idnits 2.17.1 draft-gont-v6ops-ipv6-ehs-packet-drops-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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 4, 2016) is 3003 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC2460' is defined on line 465, but no explicit reference was found in the text == Unused Reference: 'RIPE-Atlas' is defined on line 643, but no explicit reference was found in the text ** Obsolete normative reference: RFC 2460 (Obsoleted by RFC 8200) ** Obsolete normative reference: RFC 6145 (Obsoleted by RFC 7915) == Outdated reference: A later version (-08) exists of draft-ietf-6man-deprecate-atomfrag-generation-05 == Outdated reference: A later version (-13) exists of draft-ietf-6man-rfc2460bis-03 -- Obsolete informational reference (is this intentional?): RFC 5575 (Obsoleted by RFC 8955) Summary: 2 errors (**), 0 flaws (~~), 5 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 IPv6 Operations Working Group (v6ops) F. Gont 3 Internet-Draft SI6 Networks / UTN-FRH 4 Intended status: Informational N. Hilliard 5 Expires: August 7, 2016 INEX 6 G. Doering 7 SpaceNet AG 8 W. Liu 9 Huawei Technologies 10 W. Kumari 11 Google 12 February 4, 2016 14 Operational Implications of IPv6 Packets with Extension Headers 15 draft-gont-v6ops-ipv6-ehs-packet-drops-02 17 Abstract 19 This document summarizes the security and operational implications of 20 IPv6 extension headers, and attempts to analyze reasons why packets 21 with IPv6 extension headers may be dropped in the public Internet. 23 Status of This Memo 25 This Internet-Draft is submitted in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at http://datatracker.ietf.org/drafts/current/. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 This Internet-Draft will expire on August 7, 2016. 40 Copyright Notice 42 Copyright (c) 2016 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (http://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 58 2. Previous Work on IPv6 Extension Headers . . . . . . . . . . . 3 59 3. Security Implications . . . . . . . . . . . . . . . . . . . . 3 60 4. Operational Implications . . . . . . . . . . . . . . . . . . 5 61 4.1. Requirement to process required layer-3/layer-4 62 information . . . . . . . . . . . . . . . . . . . . . . . 5 63 4.2. Route-Processor Protection . . . . . . . . . . . . . . . 7 64 4.3. Inability to Perform Fine-grained Filtering . . . . . . . 8 65 5. A Possible Attack Vector . . . . . . . . . . . . . . . . . . 8 66 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 10 67 7. Security Considerations . . . . . . . . . . . . . . . . . . . 10 68 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 69 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 70 9.1. Normative References . . . . . . . . . . . . . . . . . . 10 71 9.2. Informative References . . . . . . . . . . . . . . . . . 11 72 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 74 1. Introduction 76 IPv6 Extension Headers (EHs) allow for the extension of the IPv6 77 protocol, and provide support for core functionality such as IPv6 78 fragmentation. However, common implementation limitations suggest 79 that EHs present a challenge for IPv6 packet routing equipment, and 80 evidence exists to suggest that IPv6 packets with EHs may be 81 intentionally dropped on the public Internet in some network 82 deployments. 84 The authors of this document have been involved in numerous 85 discussions about IPv6 extension headers (both within the IETF and 86 outside of it), and have noticed that a number of security and 87 operational issues were unknown to the larger audience participating 88 in these discussions. 90 This document has the following goals: 92 o Raise awareness about the security and operational implications of 93 IPv6 Extension Headers, and presents reason why some networks 94 intentionally drop packets containing IPv6 Extension Headers. 96 o Highlight areas where current IPv6 support by networking devices 97 maybe sub-optimal, such that the aforementioned support is 98 improved. 100 o Highlight operational issues associated with IPv6 extension 101 headers, such that those issues are considered in IETF 102 standardization efforts. 