idnits 2.17.1 draft-ietf-isis-reverse-metric-11.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 : ---------------------------------------------------------------------------- == There are 1 instance of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (July 2, 2018) is 2125 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Possible downref: Non-RFC (?) normative reference: ref. 'ISO10589' == Outdated reference: A later version (-07) exists of draft-shen-isis-spine-leaf-ext-03 Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Networking Working Group N. Shen 3 Internet-Draft Cisco Systems 4 Intended status: Standards Track S. Amante 5 Expires: January 3, 2019 Apple, Inc. 6 M. Abrahamsson 7 T-Systems Nordic 8 July 2, 2018 10 IS-IS Routing with Reverse Metric 11 draft-ietf-isis-reverse-metric-11 13 Abstract 15 This document describes a mechanism to allow IS-IS routing to quickly 16 and accurately shift traffic away from either a point-to-point or 17 multi-access LAN interface during network maintenance or other 18 operational events. This is accomplished by signaling adjacent IS-IS 19 neighbors with a higher reverse metric, i.e., the metric towards the 20 signaling IS-IS router. 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 January 3, 2019. 39 Copyright Notice 41 Copyright (c) 2018 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 Table of Contents 56 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 57 1.1. Node and Link Isolation . . . . . . . . . . . . . . . . . 2 58 1.2. Distributed Forwarding Planes . . . . . . . . . . . . . . 3 59 1.3. Spine-Leaf Applications . . . . . . . . . . . . . . . . . 3 60 1.4. LDP IGP Synchronization . . . . . . . . . . . . . . . . . 3 61 1.5. IS-IS Reverse Metric . . . . . . . . . . . . . . . . . . 3 62 1.6. Specification of Requirements . . . . . . . . . . . . . . 4 63 2. IS-IS Reverse Metric TLV . . . . . . . . . . . . . . . . . . 4 64 3. Elements of Procedure . . . . . . . . . . . . . . . . . . . . 6 65 3.1. Processing Changes to Default Metric . . . . . . . . . . 6 66 3.2. Multi-Topology IS-IS Support on Point-to-point links . . 7 67 3.3. Multi-Access LAN Procedures . . . . . . . . . . . . . . . 7 68 3.4. Point-To-Point Link Procedures . . . . . . . . . . . . . 8 69 3.5. LDP/IGP Synchronization on LANs . . . . . . . . . . . . . 8 70 3.6. Operational Guidelines . . . . . . . . . . . . . . . . . 9 71 4. Security Considerations . . . . . . . . . . . . . . . . . . . 9 72 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 73 6. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 10 74 7. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 75 7.1. Normative References . . . . . . . . . . . . . . . . . . 10 76 7.2. Informative References . . . . . . . . . . . . . . . . . 11 77 Appendix A. Node Isolation Challenges . . . . . . . . . . . . . 11 78 Appendix B. Link Isolation Challenges . . . . . . . . . . . . . 12 79 Appendix C. Contributors' Addresses . . . . . . . . . . . . . . 13 80 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 13 82 1. Introduction 84 The IS-IS [ISO10589] routing protocol has been widely used in 85 Internet Service Provider IP/MPLS networks. Operational experience 86 with the protocol, combined with ever increasing requirements for 87 lossless operations have demonstrated some operational issues. This 88 document describes the issues and a mechanism for mitigating them. 90 1.1. Node and Link Isolation 92 IS-IS routing mechanism has the overload-bit, which can be used by 93 operators to perform disruptive maintenance on the router. But in 94 many operational maintenance cases, it is not necessary to divert all 95 the traffic away from this node. It is necessary to avoid only a 96 single link during the maintenance. More detailed descriptions of 97 the challenges can be found in Appendix A and Appendix B of this 98 document. 