idnits 2.17.1 draft-sriram-opsec-urpf-improvements-02.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 abstract seems to contain references ([BCP38], [BCP84]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (October 30, 2017) is 2369 days in the past. Is this intentional? Checking references for intended status: Best Current Practice ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'BCP84' is mentioned on line 19, but not defined == Missing Reference: 'BCP38' is mentioned on line 16, but not defined == Missing Reference: 'AS2 AS1' is mentioned on line 328, but not defined == Missing Reference: 'AS3 AS1' is mentioned on line 396, but not defined == Missing Reference: 'AS1' is mentioned on line 405, but not defined == Missing Reference: 'AS5 AS1' is mentioned on line 323, but not defined == Unused Reference: 'RFC6811' is defined on line 562, but no explicit reference was found in the text Summary: 1 error (**), 0 flaws (~~), 8 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Opsec Working Group K. Sriram 3 Internet-Draft D. Montgomery 4 Intended status: Best Current Practice US NIST 5 Expires: May 3, 2018 J. Haas 6 Juniper Networks, Inc. 7 October 30, 2017 9 Enhanced Feasible-Path Unicast Reverse Path Filtering 10 draft-sriram-opsec-urpf-improvements-02 12 Abstract 14 This document identifies a need for improvement of the unicast 15 Reverse Path Filtering techniques (uRPF) [BCP84] for source address 16 validation (SAV) [BCP38]. The strict uRPF is inflexible about 17 directionality, the loose uRPF is oblivious to directionality, and 18 the current feasible-path uRPF attempts to strike a balance between 19 the two [BCP84]. However, as shown in this draft, the existing 20 feasible-path uRPF still has short comings. This document proposes 21 an enhanced feasible-path uRPF technique, which aims to be more 22 flexible (in a meaningful way) about directionality than the 23 feasible-path uRPF. It can potentially alleviate ISPs' concerns 24 about the possibility of disrupting service for their customers, and 25 encourage greater deployment of uRPF techniques. 27 Status of This Memo 29 This Internet-Draft is submitted in full conformance with the 30 provisions of BCP 78 and BCP 79. 32 Internet-Drafts are working documents of the Internet Engineering 33 Task Force (IETF). Note that other groups may also distribute 34 working documents as Internet-Drafts. The list of current Internet- 35 Drafts is at https://datatracker.ietf.org/drafts/current/. 37 Internet-Drafts are draft documents valid for a maximum of six months 38 and may be updated, replaced, or obsoleted by other documents at any 39 time. It is inappropriate to use Internet-Drafts as reference 40 material or to cite them other than as "work in progress." 42 This Internet-Draft will expire on May 3, 2018. 44 Copyright Notice 46 Copyright (c) 2017 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents 51 (https://trustee.ietf.org/license-info) in effect on the date of 52 publication of this document. Please review these documents 53 carefully, as they describe your rights and restrictions with respect 54 to this document. Code Components extracted from this document must 55 include Simplified BSD License text as described in Section 4.e of 56 the Trust Legal Provisions and are provided without warranty as 57 described in the Simplified BSD License. 59 Table of Contents 61 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 62 1.1. Requirements Language . . . . . . . . . . . . . . . . . . 3 63 2. Review of Existing Source Address Validation Techniques . . . 3 64 2.1. SAV using Access Control List . . . . . . . . . . . . . . 4 65 2.2. SAV using Strict Unicast Reverse Path Filtering . . . . . 4 66 2.3. SAV using Feasible-Path Unicast Reverse Path Filtering . 5 67 2.4. SAV using Loose Unicast Reverse Path Filtering . . . . . 