104 Section 2 of this document summarizes the previous work that has been 105 done in the area of IPv6 extension headers. Section 3 briefly 106 discusses the security implications of IPv6 Extension Headers, while 107 Section 4 discusses their operational implications. 109 2. Previous Work on IPv6 Extension Headers 111 Some of the implications of IPv6 Extension Headers have been 112 discussed in IETF circles. For example, [I-D.taylor-v6ops-fragdrop] 113 discusses a rationale for which operators drop IPv6 fragments. 114 [I-D.wkumari-long-headers] discusses possible issues arising from 115 "long" IPv6 header chains. [RFC7045] clarifies how intermediate 116 nodes should deal with IPv6 extension headers. [RFC7112] discusses 117 the issues arising in a specific fragmentation case where the IPv6 118 header chain is fragmented into two or more fragments (and formally 119 forbids such fragmentation case). 120 [I-D.kampanakis-6man-ipv6-eh-parsing] describes how inconsistencies 121 in the way IPv6 packets with extension headers are parsed by 122 different implementations may result in evasion of security controls, 123 and presents guidelines for parsing IPv6 extension headers with the 124 goal of providing a common and consistent parsing methodology for 125 IPv6 implementations. [RFC6980] analyzes the security implications 126 of employing IPv6 fragmentation with Neighbor Discovery for IPv6, and 127 formally recommends against such usage. Finally, [RFC7123] discusses 128 how some popular RA-Guard implementations are subject to evasion by 129 means of IPv6 extension headers. 131 Some preliminary measurements regarding the extent to which packet 132 containing IPv6 EHs are dropped in the public Internet have been 133 presented in [PMTUD-Blackholes], [Gont-IEPG88], [Gont-Chown-IEPG89], 134 and [Linkova-Gont-IEPG90]. [I-D.ietf-v6ops-ipv6-ehs-in-real-world] 135 presents more comprehensive results and documents the methodology for 136 obtaining the presented results. 138 3. Security Implications 140 The security implications of IPv6 Extension Headers generally fall 141 into one or more of these categories: 143 o Evasion of security controls 144 o DoS due to processing requirements 146 o DoS due to implementation errors 148 o Extension Header-specific issues 150 Unlike IPv4 packets where the upper-layer protocols can be trivially 151 found by means of the "IHL" ("Internet Header Length") IPv4 header 152 field, the structure of IPv6 packets is more flexible and complex. 153 Locating upper-layer protocol information requires that all IPv6 154 extension headers be examined. This has presented implementation 155 difficulties, and packet filtering mechanisms that require upper- 156 layer information (even if just the upper layer protocol type) on 157 several security devices can be trivially evaded by inserting IPv6 158 Extension Headers between the main IPv6 header and the upper layer 159 protocol. [RFC7113] describes this issue for the RA-Guard case, but 160 the same techniques can be employed to circumvent other IPv6 firewall 161 and packet filtering mechanisms. Additionally, implementation 162 inconsistencies in packet forwarding engines may result in evasion of 163 security controls [I-D.kampanakis-6man-ipv6-eh-parsing] [Atlasis2014] 164 [BH-EU-2014]. 166 Packets that use IPv6 Extension Headers may have a negative 167 performance impact on the handling devices. Unless appropriate 168 mitigations are put in place (e.g., packet dropping and/or rate- 169 limiting), an attacker could simply send a large amount of IPv6 170 traffic employing IPv6 Extension Headers with the purpose of 171 performing a Denial of Service (DoS) attack (see Section 4 for 172 further details). 174 NOTE: 175 In the most trivial case, a packet that includes a Hop-by-Hop 176 Options header will typically go through the slow forwarding path, 177 and be processed by the router's CPU. Another possible case might 178 be that in which a router that has been configured to enforce an 179 ACL based on upper-layer information (e.g., upper layer protocol 180 or TCP Destination Port), needs to process the entire IPv6 header 181 chain (in order to find the required information) and this causes 182 the packet to be processed in the slow path [Cisco-EH-Cons]. We 183 note that, for obvious reasons, the aforementioned performance 184 issues may also affect other devices such as firewalls, Network 185 Intrusion Detection Systems (NIDS), etc. [Zack-FW-Benchmark]. 