100 1.2. Distributed Forwarding Planes 102 In a distributed forwarding platform, different forwarding line-cards 103 may have interfaces and IS-IS connections to neighbor routers. If 104 one of the line-card's software resets, it may take some time for the 105 forwarding entries to be fully populated on the line-card, in 106 particular if the router is a PE (Provider Edge) router in ISP's MPLS 107 VPN. An IS-IS adjacency may be established with a neighbor router 108 long before the entire BGP VPN prefixes are downloaded to the 109 forwarding table. It is important to signal to the adjacent IS-IS 110 routers to raise metric values and not to use the corresponding IS-IS 111 adjacency inbound to this router if possible. Temporarily signaling 112 the 'Reverse Metric' over this link to discourage the traffic via the 113 corresponding line-card will help to reduce the traffic loss in the 114 network. In the meantime, the remote PE routers will select a 115 different set of PE routers for the BGP best path calculation or use 116 a different link towards the same PE router on which a line-card is 117 resetting. 119 1.3. Spine-Leaf Applications 121 In the IS-IS Spine-Leaf extension [I-D.shen-isis-spine-leaf-ext], the 122 leaf nodes will perform equal-cost or unequal-cost load sharing 123 towards all the spine nodes. In certain operational cases, for 124 instance, when one of the backbone links on a spine node is 125 congested, a spine node can push a higher metric towards the 126 connected leaf nodes to reduce the transit traffic through the 127 corresponding spine node or link. 129 1.4. LDP IGP Synchronization 131 In the [RFC5443], a mechanism is described to achieve LDP IGP 132 synchronization by using the maximum link metric value on the 133 interface. But in the case of a new IS-IS node joining the broadcast 134 network (LAN), it is not optimal to change all the nodes on the LAN 135 to the maximum link metric value, as described in [RFC6138]. In this 136 case, the Reverse Metric can be used to discourage both outbound and 137 inbound traffic without affecting the traffic of other IS-IS nodes on 138 the LAN. 140 1.5. IS-IS Reverse Metric 142 This document uses the routing protocol itself as the transport 143 mechanism to allow one IS-IS router to advertise a "reverse metric" 144 in an IS-IS Hello (IIH) PDU to an adjacent node on a point-to-point 145 or multi-access LAN link. This would allow the provisioning to be 146 performed only on a single node, setting a "reverse metric" on a link 147 and have traffic bidirectionally shift away from that link gracefully 148 to alternate, viable paths. 150 This Reverse Metric mechanism is used for both point-to-point and 151 multi-access LAN links. Unlike the point-to-point links, the IS-IS 152 protocol currently does not have a way to influence the traffic 153 towards a particular node on LAN links. This mechanism provides IS- 154 IS routing the capability of altering traffic in both directions on 155 either a point-to-point link or a multi-access link of an IS-IS node. 157 The metric value in the "reverse metric" TLV and the TE metric in the 158 sub-TLV being advertised is an offset or relative metric to be added 159 to the existing local link and TE metric values of the receiver. 161 1.6. Specification of Requirements 163 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 164 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 165 "OPTIONAL" in this document are to be interpreted as described in BCP 166 14 [RFC2119] [RFC8174] when, and only when, they appear in all 167 capitals, as shown here. 169 2. IS-IS Reverse Metric TLV 171 The Reverse Metric TLV is composed of a 1 octet field of Flags, a 3 172 octet field containing an IS-IS Metric Value, and a 1 octet Traffic 173 Engineering (TE) sub-TLV length field representing the length of a 174 variable number of Extended Intermediate System (IS) Reachability 175 sub-TLVs. If the "sub-TLV len" is non-zero, then the Value field 176 MUST also contain one or more Extended IS Reachability sub-TLVs. 178 The Reverse Metric TLV is optional. The Reverse Metric TLV may be 179 present in any IS-IS Hello PDU. A sender MUST only transmit a single 180 Reverse Metric TLV in a IS-IS Hello PDU. If a received IS-IS Hello 181 PDU contains more than one Reverse Metric TLV, an implementation 182 SHOULD ignore all the Reverse Metric TLVs and treat it as an error 183 condition. 