6 68 3. Proposed New Technique: SAV using Enhanced Feasible-Path uRPF 7 69 3.1. Description of the Method . . . . . . . . . . . . . . . . 7 70 3.2. Operational Recommendations . . . . . . . . . . . . . . . 8 71 3.3. A Challenging Scenario . . . . . . . . . . . . . . . . . 9 72 3.4. Overcoming the Above Challenge: Algorithm with Full 73 Flexibility Across Customer Cone . . . . . . . . . . . . 10 74 3.5. Implementation Considerations . . . . . . . . . . . . . . 11 75 3.5.1. Impact on FIB Memory Size Requirement . . . . . . . . 11 76 4. Security Considerations . . . . . . . . . . . . . . . . . . . 12 77 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 12 78 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 13 79 7. Informative References . . . . . . . . . . . . . . . . . . . 13 80 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 82 1. Introduction 84 This internet draft identifies a need for improvement of the unicast 85 Reverse Path Filtering (uRPF) techniques [RFC2827] for source address 86 validation (SAV) [RFC3704]. The strict uRPF is inflexible about 87 directionality, the loose uRPF is oblivious to directionality, and 88 the current feasible-path uRPF attempts to strike a balance between 89 the two [RFC3704]. However, as shown in this draft, the existing 90 feasible-path uRPF still has short comings. Even with the feasible- 91 path uRPF, ISPs are often apprehensive that they may be dropping 92 customers' data packets with legitimate source addresses. 94 This document proposes an enhanced feasible-path uRPF technique, 95 which aims to be more flexible (in a meaningful way) about 96 directionality than the feasible-path uRPF. It is based on the 97 principle that if BGP updates for multiple prefixes with the same 98 origin AS were received on different interfaces (at border routers), 99 then incoming data packets with source addresses in any of those 100 prefixes should be accepted on any of those interfaces (described in 101 Section 3.1). For some challenging ISP-customer scenarios (see 102 Section 3.3), we further propose (a) Forming a list of all unique 103 prefixes in the collection of routes received on all customer 104 interfaces; and (b) Including that list in the RPF list of each 105 customer interface (described in Section 3.4). Implementation 106 considerations are discussed in Section 3.5. 108 Note: Definition of Reverse Path Filtering (RPF) list: The list of 109 permissible source address prefixes for incoming data packets on a 110 given interface. 112 The proposed techniques are expected to add greater operational 113 robustness and efficacy to uRPF, while minimizing ISPs' concerns 114 about accidental service disruption for their customers. It is 115 expected that this will encourage more deployment of uRPF so as to 116 realize its DDoS prevention benefits network wide. 118 1.1. Requirements Language 120 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 121 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 122 document are to be interpreted as described in RFC 2119 [RFC2119]. 124 2. Review of Existing Source Address Validation Techniques 126 There are various existing techniques for mitigation against DDoS 127 attacks with spoofed addresses [RFC2827] [RFC3704]. There are also 128 some techniques used for mitigating reflection attacks [RRL] 129 [TA14-017A], which are used to amplify the impact in DDoS attacks. 130 Employing a combination of these preventive techniques in enterprise 131 and ISP border routers, DNS servers, broadband and wireless access 132 networks, and data centers provides reasonably effective protection 133 against DDoS attacks. 135 Source address validation (SAV) is performed in network edge devices 136 such as border routers, Cable Modem Termination Systems (CMTS), 137 Digital Subscriber Line Access Multiplexers (DSLAM), and Packet Data 138 Network (PDN) gateways in mobile networks. Ingress Access Control 139 List (ACL) and unicast Reverse Path Filtering (uRPF) are techniques 140 employed for implementing SAV [RFC2827] [RFC3704] [ISOC]. 