186 The extent to which these devices are affected will typically be 187 implementation-dependent. 189 IPv6 implementations, like all other software, tend to mature with 190 time and wide-scale deployment. While the IPv6 protocol itself has 191 existed for almost 20 years, serious bugs related to IPv6 Extension 192 Header processing continue to be discovered. Because there is 193 currently little operational reliance on IPv6 Extension headers, the 194 corresponding code paths are rarely exercised, and there is the 195 potential that bugs still remain to be discovered in some 196 implementations. 198 IPv6 Fragment Headers are employed to allow fragmentation of IPv6 199 packets. While many of the security implications of the 200 fragmentation / reassembly mechanism are known from the IPv4 world, 201 several related issues have crept into IPv6 implementations. These 202 range from denial of service attacks to information leakage, for 203 example [I-D.ietf-6man-predictable-fragment-id], [Bonica-NANOG58] and 204 [Atlasis2012]). 206 4. Operational Implications 208 4.1. Requirement to process required layer-3/layer-4 information 210 Intermediate systems and middleboxes that need to find the layer-4 211 header must process the entire IPv6 extension header chain. When 212 such devices are unable to obtain the required information, they may 213 simply drop the corresponding packets. The following subsections 214 discuss some of reasons for which such layer-4 information may be 215 needed by an intermediate systems or middlebox, and why packets 216 containing IPv6 extension headers may represent a challenge in such 217 scenarios. 219 4.1.1. Packet Forwarding Engine Constraints 221 Most modern routers use dedicated hardware (e.g. ASICs or NPUs) to 222 determine how to forward packets across their internal fabrics (see 223 [IEPG94-Scudder] for details). One of the common methods of handling 224 next-hop lookup is to send a small portion of the ingress packet to a 225 lookup engine with specialised hardware (e.g. ternary CAM or RLDRAM) 226 to determine the packet's next-hop. Technical constraints mean that 227 there is a trade-off between the amount of data sent to the lookup 228 engine and the overall performance of the lookup engine. If more 229 data is sent, the lookup engine can inspect further into the packet, 230 but the overall performance of the system will be reduced. If less 231 data is sent, the overall performance of the router will be increased 232 but the packet lookup engine may not be able to inspect far enough 233 into a packet to determine how it should be handled. 235 NOTE: 236 For example, current high-end routers at the time of authorship of 237 this document can use up to 192 bytes of header (Cisco ASR9000 238 Typhoon) or 384 bytes of header (Juniper MX Trio) 240 If a hardware forwarding engine on a modern router cannot make a 241 forwarding decision about a packet because critical information is 242 not sent to the look-up engine, then the router will normally drop 243 the packet. Historically, some packet forwarding engines punted 244 packets of this form to the control plane for more in-depth analysis, 245 but this is unfeasible on most current router architectures as a 246 result of the vast difference between the hardware forwarding 247 capacity of the router and processing capacity of the control plane 248 and the size of the management link which connects the control plane 249 to the forwarding plane. 251 If an IPv6 header chain is sufficiently long that its header exceeds 252 the packet look-up capacity of the router, then it may be dropped due 253 to hardware inability to determine how it should be handled. 