185 0 1 2 3 186 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 187 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 188 | Type | Length | Flags | Metric 189 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 190 Metric (Continue) | sub-TLV Len |Optional sub-TLV 191 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 193 Reverse Metric TLV 195 TYPE: TBD (be replaced by the value that IANA allocates) 196 LENGTH: variable (5 - 255 octets) 197 VALUE: 199 Flags (1 octet) 200 Metric (3 octets) 201 sub-TLV length (1 octet) 202 sub-TLV data (0 - 250 octets) 204 0 1 2 3 4 5 6 7 205 +-+-+-+-+-+-+-+-+ 206 | Reserved |U|W| 207 +-+-+-+-+-+-+-+-+ 209 Figure 1: Flags 211 The Metric field contains a 24-bit unsigned integer. This value is a 212 metric offset that a neighbor SHOULD add to the existing, configured 213 "default metric" for the IS-IS link. Refer to "Elements of 214 Procedure", in Section 3 for details on how an IS-IS router should 215 process the Metric field in a Reverse Metric TLV. 217 There are currently only two Flag bits defined. 219 W bit (0x01): The "Whole LAN" bit is only used in the context of 220 multi-access LANs. When a Reverse Metric TLV is transmitted from a 221 node to the Designated Intermediate System (DIS), if the "Whole LAN" 222 bit is set (1), then a DIS SHOULD add the received Metric value in 223 the Reverse Metric TLV to each node's existing "default metric" in 224 the Pseudonode LSP. If the "Whole LAN" bit is not set (0), then a 225 DIS SHOULD add the received Metric value in the Reverse Metric TLV to 226 the existing "default metric" in the Pseudonode LSP for the single 227 node from whom the Reverse Metric TLV was received. Please refer to 228 "Multi-Access LAN Procedures", in Section 3.3, for additional 229 details. The W bit MUST be clear when a Reverse Metric TLV is 230 transmitted in an IIH PDU on a point-to-point link, and MUST be 231 ignored when received on a point-to-point link. 233 U bit (0x02): The "Unreachable" bit specifies that the metric 234 calculated by addition of the reverse metric value to the "default 235 metric" is limited to (2^24-1). This "U" bit applies to both the 236 default metric in the Extended IS Reachability TLV and the TE 237 default-metric sub-TLV of the link. This is only relevant to the IS- 238 IS "wide" metric mode. 240 The Reverse Metric TLV can include sub-TLVs when an IS-IS router 241 wishes to signal to its neighbor to raise its Traffic Engineering 242 (TE) Metric over the link. In this document, only the "Traffic 243 Engineering Default Metric" sub-TLV [RFC5305], sub-TLV Type 18, is 244 defined and MAY be included in the Reverse Metric TLV, because that 245 is a similar 'reverse metric' operation to be used in TE 246 computations. Upon receiving this TE METRIC sub-TLV in a Reverse 247 Metric TLV, a node SHOULD add the received TE metric offset value to 248 its existing, configured TE default metric within its Extended IS 249 Reachability TLV. Use of other sub-TLVs is outside the scope of this 250 document. The "sub-TLV Len" value MUST be set to zero when an IS-IS 251 router does not have TE sub-TLVs that it wishes to send to its IS-IS 252 neighbor. 254 3. Elements of Procedure 256 3.1. Processing Changes to Default Metric 258 The Metric field, in the Reverse Metric TLV, is a "reverse offset 259 metric" that will either be in the range of 0 - 63 when a "narrow" 260 IS-IS metric is used (IS Neighbors TLV, Pseudonode LSP) [RFC1195] or 261 in the range of 0 - (2^24 - 2) when a "wide" Traffic Engineering 262 metric value is used, (Extended IS Reachability TLV) [RFC5305] 263 [RFC5817]. It is important to use the same IS-IS metric mode on both 264 ends of the link. On the receiving side of the 'reverse-metric' TLV, 265 the accumulated value of configured metric and the reverse-metric 266 needs to be limited to 63 in "narrow" metric mode and to (2^24 - 2) 267 in "wide" metric mode. This applies to both the default metric of 268 Extended IS Reachability TLV and the TE default-metric sub-TLV in LSP 269 or Pseudonode LSP for the "wide" metric mode case. If the "U" bit is 270 present in the flags, the