The procedures described in this document can be applied to reduce or 108 avoid packet loss for outbound and inbound traffic flows initially 109 forwarded along the peering link to be shut down. These procedures 110 trigger, in both involved ASes, rerouting to the alternate path, 111 while allowing routers to keep using old paths until alternate ones 112 are learned, installed in the RIB and in the FIB. This ensures that 113 routers always have a valid route available during the convergence 114 process. 116 The goal of the document is to meet the requirements described in 117 [REQS] at best, without changing the BGP protocol. 119 Still, it explains why reserving a community value for the purpose of 120 BGP session graceful shutdown would reduce the management overhead 121 bound with the solution. It would also allow vendors to provide an 122 automatic graceful shutdown mechanism that does not require any 123 router reconfiguration at maintenance time. 125 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 126 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 127 document are to be interpreted as described in RFC 2119 [RFC2119]. 129 2. Terminology 131 g-shut initiator: a router on which the session shutdown is performed 132 for the maintenance. 134 g-shut neighbor: a router that peers with the g-shut initiator via 135 (one of) the session(s) to be shut down. 137 Initiator AS: the Autonomous System of the g-shut initiator. 139 Neighbor AS: the Autonomous System of the g-shut neighbor. 141 Loss of Connectivity (LoC: the state when a router has no path 142 towards an affected prefix. 144 3. Packet loss upon manual eBGP session shutdown 146 Packets can be lost during a manual shutdown of an eBGP session for 147 two reasons. 149 First, routers involved in the convergence process can transiently 150 lack of paths towards an affected prefix, and drop traffic destined 151 to this prefix. This is because alternate paths can be hidden by 152 nodes of an AS. This happens when the paths are not selected as best 153 by the ASBR that receive them on an eBGP session, or by Route 154 Reflectors that do not propagate them further in the iBGP topology 155 because they do not select them as best. 157 Second, within the AS, the FIB of routers can be transiently 158 inconsistent during the BGP convergence and packets towards affected 159 prefixes can loop and be dropped. Note that these loops only happen 160 when ASBR-to-ASBR encapsulation is not used within the AS. 162 This document only addresses the first reason. 164 4. Practices to avoid packet losses 166 This section describes means for an ISP to reduce the transient loss 167 of packets upon a manual shutdown of a BGP session. 169 4.1. Improving availability of alternate paths 171 All solutions that increase the availability of alternate BGP paths 172 at routers performing packet lookups in BGP tables such as 173 [BestExternal] and [AddPath] help in reducing the LoC bound with 174 manual shutdown of eBGP sessions. 176 One of such solutions increasing diversity in such a way that, at any 177 single step of the convergence process following the eBGP session 178 shutdown, a BGP router does not receive a message withdrawing the 179 only path it currently knows for a given NLRI, allows for a 180 simplified g-shut procedure. 182 Note that the LoC for the inbound traffic of the maintained router, 183 induced by a lack of alternate path propagation within the iBGP 184 topology of a neighboring AS is not under the control of the operator 185 performing the maintenance. The part of the procedure aimed at 186 avoiding LoC for incoming paths can thus be applied even if no LoC 187 are expected for the outgoing paths. 189 4.2. Make before break convergence: g-shut 191 This section describes configurations and actions to be performed for 192 the graceful shutdown of eBGP peering links. 194 The goal of this procedure is to let the paths being shutdown 195 visible, but with a lower LOCAL_PREF value, while alternate paths 196 spread through the iBGP topology. Instead of withdrawing the path, 197 routers of an AS will keep on using it until they become aware of 198 alternate paths. 