255 4.1.2. ECMP and Hash-based Load-Sharing 257 In the case of ECMP (equal cost multi path) load sharing, the router 258 on the sending side of the link needs to make a decision regarding 259 which of the links to use for a given packet. Since round-robin 260 usage of the links is usually avoided in order to prevent packet 261 reordering, forwarding engines need to use a mechanism which will 262 consistently forward the same data streams down the same forwarding 263 paths. Most forwarding engines achieve this by calculating a simple 264 hash using an n-tuple gleaned from a combination of layer-2 through 265 to layer-4 packet header information. This n-tuple will typically 266 use the src/dst MAC address, src/dst IP address, and if possible 267 further layer-4 src/dst port information. As layer-4 port 268 information increases the entropy of the hash, it is highly desirable 269 to use it where possible. 271 We note that in the IPv6 world, flows are expected to be identified 272 by means of the IPv6 Flow Label [RFC6437]. Thus, ECMP and Hash-based 273 Load-Sharing would be possible without the need to process the entire 274 IPv6 header chain to obtain upper-layer information to identify 275 flows. However, we note that for a long time many IPv6 276 implementations failed to set the Flow Label, and ECMP and Hash-based 277 Load-Sharing devices also did not employ the Flow Label for 278 performing their task. 280 Clearly, widespread support of [RFC6437] would relieve middle-boxes 281 from having to process the entire IPv6 header chain, making Flow 282 Label-based ECMP and Hash-based Load-Sharing [RFC6438] feasible. 284 4.1.3. Enforcing infrastructure ACLs 286 Generally speaking, infrastructure ACLs (iACLs) drop unwanted packets 287 destined to parts of a provider's infrastructure, because they are 288 not operationally needed and can be used for attacks of different 289 sorts against the router's control plane. Some traffic needs to be 290 differentiated depending on layer-3 or layer-4 criteria to achieve a 291 useful balance of protection and functionality, for example: 293 o Permit some amount of ICMP echo (ping) traffic towards the 294 router's addresses for troubleshooting. 296 o Permit BGP sessions on the shared network of an exchange point 297 (potentially differentiating between the amount of packets/seconds 298 permitted for established sessions and connection establishment), 299 but do not permit other traffic from the same peer IP addresses. 301 4.1.4. DDoS Management and Customer Requests for Filtering 303 The case of customer DDoS protection and edge-to-core customer 304 protection filters is similar in nature to the infrastructure ACL 305 protection. Similar to infrastructure ACL protection, layer-4 ACLs 306 generally need to be applied as close to the edge of the network as 307 possible, even though the intent is usually to protect the customer 308 edge rather than the provider core. Application of layer-4 DDoS 309 protection to a network edge is often automated using Flowspec 310 [RFC5575]. 312 For example, a web site which normally only handled traffic on TCP 313 ports 80 and 443 could be subject to a volumetric DDoS attack using 314 NTP and DNS packets with randomised source IP address, thereby 315 rendering useless traditional [RFC5635] source-based real-time black 316 hole mechanisms. In this situation, DDoS protection ACLs could be 317 configured to block all UDP traffic at the network edge without 318 impairing the web server functionality in any way. Thus, being able 319 to block arbitrary protocols at the network edge can avoid DDoS- 320 related problems both in the provider network and on the customer 321 edge link. 323 4.2. Route-Processor Protection 325 Most modern routers have a fast hardware-assisted forwarding plane 326 and a loosely coupled control plane, connected together with a link 327 that has much less capacity than the forwarding plane could handle. 328 Traffic differentiation cannot be done by the control plane side, 329 because this would overload the internal link connecting the 330 forwarding plane to the control plane. 