accumulated metric value is to be limited 271 to (2^24 - 1) for both the normal link metric and TE metric in IS-IS 272 "wide" metric mode. 274 If an IS-IS router is configured to originate a TE Default Metric 275 sub-TLV for a link, but receives a Reverse Metric TLV from its 276 neighbor that does not contain a TE Default Metric sub-TLV, then the 277 IS-IS router MUST NOT change the value of its TE Default Metric sub- 278 TLV for that link. 280 3.2. Multi-Topology IS-IS Support on Point-to-point links 282 The Reverse Metric TLV is applicable to Multi-Topology IS-IS (M-ISIS) 283 [RFC5120]. On point-to-point links, if an IS-IS router is configured 284 for M-ISIS, it MUST send only a single Reverse Metric TLV in IIH PDUs 285 toward its neighbor(s) on the designated link. When an M-ISIS router 286 receives a Reverse Metric TLV, it MUST add the received Metric value 287 to its default metric in all Extended IS Reachability TLVs for all 288 topologies. If an M-ISIS router receives a Reverse Metric TLV with a 289 TE Default Metric sub-TLV, then the M-ISIS router MUST add the 290 received TE Default Metric value to each of its TE Default Metric 291 sub-TLVs in all of its MT Intermediate Systems TLVs. If an M-ISIS 292 router is configured to advertise TE Default Metric sub-TLVs for one 293 or more topologies, but does not receive a TE Default Metric sub-TLV 294 in a Reverse Metric TLV, then the M-ISIS router MUST NOT change the 295 value in each of the TE Default Metric sub-TLVs for all topologies. 297 3.3. Multi-Access LAN Procedures 299 On a Multi-Access LAN, only the DIS SHOULD act upon information 300 contained in a received Reverse Metric TLV. All non-DIS nodes MUST 301 silently ignore a received Reverse Metric TLV. The decision process 302 of the routers on the LAN MUST follow the procedure in section 303 7.2.8.2 of [ISO10589], and use the "Two-way connectivity check" 304 during the topology and route calculation. 306 The Reverse Metric TE sub-TLV also applies to the DIS. If a DIS is 307 configured to apply TE over a link and it receives TE metric sub-TLV 308 in a Reverse Metric TLV, it should update the TE Default Metric sub- 309 TLV value of the corresponding Extended IS Reachability TLV or insert 310 a new one if not present. 312 In the case of multi-access LANs, the "W" Flags bit is used to signal 313 from a non-DIS to the DIS whether to change the metric and, 314 optionally Traffic Engineering parameters for all nodes in the 315 Pseudonode LSP or solely the node on the LAN originating the Reverse 316 Metric TLV. 318 A non-DIS node, e.g., Router B, attached to a multi-access LAN will 319 send the DIS a Reverse Metric TLV with the W bit clear when Router B 320 wishes the DIS to add the Metric value to the default metric 321 contained in the Pseudonode LSP specific to just Router B. Other 322 non-DIS nodes, e.g., Routers C and D, may simultaneously send a 323 Reverse Metric TLV with the W bit clear to request the DIS to add 324 their own Metric value to their default metric contained in the 325 Pseudonode LSP. When the DIS receives a properly formatted Reverse 326 Metric TLV with the W bit clear, the DIS MUST only add the default 327 metric contained in its Pseudonode LSP for the specific neighbor that 328 sent the corresponding Reverse Metric TLV. 330 As long as at least one IS-IS node on the LAN sending the signal to 331 DIS with the W bit set, the DIS would add the metric value in the 332 Reverse Metric TLV to all neighbor adjacencies in the Pseudonode LSP, 333 regardless if some of the nodes on the LAN advertise the Reverse 334 Metric TLV without the W bit set. The DIS MUST use the reverse 335 metric of the highest source MAC address Non-DIS advertising the 336 Reverse Metric TLV with the W bit set. The DIS MUST use the metric 337 value towards the nodes which explicitly advertise the Reverse Metric 338 TLV. 340 Local provisioning on the DIS to adjust the default metric(s) 341 contained in the Pseudonode LSP MUST take precedence over received 342 Reverse Metric TLVs. For instance, local policy on the DIS may be 343 provisioned to ignore the W bit signaling on a LAN. 345 Multi-Topology IS-IS [RFC5120] specifies there is no change to 346 construction of the Pseudonode LSP, regardless of the Multi-Topology 347 capabilities of a multi-access LAN. If any MT capable node on the 348 LAN advertises the Reverse Metric TLV to the DIS, the DIS should 349 update, as appropriate, the default metric contained in the 350 Pseudonode LSP. If the DIS updates the default metric in and floods 351 a new Pseudonode LSP, those default metric values will be applied to 352 all topologies during Multi-Topology SPF calculations. 