200 4.2.1. eBGP g-shut 202 4.2.1.1. Pre-configuration 204 On each ASBR supporting the g-shut procedure, an outbound BGP route 205 policy is applied on all iBGP sessions of the ASBR, that: 206 o matches the g-shut community 207 o sets the LOCAL_PREF attribute of the paths tagged with the 208 g-shut community to a low value 209 o removes the g-shut community from the paths. 210 o optionally, adds an AS specific g-shut community on these paths 211 to indicate that these are to be withdrawn soon. If some 212 ingress ASBRs reset the LOCAL_PREF attribute, this AS specific 213 g-shut community will be used to override other LOCAL_PREF 214 preference changes. 216 Note that in the case where an AS is aggregating multiple routes 217 under a covering prefix, it is recommended to filter out the g-shut 218 community from the resulting aggregate BGP route. By doing so, the 219 setting of the g-shut community on one of the aggregated routes will 220 not let the entire aggregate inherit the community. Not doing so 221 would let the entire aggregate undergo the g-shut behavior. 223 4.2.1.2. Operations at maintenance time 225 On the g-shut initiator, upon maintenance time, it is required to: 226 o apply an outbound BGP route policy on the maintained eBGP session 227 to tag the paths propagated over the session with the g-shut 228 community. This will trigger the BGP implementation to re- 229 advertise all active routes previously advertised, and tag them 230 with the g-shut community. 231 o apply an inbound BGP route policy on the maintained eBGP session 232 to tag the paths received over the session with the g-shut 233 community. 235 o wait for convergence to happen. 236 o perform a BGP session shutdown. 238 4.2.1.3. BGP implementation support for G-Shut 240 A BGP router implementation MAY provide features aimed at automating 241 the application of the graceful shutdown procedures described above. 243 Upon a session shutdown specified as graceful by the operator, a BGP 244 implementation supporting a g-shut feature SHOULD: 246 1. On the eBGP side, update all the paths propagated over the 247 corresponding eBGP session, tagging the GSHUT community to them. 248 Any subsequent update sent to the session being gracefully shut 249 down would be tagged with the GSHUT community. 250 2. On the iBGP side, lower the LOCAL_PREF value of the paths 251 received over the eBGP session being shut down, upon their 252 propagation over iBGP sessions. Optionally, also tag these 253 paths with an AS specific g-shut community. Note that 254 alternatively, the LOCAL_PREF of the paths received over the 255 eBGP session can be lowered on the g-shut initiator itself, 256 instead of only when propagating over its iBGP sessions. 257 3. Optionally shut down the session after a configured time. 258 4. Prevent the GSHUT community from being inherited by a path that 259 would aggregate some paths tagged with the GSHUT community. 260 This behavior avoids the GSHUT procedure to be applied to the 261 aggregate upon the graceful shutdown of one of its covered 262 prefixes. 264 A BGP implementation supporting a g-shut feature SHOULD also 265 automatically install the BGP policies that are supposed to be 266 configured, as decribed in Section 4.2.1.1 for sessions over which 267 g-shut is to be supported. 269 4.2.2. iBGP g-shut 271 If the iBGP topology is viable after the maintenance of the session, 272 i.e, if all BGP speakers of the AS have an iBGP signaling path for 273 all prefixes advertised on this g-shut iBGP session, then the 274 shutdown of an iBGP session does not lead to transient 275 unreachability. 277 4.2.3. Router g-shut 279 In the case of a shutdown of a router, a reconfiguration of the 280 outbound BGP route policies of the g-shut initiator SHOULD be 281 performed to set a low LOCAL_PREF value for the paths originated by 282 the g-shut initiator (e.g, BGP aggregates redistributed from other 283 protocols, including static routes). 