142 2.1. SAV using Access Control List 144 Ingress/egress Access Control Lists (ACLs) are maintained which list 145 acceptable (or alternatively, unacceptable) prefixes for the source 146 addresses in the incoming Internet Protocol (IP) packets. Any packet 147 with a source address that does not match the filter is dropped. The 148 ACLs for the ingress/egress filters need to be maintained to keep 149 them up to date. Updating the ACLs is an operator driven manual 150 process, and hence operationally difficult or infeasible. 152 Typically, the egress ACLs in access aggregation devices (e.g. CMTS, 153 DSLAM) permit source addresses only from the address spaces 154 (prefixes) that are associated with the interface on which the 155 customer network is connected. Ingress ACLs are typically deployed 156 on border routers, and drop ingress packets when the source address 157 is spoofed (i.e. belongs to obviously disallowed prefix blocks, RFC 158 1918 prefixes, or provider's own prefixes). 160 2.2. SAV using Strict Unicast Reverse Path Filtering 162 In the strict unicast Reverse Path Filtering (uRPF) method, an 163 ingress packet at border router is accepted only if the Forwarding 164 Information Base (FIB) contains a prefix that encompasses the source 165 address and forwarding information for that destination prefix points 166 back to the interface over which the packet was received. In other 167 words, the reverse path for routing to that source address (if it 168 were used as a destination address) should use the same interface 169 over which the packet was received. It is well known that this 170 method has limitations when networks are multi-homed and there is 171 asymmetric routing of packets. Asymmetric routing occurs (see 172 Figure 1) when a customer AS announces one prefix (P1) to one transit 173 provider (ISP-a) and a different prefix (P2) to another transit 174 provider (ISP-b), but routes data packets with source addresses in 175 the second prefix (P2) to the first transit provider (ISP-a) or vice 176 versa. 178 +------------+ ---- P1[AS2 AS1] ---> +------------+ 179 | AS2(ISP-a) | <----P2[AS3 AS1] ---- | AS3(ISP-b)| 180 +------------+ +------------+ 181 /\ /\ 182 \ / 183 \ / 184 \ / 185 P1[AS1]\ /P2[AS1] 186 \ / 187 +-----------------------+ 188 | AS1(customer) | 189 +-----------------------+ 190 P1, P2 (prefixes originated) 192 Consider data packets received at AS2 193 (1) from AS1 with source address in P2, or 194 (2) from AS3 that originated from AS1 195 with source address in P1: 196 * Strict uRPF fails 197 * Feasible-path uRPF fails 198 * Loose uRPF works (but ineffective in IPv4) 199 * Enhanced Feasible-path uRPF works best 201 Figure 1: Scenario 1 for illustration of efficacy of uRPF schemes. 203 2.3. SAV using Feasible-Path Unicast Reverse Path Filtering 205 The feasible-path uRPF helps partially overcome the problem 206 identified with the strict uRPF in the multi-homing case. The 207 feasible-path uRPF is similar to the strict uRPF, but in addition to 208 inserting the best-path prefix, additional prefixes from alternative 209 announced routes are also included in the RPF table. This method 210 relies on announcements for the same prefixes (albeit some may be 211 prepended to effect lower preference) propagating to all routers 212 performing feasible-path uRPF checks. Therefore, in the multi-homing 213 scenario, if the customer AS announces routes for both prefixes (P1, 214 P2) to both transit providers (with suitable prepends if needed for 215 traffic engineering), then the feasible-path uRPF method works (see 216 Figure 2). It should be mentioned that the feasible-path uRPF works 217 in this scenario only if customer routes are preferred at AS2 and AS3 218 over a shorter non-customer route. 