354 3.4. Point-To-Point Link Procedures 356 On a point-to-point link, there is already a "configured" IS-IS 357 interface metric to be applied over the link towards the IS-IS 358 neighbor. 360 When IS-IS receives the IIH PDU with the "Reverse Metric" on a point- 361 to-point link and if the local policy allows the supporting of 362 "Reverse Metric", it MUST add the metric value in "reverse metric" 363 TLV according to the rules described in Section 3.1 and Section 3.2. 365 3.5. LDP/IGP Synchronization on LANs 367 As described in [RFC6138] when a new IS-IS node joins a broadcast 368 network, it is unnecessary and sometimes even harmful for all IS-IS 369 nodes on the LAN to advertise maximum link metric. [RFC6138] 370 proposes a solution to have the new node not advertise its adjacency 371 towards the pseudo-node when it is not in a "cut-edge" position. 373 With the introduction of Reverse Metric in this document, a simpler 374 alternative solution to the above mentioned problem can be used. The 375 Reverse Metric allows the new node on the LAN to advertise its 376 inbound metric value to be the maximum and this puts the link of this 377 new node in the last resort position without impacting the other IS- 378 IS nodes on the same LAN. 380 Specifically, when IS-IS adjacencies are being established by the new 381 node on the LAN, besides setting the maximum link metric value (2^24 382 - 2) on the interface of the LAN for LDP IGP synchronization as 383 described in [RFC5443], it SHOULD advertise the maximum metric offset 384 value in the Reverse Metric TLV in its IIH PDU sent on the LAN. It 385 SHOULD continue this advertisement until it completes all the LDP 386 label binding exchanges with all the neighbors over this LAN, either 387 by receiving the LDP End-of-LIB [RFC5919] for all the sessions or by 388 exceeding the provisioned timeout value for the node LDP/IGP 389 synchronization. 391 3.6. Operational Guidelines 393 A router MUST advertise a Reverse Metric TLV toward a neighbor only 394 for the period during which it wants a neighbor to temporarily update 395 its IS-IS metric or TE parameters towards it. 397 The use of Reverse Metric does not alter IS-IS metric parameters 398 stored in a router's persistent provisioning database. 400 Routers that receive a Reverse Metric TLV MAY send a syslog message 401 or SNMP trap, in order to assist in rapidly identifying the node in 402 the network that is advertising an IS-IS metric or Traffic 403 Engineering parameters different from that which is configured 404 locally on the device. 406 When the link TE metric is raised to (2^24 - 1) [RFC5817], either due 407 to the reverse-metric mechanism or by explicit user configuration, 408 this SHOULD immediately trigger the CSPF re-calculation to move the 409 TE traffic away from that link. It is RECOMMENDED also that the CSPF 410 does the immediate CSPF re-calculation when the TE metric is raised 411 to (2^24 - 2) to be the last resort link. 413 It is RECOMMENDED that implementations provide a capability to 414 disable any changes to a node's individual interface default metric 415 or Traffic Engineering parameters based upon receiving a properly 416 formatted Reverse Metric TLVs. 418 4. Security Considerations 420 The enhancement in this document makes it possible for one IS-IS 421 router to manipulate the IS-IS default metric and, optionally, 422 Traffic Engineering parameters of adjacent IS-IS neighbors. Although 423 IS-IS routers within a single Autonomous System nearly always are 424 under the control of a single administrative authority, it is highly 