285 This behavior is equivalent to the recommended behavior for paths 286 "redistributed" from eBGP sessions to iBGP sessions in the case of 287 the shutdown of an ASBR. 289 5. Forwarding modes and transient forwarding loops during convergence 291 The g-shut procedure or the solutions improving the availability of 292 alternate paths, do not change the fact that BGP convergence and the 293 subsequent FIB updates are run independently on each router of the 294 ASes. If the AS applying the solution does not rely on encapsulation 295 to forward packets from the Ingress Border Router to the Egress 296 Border Router, then transient forwarding loops and consequent packet 297 losses can occur during the convergence process. If zero LoC is 298 required, encapsulation is required between ASBRs of the AS. 300 6. Link Up cases 302 We identify two potential causes for transient packet losses upon an 303 eBGP link up event. The first one is local to the g-no-shut 304 initiator, the second one is due to the BGP convergence following the 305 injection of new best paths within the iBGP topology. 307 6.1. Unreachability local to the ASBR 309 An ASBR that selects as best a path received over a newly brought up 310 eBGP session may transiently drop traffic. This can typically happen 311 when the nexthop attribute differs from the IP address of the eBGP 312 peer, and the receiving ASBR has not yet resolved the MAC address 313 associated with the IP address of that "third party" nexthop. 315 A BGP speaker implementation could avoid such losses by ensuring that 316 "third party" nexthops are resolved before installing paths using 317 these in the RIB. 319 If the link up event corresponds to an eBGP session that is being 320 manually brought up, over an already up multi-access link, then the 321 operator can ping third party nexthops that are expected to be used 322 before actually bringing the session up, or ping directed broadcast 323 the subnet IP address of the link. By proceeding like this, the MAC 324 addresses associated with these third party nexthops will be resolved 325 by the g-no-shut initiator. 327 6.2. iBGP convergence 329 Corner cases leading to LoC can occur during an eBGP link up event. 331 A typical example for such transient unreachability for a given 332 prefix is the following: 334 Let's consider 3 route reflectors RR1, RR2, RR3. There is a full 335 mesh of iBGP session between them. 337 1. RR1 is initially advertising the current best path to the 338 members of its iBGP RR full-mesh. It propagated that path 339 within its RR full-mesh. RR2 knows only that path towards the 340 prefix. 341 2. RR3 receives a new best path originated by the "g-no-shut" 342 initiator, being one of its RR clients. RR3 selects it as best, 343 and propagates an UPDATE within its RR full-mesh, i.e., to RR1 344 and RR2. 345 3. RR1 receives that path, reruns its decision process, and 346 picks this new path as best. As a result, RR1 withdraws its 347 previously announced best-path on the iBGP sessions of its RR 348 full-mesh. 349 4. If, for any reason, RR3 processes the withdraw generated in 350 step 3, before processing the update generated in step 2, RR3 351 transiently suffers from unreachability for the affected prefix. 353 The use of [BestExternal] among the RR of the iBGP full-mesh can 354 solve these corner cases by ensuring that within an AS, the 355 advertisement of a new route is not translated into the withdraw of a 356 former route. 358 Indeed, "best-external" ensures that an ASBR does not withdraw a 359 previously advertised (eBGP) path when it receives an additional, 360 preferred path over an iBGP session. Also, "best-intra-cluster" 361 ensures that a RR does not withdraw a previously advertised (iBGP) 362 path to its non clients (e.g. other RRs in a mesh of RR) when it 363 receives a new, preferred path over an iBGP session. 365 7. IANA assigned g-shut BGP community 367 Applying the g-shut procedure is rendered much easier with the use of 368 a single g-shut community value which could be used on all eBGP 369 sessions, for both inbound and outbound signaling. The community 370 value 0xFFFF0000 has been assigned by IANA for this purpose. 372 For Internet routes, a non transitive extended community will be 373 reserved from the pool defined in [EXT_POOL]. Using such a community 374 type allows for not leaking graceful signaling out of the AS 375 boundaries, without the need to explicitly configure filters to strip 376 the community off upon path propagation. 