220 +------------+ routes for P1, P2 +-----------+ 221 | AS2(ISP-a) |<-------------------->| AS3(ISP-b)| 222 +------------+ (p2p) +-----------+ 223 /\ /\ 224 \ / 225 P1[AS1]\ /P2[AS1] 226 \ / 227 P2[AS1 AS1 AS1]\ /P1[AS1 AS1 AS1] 228 \ / 229 +-----------------------+ 230 | AS1(customer) | 231 +-----------------------+ 232 P1, P2 (prefixes originated) 234 Consider data packets received at AS2 via AS3 235 that originated from AS1 and have source address in P1: 236 * Feasible-path uRPF works (if customer route to P1 237 is preferred at AS3 over shorter path) 238 * Feasible-path uRPF fails (if shorter path to P1 239 is preferred at AS3 over customer route) 240 * Loose uRPF works (but ineffective in IPv4) 241 * Enhanced Feasible-path uRPF works best 243 Figure 2: Scenario 2 for illustration of efficacy of uRPF schemes. 245 However, the feasible-path uRPF method has limitations as well. One 246 form of limitation naturally occurs when the recommendation of 247 propagating the same prefixes to all routers is not followed. 248 Another form of limitation can be described as follows. In Scenario 249 2 (described above, illustrated in Figure 2), it is possible that the 250 second transit provider (ISP-b or AS3) does not propagate the 251 prepended route for prefix P1 to the first transit provider (ISP-a or 252 AS2). This is because AS3's decision policy permits giving priority 253 to a shorter route to prefix P1 via a peer (AS2) over a longer route 254 learned directly from the customer (AS1). In such a scenario, AS3 255 would not send any route announcement for prefix P1 to AS2. Then a 256 data packet with source address in prefix P1 that originates from AS1 257 and traverses via AS3 to AS2 will get dropped at AS2. 259 2.4. SAV using Loose Unicast Reverse Path Filtering 261 In the loose unicast Reverse Path Filtering (uRPF) method, an ingress 262 packet at the border router is accepted only if the FIB has one or 263 more prefixes that encompass the source address. That is, a packet 264 is dropped if no route exists in the FIB for the source address. 265 Loose uRPF sacrifices directionality. This method is not effective 266 for prevention of address spoofing since there is little unrouted 267 address space in IPv4. It only drops packets if the spoofed address 268 is unreachable in the current FIB (e.g. RFC 1918, unallocated, 269 allocated but currently not routed). 271 3. Proposed New Technique: SAV using Enhanced Feasible-Path uRPF 273 3.1. Description of the Method 275 Enhanced feasible-path uRPF adds greater operational robustness and 276 efficacy to existing uRPF methods discussed in Section 2. The 277 proposed technique is based on the principle that if BGP updates for 278 multiple prefixes with the same origin AS were received on different 279 interfaces (at border routers), then incoming data packets with 280 source addresses in any of those prefixes should be accepted on any 281 of those interfaces. It can be best explained with an example as 282 follows: 284 Let us say, a border router of ISP-A has in its Adj-RIB-in the set of 285 prefixes {Q1, Q2, Q3} each of which has AS-x as its origin and AS-x 286 is in ISP-A's customer cone. Further, the border router received a 287 route for prefix Q1 over a customer facing interface, while it 288 learned routes for prefixes Q2 and Q3 from a lateral peer and an 289 upstream transit provider, respectively. All these routes passed 290 route filtering and/or origin validation (i.e. the origin AS-x is 291 deemed legitimate). That is, the route announcements are considered 292 legitimate. In this example scenario, the enhanced feasible-path 293 uRPF method allows source addresses to belong in {Q1, Q2, Q3} on any 294 of the three specific interfaces in question (customer, peer, 295 provider) on which the three routes were learned. 