425 RECOMMENDED that operators configure authentication of IS-IS PDUs to 426 mitigate use of the Reverse Metric TLV as a potential attack vector, 427 particularly on multi-access LANs. 429 5. IANA Considerations 431 This document requests that IANA allocate from the IS-IS TLV 432 Codepoints Registry a new TLV, referred to as the "Reverse Metric" 433 TLV, possibly from the "Unassigned" range of 244-250, with the 434 following attributes: IIH = y, LSP = n, SNP = n, Purge = n. 436 6. Acknowledgments 438 The authors would like to thank Mike Shand, Dave Katz, Guan Deng, 439 Ilya Varlashkin, Jay Chen, Les Ginsberg, Peter Ashwood-Smith, Uma 440 Chunduri, Alexander Okonnikov, Jonathan Harrison, Dave Ward, Himanshu 441 Shah, Wes George, Danny McPherson, Ed Crabbe, Russ White, Robert 442 Razsuk, Tom Petch and Acee Lindem for their comments and 443 contributions. 445 This document was produced using Marshall Rose's xml2rfc tool. 447 7. References 449 7.1. Normative References 451 [ISO10589] 452 ISO, "Intermediate system to Intermediate system routeing 453 information exchange protocol for use in conjunction with 454 the Protocol for providing the Connectionless-mode Network 455 Service (ISO 8473)", ISO/IEC 10589:2002. 457 [RFC1195] Callon, R., "Use of OSI IS-IS for routing in TCP/IP and 458 dual environments", RFC 1195, DOI 10.17487/RFC1195, 459 December 1990, . 461 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 462 Requirement Levels", BCP 14, RFC 2119, 463 DOI 10.17487/RFC2119, March 1997, . 466 [RFC5120] Przygienda, T., Shen, N., and N. Sheth, "M-ISIS: Multi 467 Topology (MT) Routing in Intermediate System to 468 Intermediate Systems (IS-ISs)", RFC 5120, 469 DOI 10.17487/RFC5120, February 2008, . 472 [RFC5305] Li, T. and H. Smit, "IS-IS Extensions for Traffic 473 Engineering", RFC 5305, DOI 10.17487/RFC5305, October 474 2008, . 476 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 477 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 478 May 2017, . 480 7.2. Informative References 482 [I-D.shen-isis-spine-leaf-ext] 483 Shen, N., Ginsberg, L., and S. Thyamagundalu, "IS-IS 484 Routing for Spine-Leaf Topology", draft-shen-isis-spine- 485 leaf-ext-03 (work in progress), March 2017. 487 [RFC5443] Jork, M., Atlas, A., and L. Fang, "LDP IGP 488 Synchronization", RFC 5443, DOI 10.17487/RFC5443, March 489 2009, . 491 [RFC5817] Ali, Z., Vasseur, JP., Zamfir, A., and J. Newton, 492 "Graceful Shutdown in MPLS and Generalized MPLS Traffic 493 Engineering Networks", RFC 5817, DOI 10.17487/RFC5817, 494 April 2010, . 496 [RFC5919] Asati, R., Mohapatra, P., Chen, E., and B. Thomas, 497 "Signaling LDP Label Advertisement Completion", RFC 5919, 498 DOI 10.17487/RFC5919, August 2010, . 501 [RFC6138] Kini, S., Ed. and W. Lu, Ed., "LDP IGP Synchronization for 502 Broadcast Networks", RFC 6138, DOI 10.17487/RFC6138, 503 February 2011, . 505 Appendix A. Node Isolation Challenges 507 On rare occasions, it is necessary for an operator to perform 508 disruptive network maintenance on an entire IS-IS router node, i.e., 509 major software upgrades, power/cooling augments, etc. In these 510 cases, an operator will set the IS-IS Overload Bit (OL-bit) within 511 the Link State Protocol Data Units (LSPs) of the IS-IS router about 512 to undergo maintenance. The IS-IS router immediately floods its 513 updated LSPs to all IS-IS routers in the IS-IS domain. Upon receipt 514 of the updated LSPs, all IS-IS routers recalculate their Shortest 515 Path First (SPF) tree excluding IS-IS routers whose LSPs have the OL- 516 bit set. This effectively removes the IS-IS router about to undergo 517 maintenance from the topology, thus preventing it from receiving any 518 transit traffic during the maintenance period. 520 After the maintenance activity has completed, the operator resets the 521 IS-IS Overload Bit within the LSPs of the original IS-IS router 522 causing it to flood updated IS-IS LSPs throughout the IS-IS domain. 523 All IS-IS routers recalculate their SPF tree and now include the 524 original IS-IS router in their topology calculations, allowing it to 525 be used for transit traffic again. 