378 8. Security Considerations 380 By providing the g-shut service to a neighboring AS, an ISP provides 381 means to this neighbor to lower the LOCAL_PREF value assigned to the 382 paths received from this neighbor. 384 The neighbor could abuse the technique and do inbound traffic 385 engineering by declaring some prefixes as undergoing a maintenance so 386 as to switch traffic to another peering link. 388 If this behavior is not tolerated by the ISP, it SHOULD monitor the 389 use of the g-shut community by this neighbor. 391 ASes using the regular (transitive) g-shut community SHOULD remove 392 the community from neighboring ASes that do not support the g-shut 393 procedure. Doing so prevents malignant remote ASes from using the 394 community through intermediate ASes that do not support the feature, 395 in order to perform inbound traffic engineering. ASes using the non- 396 transitive extended community do not need to do this as the community 397 is non transitive and hence cannot be used by remote ASes. 399 9. Acknowledgments 401 The authors wish to thank Olivier Bonaventure and Pradosh Mohapatra 402 for their useful comments on this work. 404 10. References 406 10.1. Normative References 408 [REQS] Decraene, B., Francois, P., Pelsser, C., Ahmad, Z., 409 Armengol, A., and T. Takeda, "Requirements for the 410 graceful shutdown of BGP sessions", RFC 6198. 412 [EXT_POOL] 413 Decraene, B. and P. Francois, "Assigned BGP extended 414 communities", 415 draft-ietf-idr-reserved-extended-communities-06. 417 [RFC4360] Sangli, S., Tappan, D., and Y. Rekhter, "BGP Extended 418 Communities Attribute", RFC 4360, February 2006. 420 [BGPWKC] "http://www.iana.org/assignments/ 421 bgp-well-known-communities". 423 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 424 Requirement Levels", BCP 14, RFC 2119, March 1997. 426 10.2. Informative References 428 [AddPath] D. Walton, E. Chen, A. Retana, and J. Scudder, 429 "Advertisement of Multiple Paths in BGP", 430 draft-ietf-idr-add-paths-09.txt (work in progress). 432 [BestExternal] 433 Marques, P., Fernando, R., Chen, E., Mohapatra, P., and H. 434 Gredler, "Advertisement of the best-external route to 435 IBGP", draft-ietf-idr-best-external-05.txt. 437 Appendix A. Alternative techniques with limited applicability 439 A few alternative techniques have been considered to provide g-shut 440 capabilities but have been rejected due to their limited 441 applicability. This section describe them for possible reference. 443 A.1. Multi Exit Discriminator tweaking 445 The MED attribute of the paths to be avoided can be increased so as 446 to force the routers in the neighboring AS to select other paths. 448 The solution only works if the alternate paths are as good as the 449 initial ones with respect to the Local-Pref value and the AS Path 450 Length value. In the other cases, increasing the MED value will not 451 have an impact on the decision process of the routers in the 452 neighboring AS. 454 A.2. IGP distance Poisoning 456 The distance to the BGP nexthop corresponding to the maintained 457 session can be increased in the IGP so that the old paths will be 458 less preferred during the application of the IGP distance tie-break 459 rule. However, this solution only works for the paths whose 460 alternates are as good as the old paths with respect to their Local- 461 Pref value, their AS Path length, and their MED value. 463 Also, this poisoning cannot be applied when nexthop self is used as 464 there is no nexthop specific to the maintained session to poison in 465 the IGP. 467 Authors' Addresses 469 Pierre Francois 470 Institute IMDEA Networks 471 Avda. del Mar Mediterraneo, 22 472 Leganese 28918 473 ES 475 Email: pierre.francois@imdea.org 477 Bruno Decraene 478 Orange 479 38-40 rue du General Leclerc 480 92794 Issy Moulineaux cedex 9 481 FR 483 Email: bruno.decraene@orange.com 485 Cristel Pelsser 486 Internet Initiative Japan 487 Jinbocho Mitsui Bldg. 488 1-105 Kanda Jinbo-cho 489 Tokyo 101-0051 490 JP 492 Email: cristel@iij.ad.jp 494 Keyur Patel 495 Cisco Systems 496 170 West Tasman Dr 497 San Jose, CA 95134 498 US 500 Email: keyupate@cisco.com 502 Clarence Filsfils 503 Cisco Systems 504 De kleetlaan 6a 505 Diegem 1831 506 BE 508 Email: cfilsfil@cisco.com