297 Thus, enhanced feasible-path uRPF defines feasible paths in a more 298 generalized but precise way (as compared to feasible-path uRPF). In 299 the above example, routes for prefixes Q2 and Q3 were not received on 300 a customer facing interface at the border router, yet data packets 301 with source addresses in Q2 or Q3 are accepted by the router if they 302 come in on the same customer interface on which the route for prefix 303 Q1 was received (based on these prefix routes having the same origin 304 AS). 306 Looking back at Scenarios 1 and 2 (Figure 1 and Figure 2), the 307 enhanced feasible-path uRPF provides comparable or better performance 308 than the other uRPF methods. Scenario 3 (Figure 3) further 309 illustrates the enhanced feasible-path uRPF method with a more 310 concrete example. In this scenario, the focus is on operation of the 311 feasible-path uRPF at ISP4 (AS4). ISP4 learns a route for prefix P1 312 via a customer-to-provider (C2P) interface from customer ISP2 (AS2). 313 This route for P1 has origin AS1. ISP4 also learns a route for P2 314 via another C2P interface from customer ISP3 (AS3). Additionally, 315 AS4 learns an alternate route for P2 via a peer-to-peer (p2p) 316 interface from ISP5 (AS5). Both routes for P2 have the same origin 317 AS (i.e. AS1) as does the route for P1. Using the proposed enhanced 318 feasible-path uRPF scheme, given the commonality of the origin AS 319 across the above-mentioned routes for P1 and P2, AS4 would permit 320 source addresses belonging to either P1 or P2 in data packets 321 received on any of the three interfaces (from AS2, AS3, and AS5). 323 +----------+ P2[AS5 AS1] +------------+ 324 | AS4(ISP4)|<---------------| AS5(ISP5) | 325 +----------+ (p2p) +------------+ 326 /\ /\ /\ 327 / \ / 328 P1[AS2 AS1]/ \P2[AS3 AS1] / 329 (C2P)/ \(C2P) / 330 / \ / 331 +----------+ +----------+ / 332 | AS2(ISP2)| | AS3(ISP3)| / 333 +----------+ +----------+ / 334 /\ /\ / 335 \ / / 336 P1[AS1]\ /P2[AS1] /P2[AS1] 337 (C2P)\ /(C2P) /(C2P) 338 \ / / 339 +----------------+ / 340 | AS1(customer) |/ 341 +----------------+ 342 P1, P2 (prefixes originated) 344 Consider that data packets (sourced from AS1) 345 may be received at AS4 with source address 346 in P1 or P2 via any of the neighbors (AS2, AS3, AS5): 347 * Feasible-path uRPF fails 348 * Loose uRPF works (but not desirable) 349 * Enhanced Feasible-path uRPF works best 351 Figure 3: Scenario 3 for illustration of efficacy of uRPF schemes. 353 Based on the above, the proposed enhanced feasible-path uRPF method 354 would reduce ISP concerns about possible service disruption affecting 355 their customers and encourage greater adoption of uRPF. 357 3.2. Operational Recommendations 359 The following operational recommendations will make the operation of 360 the proposed enhanced feasible-path uRPF robust: 362 For multi-homed stub AS: 364 o A multi-homed stub AS SHOULD announce at least one of the prefixes 365 it originates to each of its transit provider ASes. 367 For non-stub AS: 369 o A non-stub AS SHOULD also announce at least one of the prefixes it 370 originates to each of its transit provider ASes. 372 o Additionally, from the routes it has learned from customers, a 373 non-stub AS SHOULD announce at least one route per origin AS to 374 each of its transit provider ASes. 376 (Note: It is worth noting that in the above recommendations if "at 377 least one" is replaced with "all", then even traditional feasible- 378 path uRPF will work as desired.) 380 3.3. A Challenging Scenario 382 It should be observed that in the absence of ASes adhering the above 383 recommendations, the following example scenarios may be constructed 384 which pose a challenge for the enhanced feasible-path uRPF (as well 385 as for traditional feasible-path uRPF). In the scenario illustrated 386 in Figure 4, since routes for neither P1 nor P2 are propagated on the 387 AS2-AS4 interface, the enhanced feasible-path uRPF at AS4 will reject 388 data packets received on that interface with source addresses in P1 389 or P2. 