332 The Hop-by-Hop Options header is particularly challenging since, in 333 most (if not all) implementations, it causes the corresponding packet 334 to be punted to a software path. As a result, operators usually drop 335 IPv6 packets containing this extension header. Please see [RFC6192] 336 for advice regarding protection of the router control plane. 338 4.3. Inability to Perform Fine-grained Filtering 340 Some routers lack of fine-grained filtering of IPv6 extension 341 headers. For example, an operator may want to drop packets 342 containing Routing Header Type 0 (RHT0) but may only be able to 343 filter on the extension header type (Routing Header). As a result, 344 the operator may end up enforcing a more coarse filtering policy 345 (e.g. "drop all packets containing a Routing Header" vs. "only drop 346 packets that contain a Routing Header Type 0"). 348 5. A Possible Attack Vector 350 The widespread drop of IPv6 packets employing IPv6 Extension Headers 351 can, in some scenarios, be exploited for malicious purposes: if 352 packets employing IPv6 EHs are known to be dropped on the path from 353 system A to system B, an attacker could cause packets sent from A to 354 B to be dropped by sending a forged ICMPv6 Packet Too Big (PTB) 355 [RFC4443] error message to A (advertising an MTU smaller than 1280), 356 such that subsequent packets from A to B include a fragment header 357 (i.e., they result in atomic fragments [RFC6946]). 359 Possible scenarios where this attack vector could be exploited 360 include (but are not limited to): 362 o Communication between any two systems through the public network 363 (e.g., client from/to server or server from/to server), where 364 packets with IPv6 extension headers are dropped by some 365 intermediate router 367 o Communication between two BGP peers employing IPv6 transport, 368 where these BGP peers implement ACLs to drop IPv6 fragments (to 369 avoid control-plane attacks) 371 The aforementioned attack vector is exacerbated by the following 372 factors: 374 o The attacker does not need to forge the IPv6 Source Address of his 375 attack packets. Hence, deployment of simple BCP38 filters will 376 not help as a counter-measure. 378 o Only the IPv6 addresses of the IPv6 packet embedded in the ICMPv6 379 payload need to be forged. While one could envision filtering 380 devices enforcing BCP38-style filters on the ICMPv6 payload, the 381 use of extension headers (by the attacker) could make this 382 difficult, if not impossible. 384 o Many implementations fail to perform validation checks on the 385 received ICMPv6 error messages, as recommended in Section 5.2 of 386 [RFC4443] and documented in [RFC5927]. It should be noted that in 387 some cases, such as when an ICMPv6 error message has (supposedly) 388 been elicited by a connection-less transport protocol (or some 389 other connection-less protocol being encapsulated in IPv6), it may 390 be virtually impossible to perform validation checks on the 391 received ICMPv6 error messages. And, because of IPv6 extension 392 headers, the ICMPv6 payload might not even contain any useful 393 information on which to perform validation checks. 395 o Upon receipt of one of the aforementioned ICMPv6 "Packet Too Big" 396 error messages, the Destination Cache [RFC4861] is usually updated 397 to reflect that any subsequent packets to such destination should 398 include a Fragment Header. This means that a single ICMPv6 399 "Packet Too Big" error message might affect multiple communication 400 instances (e.g. TCP connections) with such destination. 402 o A router or other middlebox cannot simply drop all incoming ICMPv6 403 Packet Too Big error messages, as this would create a PMTUD 404 blackhole. 406 Possible mitigations for this issue include: 408 o Dropping incoming ICMPv6 Packet Too Big error messages that 409 advertise an MTU smaller than 1280 bytes. 411 o Artificially reducing the MTU to 1280 bytes and dropping incoming 412 ICMPv6 PTB error messages. 414 Both of these mitigations come at the expense of possibly preventing 415 communication through SIIT [RFC6145], that relies on IPv6 atomic 416 fragments (see [I-D.ietf-6man-deprecate-atomfrag-generation]), and 417 also implies that the filtering device has the ability to filter ICMP 418 PTB messages based on the contents of the MTU field. 