527 Isolating an entire IS-IS router from the topology can be especially 528 disruptive due to the displacement of a large volume of traffic 529 through an entire IS-IS router to other, sub-optimal paths, (e.g., 530 those with significantly larger delay). Thus, in the majority of 531 network maintenance scenarios, where only a single link or LAN needs 532 to be augmented to increase its physical capacity or is experiencing 533 an intermittent failure, it is much more common and desirable to 534 gracefully remove just the targeted link or LAN from service, 535 temporarily, so that the least amount of user-data traffic is 536 affected during the link-specific network maintenance. 538 Appendix B. Link Isolation Challenges 540 Before network maintenance events are performed on individual 541 physical links or LANs, operators substantially increase the IS-IS 542 metric simultaneously on both devices attached to the same link or 543 LAN. In doing so, the devices generate new Link State Protocol Data 544 Units (LSPs) that are flooded throughout the network and cause all 545 routers to gradually shift traffic onto alternate paths with very 546 little or no disruption to in-flight communications by applications 547 or end-users. When performed successfully, this allows the operator 548 to confidently perform disruptive augmentation, fault diagnosis or 549 repairs on a link without disturbing ongoing communications in the 550 network. 552 There are a number of challenges with the above solution. First, it 553 is quite common to have routers with several hundred interfaces and 554 individual interfaces that are from several hundred Gigabits/second 555 to Terabits/second of traffic. Thus, it is imperative that operators 556 accurately identify the same point-to-point link on two, separate 557 devices in order to increase (and, afterward, decrease) the IS-IS 558 metric appropriately. Second, the aforementioned solution is very 559 time consuming and even more error-prone to perform when it's 560 necessary to temporarily remove a multi-access LAN from the network 561 topology. Specifically, the operator needs to configure ALL devices 562 that have interfaces attached to the multi-access LAN with an 563 appropriately high IS-IS metric, (and then decrease the IS-IS metric 564 to its original value afterward). Finally, with respect to multi- 565 access LANs, there is currently no method to bidirectionally isolate 566 only a single node's interface on the LAN when performing more fine- 567 grained diagnosis and repairs to the multi-access LAN. 569 In theory, use of a Network Management System (NMS) could improve the 570 accuracy of identifying the appropriate subset of routers attached to 571 either a point-to-point link or a multi-access LAN as well as 572 signaling from the NMS to those devices, using a network management 573 protocol to adjust the IS-IS metrics on the pertinent set of 574 interfaces. The reality is that NMSs are, to a very large extent, 575 not used within Service Provider's networks for a variety of reasons. 576 In particular, NMSs do not interoperate very well across different 577 vendors or even separate platform families within the same vendor. 579 The risks of misidentifying one side of a point-to-point link or one 580 or more interfaces attached to a multi-access LAN and subsequently 581 increasing its IS-IS metric and potentially increased latency, jitter 582 or packet loss. This is unacceptable given the necessary performance 583 requirements for a variety of reasons including the customer 584 perception for near lossless operations and the associated demanding 585 Service Level Agreement's (SLAs) for all network services. 587 Appendix C. Contributors' Addresses 589 Tony Li 591 Email: tony.li@tony.li 593 Authors' Addresses 595 Naiming Shen 596 Cisco Systems 597 560 McCarthy Blvd. 598 Milpitas, CA 95035 599 USA 601 Email: naiming@cisco.com 603 Shane Amante 604 Apple, Inc. 605 1 Infinite Loop 606 Cupertino, CA 95014 607 USA 609 Email: samante@apple.com 610 Mikael Abrahamsson 611 T-Systems Nordic 612 Kistagangen 26 613 Stockholm 614 SE 616 Email: Mikael.Abrahamsson@t-systems.se