391 +----------+ 392 | AS4(ISP4)| 393 +----------+ 394 /\ /\ 395 / \ P1[AS3 AS1] 396 P1 and P2 not / \ P2[AS3 AS1] 397 propagated / \ (C2P) 398 (C2P) / \ 399 +----------+ +----------+ 400 | AS2(ISP2)| | AS3(ISP3)| 401 +----------+ +----------+ 402 /\ /\ 403 \ / P1[AS1] 404 P1[AS1] NO_EXPORT \ / P2[AS1] 405 P2[AS1] NO_EXPORT \ / (C2P) 406 (C2P) \ / 407 +----------------+ 408 | AS1(customer) | 409 +----------------+ 410 P1, P2 (prefixes originated) 412 Figure 4: Illustration of a challenging scenario. 414 3.4. Overcoming the Above Challenge: Algorithm with Full Flexibility 415 Across Customer Cone 417 Adding further flexibility to the enhanced feasible-path uRPF method 418 can help address the potential limitation identified above using the 419 scenario in Figure 4 (Section 3.3). In the following, "route" refers 420 to a route currently existing in the Adj-RIB-in. Including the 421 additional degree of flexibility, the modified algorithm can be 422 described as follows: 424 o Let I = {I1, I2, ..., In} represent the set of all directly- 425 connected customer interfaces at customer-facing edge routers in a 426 transit provider's AS. 428 o Let P = {P1, P2, ..., Pm} represent the set of all unique prefixes 429 for which routes were received over the interfaces in Set I. 431 o Let A = {AS1, AS2, ..., ASk} represent the set of all unique 432 origin ASes seen in the routes that were received over the 433 interfaces in Set I. 435 o Let Q = {Q1, Q2, ..., Qj} represent the set of all unique prefixes 436 for which routes were received over peer or provider interfaces 437 such that each of the routes has its origin AS belonging in Set A. 439 o Then, Set Z = Union(P,Q) represents the RPF list for each 440 customer-facing edge router in the AS in question. That is, over 441 each interface in Set I, the edge router SHOULD permit only those 442 ingress data packets that have SA in any of the prefixes in Set Z. 444 When this algorithmic flexibility is incorporated, then the type of 445 limitation identified in Figure 4 (Section 3.3) goes away. This 446 should significantly reduce the possibility of blocking legitimate 447 customer-data packets in uRPF implementations. 449 3.5. Implementation Considerations 451 The existing RPF checks in edge routers take advantage of existing 452 line card implementations to perform the RPF functions. For 453 implementation of the proposed technique, the general necessary 454 feature would be to extend the line cards to take arbitrary RPF lists 455 that are not necessarily the same as the existing FIB contents. For 456 example, in the proposed method, the RPF lists are constructed by 457 applying a set of rules to all received BGP routes (not just those 458 selected as best path and installed in FIB). 460 3.5.1. Impact on FIB Memory Size Requirement 462 The proposed technique requires that there should be FIB memory 463 (i.e., TCAM) available to store the RPF lists in line cards. For an 464 ISP's AS, the RPF list size for each line card will roughly and 465 conservatively equal the total number of prefixes in its customer 466 cone (assuming the algorithm in Section 3.4 is used). The following 467 table shows the measured customer cone sizes for various types of 468 ISPs [sriram-ripe63]: 470 +---------------------------------+---------------------------------+ 471 | Type of ISP | Measured Customer Cone Size in | 472 | | # Prefixes (in turn this is an | 473 | | estimate for RPF list size on | 474 | | line card) | 475 +---------------------------------+---------------------------------+ 476 | Very Large Global ISP | 32392 | 477 | ------------------------------- | ------------------------------- | 478 | Very Large Global ISP | 29528 | 479 | ------------------------------- | ------------------------------- | 480 | Large Global ISP | 20038 | 481 | ------------------------------- | ------------------------------- | 