420 [I-D.ietf-6man-deprecate-atomfrag-generation] documents while the 421 generation of IPv6 atomic fragments is considered harmful, and 422 documents why this functionality is being removed from the upcoming 423 revision of the core IPv6 protocol [I-D.ietf-6man-rfc2460bis]. Thus, 424 any of the above mitigations would eliminate the attack vector 425 without any interoperability implications. 427 6. IANA Considerations 429 There are no IANA registries within this document. The RFC-Editor 430 can remove this section before publication of this document as an 431 RFC. 433 7. Security Considerations 435 The security implications of IPv6 extension headers are discussed in 436 Section 3. A specific attack vector that could leverage the 437 widespread dropping of packets with IPv6 EHs (along with possible 438 countermeasures) is discussed in Section 5. This document does not 439 introduce any new security issues. 441 8. Acknowledgements 443 The authors would like to thank (in alphabetical order) Mikael 444 Abrahamsson, Brian Carpenter, Sander Steffann, Eric Vyncke, and 445 Andrew Yourtchenko, for providing valuable comments on earlier 446 versions of this document. Additionally, the authors would like to 447 thank participants of the v6ops working group for their valuable 448 input on the topics that led to the publication of this document. 450 Fernando Gont would like to thank Sander Steffann, who took the time 451 to meet to discuss this document, even while higher priority events 452 were in place. 454 Fernando Gont would like to thank Jan Zorz / Go6 Lab 455 , and Jared Mauch / NTT America, for providing 456 access to systems and networks that were employed to perform 457 experiments and measurements involving packets with IPv6 Extension 458 Headers. Additionally, he would like to thank SixXS 459 for providing IPv6 connectivity. 461 9. References 463 9.1. Normative References 465 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 466 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 467 December 1998, . 469 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 470 Control Message Protocol (ICMPv6) for the Internet 471 Protocol Version 6 (IPv6) Specification", RFC 4443, 472 DOI 10.17487/RFC4443, March 2006, 473 . 475 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 476 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 477 DOI 10.17487/RFC4861, September 2007, 478 . 480 [RFC6145] Li, X., Bao, C., and F. Baker, "IP/ICMP Translation 481 Algorithm", RFC 6145, DOI 10.17487/RFC6145, April 2011, 482 . 484 [RFC6946] Gont, F., "Processing of IPv6 "Atomic" Fragments", 485 RFC 6946, DOI 10.17487/RFC6946, May 2013, 486 . 488 9.2. Informative References 490 [Atlasis2012] 491 Atlasis, A., "Attacking IPv6 Implementation Using 492 Fragmentation", BlackHat Europe 2012. Amsterdam, 493 Netherlands. March 14-16, 2012, 494 . 497 [Atlasis2014] 498 Atlasis, A., "A Novel Way of Abusing IPv6 Extension 499 Headers to Evade IPv6 Security Devices", May 2014, 500 . 503 [BH-EU-2014] 504 Atlasis, A., Rey, E., and R. Schaefer, "Evasion of High- 505 End IDPS Devices at the IPv6 Era", BlackHat Europe 2014, 506 2014, . 509 [Bonica-NANOG58] 510 Bonica, R., "IPv6 Extension Headers in the Real World 511 v2.0", NANOG 58. New Orleans, Louisiana, USA. June 3-5, 512 2013, . 515 [Cisco-EH-Cons] 516 Cisco, , "IPv6 Extension Headers Review and 517 Considerations", October 2006, 518 . 521 [Gont-Chown-IEPG89] 522 Gont, F. and T. Chown, "A Small Update on the Use of IPv6 523 Extension Headers", IEPG 89. London, UK. March 2, 2014, 524 . 527 [Gont-IEPG88] 528 Gont, F., "Fragmentation and Extension header Support in 529 the IPv6 Internet", IEPG 88. Vancouver, BC, Canada. 530 November 13, 2013, . 533 [I-D.ietf-6man-deprecate-atomfrag-generation] 534 Gont, F., LIU, S., and T. Anderson, "Generation of IPv6 535 Atomic Fragments Considered Harmful", draft-ietf-6man- 536 deprecate-atomfrag-generation-05 (work in progress), 537 January 2016. 