482 | Mid-size Global ISP | 8661 | 483 | ------------------------------- | ------------------------------- | 484 | Regional ISP (in Asia) | 1101 | 485 +---------------------------------+---------------------------------+ 487 Table 1: Customer cone sizes (# prefixes) for various types of ISPs. 489 For some super large global ISPs that are at the core of the 490 Internet, the customer cone size (# prefixes) can be as high as a few 491 hundred thousand [caida]. But uRPF is most effective when deployed 492 at ASes at the edges of the Internet where the customer cone sizes 493 are smaller as shown in Table 1. 495 A very large global ISP's router line card is likely to have a FIB 496 size large enough to accommodate 2 to 6 million routes [cisco1]. 497 Similarly, the line cards in routers corresponding to a large global 498 ISP, a mid-size global ISP, and a regional ISP are likely to have FIB 499 sizes large enough to accommodate about 1 million, 0.5 million, and 500 100K routes, respectively [cisco2]. Comparing these FIB size numbers 501 with the corresponding RPF list size numbers in Table 1, it can be 502 surmised that the conservatively estimated RPF list size is only a 503 small fraction of the anticipated FIB memory size under various ISP 504 scenarios. 506 4. Security Considerations 508 This document offers a technique to improve the robustness features 509 of uRPF and thus improve the security of the Internet as a whole. 510 The proposed technique does not warrant any additional security 511 considerations. 513 5. IANA Considerations 515 This document does not request new capabilities or attributes. It 516 does not create any new IANA registries. 518 6. Acknowledgements 520 The authors would like to thank Job Snijders, Marco Marzetti, Marco 521 d'Itri, Nick Hilliard, Gert Doering, Igor Gashinsky, Barry Greene, 522 and Joel Jaeggli for comments and suggestions. 524 7. Informative References 526 [caida] "Information for AS 174 (COGENT-174)", CAIDA Spoofer 527 Project , . 529 [cisco1] "Internet Routing Table Growth Causes ROUTING-FIB- 530 4-RSRC_LOW Message on Trident-Based Line Cards", Cisco 531 Trouble-shooting Tech-notes , January 2014, 532 . 536 [cisco2] "Cisco Nexus 7000 Series NX-OS Unicast Routing 537 Configuration Guide, Release 5.x (Chapter: Managing the 538 Unicast RIB and FIB)", Cisco Configuration Guides , June 539 2017, . 544 [ISOC] Vixie (Ed.), P., "Addressing the challenge of IP 545 spoofing", ISOC report , September 2015, 546 . 548 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 549 Requirement Levels", BCP 14, RFC 2119, 550 DOI 10.17487/RFC2119, March 1997, 551 . 553 [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: 554 Defeating Denial of Service Attacks which employ IP Source 555 Address Spoofing", BCP 38, RFC 2827, DOI 10.17487/RFC2827, 556 May 2000, . 558 [RFC3704] Baker, F. and P. Savola, "Ingress Filtering for Multihomed 559 Networks", BCP 84, RFC 3704, DOI 10.17487/RFC3704, March 560 2004, . 562 [RFC6811] Mohapatra, P., Scudder, J., Ward, D., Bush, R., and R. 563 Austein, "BGP Prefix Origin Validation", RFC 6811, 564 DOI 10.17487/RFC6811, January 2013, 565 . 567 [RRL] "Response Rate Limiting in the Domain Name System", 568 Redbarn blog , . 570 [sriram-ripe63] 571 Sriram, K. and R. Bush, "Estimating CPU Cost of BGPSEC on 572 a Router", Presented at RIPE-63; also at IETF-83 SIDR WG 573 Meeting, March 2012, 574 . 577 [TA14-017A] 578 "UDP-Based Amplification Attacks", US-CERT alert 579 TA14-017A , January 2014, 580 . 582 Authors' Addresses 584 Kotikalapudi Sriram 585 US NIST 586 100 Bureau Drive 587 Gaithersburg MD 20899 588 USA 590 Email: ksriram@nist.gov 592 Doug Montgomery 593 US NIST 594 100 Bureau Drive 595 Gaithersburg MD 20899 596 USA 598 Email: dougm@nist.gov 600 Jeffrey Haas 601 Juniper Networks, Inc. 602 1133 Innovation Way 603 Sunnyvale CA 94089 604 USA 606 Email: jhaas@juniper.net