539 [I-D.ietf-6man-predictable-fragment-id] 540 Gont, F., "Security Implications of Predictable Fragment 541 Identification Values", draft-ietf-6man-predictable- 542 fragment-id-10 (work in progress), October 2015. 544 [I-D.ietf-6man-rfc2460bis] 545 Deering, S. and B. Hinden, "Internet Protocol, Version 6 546 (IPv6) Specification", draft-ietf-6man-rfc2460bis-03 (work 547 in progress), January 2016. 549 [I-D.ietf-v6ops-ipv6-ehs-in-real-world] 550 Gont, F., Linkova, J., Chown, T., and S. LIU, 551 "Observations on the Dropping of Packets with IPv6 552 Extension Headers in the Real World", draft-ietf-v6ops- 553 ipv6-ehs-in-real-world-02 (work in progress), December 554 2015. 556 [I-D.kampanakis-6man-ipv6-eh-parsing] 557 Kampanakis, P., "Implementation Guidelines for parsing 558 IPv6 Extension Headers", draft-kampanakis-6man-ipv6-eh- 559 parsing-01 (work in progress), August 2014. 561 [I-D.taylor-v6ops-fragdrop] 562 Jaeggli, J., Colitti, L., Kumari, W., Vyncke, E., Kaeo, 563 M., and T. Taylor, "Why Operators Filter Fragments and 564 What It Implies", draft-taylor-v6ops-fragdrop-02 (work in 565 progress), December 2013. 567 [I-D.wkumari-long-headers] 568 Kumari, W., Jaeggli, J., Bonica, R., and J. Linkova, 569 "Operational Issues Associated With Long IPv6 Header 570 Chains", draft-wkumari-long-headers-03 (work in progress), 571 June 2015. 573 [IEPG94-Scudder] 574 Petersen, B. and J. Scudder, "Modern Router Architecture 575 for Protocol Designers", IEPG 94. Yokohama, Japan. 576 November 1, 2015, . 579 [Linkova-Gont-IEPG90] 580 Linkova, J. and F. Gont, "IPv6 Extension Headers in the 581 Real World v2.0", IEPG 90. Toronto, ON, Canada. July 20, 582 2014, . 585 [PMTUD-Blackholes] 586 De Boer, M. and J. Bosma, "Discovering Path MTU black 587 holes on the Internet using RIPE Atlas", July 2012, 588 . 591 [RFC5575] Marques, P., Sheth, N., Raszuk, R., Greene, B., Mauch, J., 592 and D. McPherson, "Dissemination of Flow Specification 593 Rules", RFC 5575, DOI 10.17487/RFC5575, August 2009, 594 . 596 [RFC5635] Kumari, W. and D. McPherson, "Remote Triggered Black Hole 597 Filtering with Unicast Reverse Path Forwarding (uRPF)", 598 RFC 5635, DOI 10.17487/RFC5635, August 2009, 599 . 601 [RFC5927] Gont, F., "ICMP Attacks against TCP", RFC 5927, 602 DOI 10.17487/RFC5927, July 2010, 603 . 605 [RFC6192] Dugal, D., Pignataro, C., and R. Dunn, "Protecting the 606 Router Control Plane", RFC 6192, DOI 10.17487/RFC6192, 607 March 2011, . 609 [RFC6437] Amante, S., Carpenter, B., Jiang, S., and J. Rajahalme, 610 "IPv6 Flow Label Specification", RFC 6437, 611 DOI 10.17487/RFC6437, November 2011, 612 . 614 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 615 for Equal Cost Multipath Routing and Link Aggregation in 616 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 617 . 619 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 620 with IPv6 Neighbor Discovery", RFC 6980, 621 DOI 10.17487/RFC6980, August 2013, 622 . 624 [RFC7045] Carpenter, B. and S. Jiang, "Transmission and Processing 625 of IPv6 Extension Headers", RFC 7045, 626 DOI 10.17487/RFC7045, December 2013, 627 . 629 [RFC7112] Gont, F., Manral, V., and R. Bonica, "Implications of 630 Oversized IPv6 Header Chains", RFC 7112, 631 DOI 10.17487/RFC7112, January 2014, 632 . 634 [RFC7113] Gont, F., "Implementation Advice for IPv6 Router 635 Advertisement Guard (RA-Guard)", RFC 7113, 636 DOI 10.17487/RFC7113, February 2014, 637 . 639 [RFC7123] Gont, F. and W. Liu, "Security Implications of IPv6 on 640 IPv4 Networks", RFC 7123, DOI 10.17487/RFC7123, February 641 2014, . 643 [RIPE-Atlas] 644 RIPE, , "RIPE Atlas", . 646 [Zack-FW-Benchmark] 647 Zack, E., "Firewall Security Assessment and Benchmarking 648 IPv6 Firewall Load Tests", IPv6 Hackers Meeting #1, 649 Berlin, Germany. June 30, 2013, 650 . 654 Authors' Addresses 655 Fernando Gont 656 SI6 Networks / UTN-FRH 657 Evaristo Carriego 2644 658 Haedo, Provincia de Buenos Aires 1706 659 Argentina 661 Phone: +54 11 4650 8472 662 Email: fgont@si6networks.com 663 URI: http://www.si6networks.com 665 Nick Hilliard 666 INEX 667 4027 Kingswood Road 668 Dublin 24 669 IE 671 Email: nick@inex.ie 673 Gert Doering 674 SpaceNet AG 675 Joseph-Dollinger-Bogen 14 676 Muenchen D-80807 677 Germany 679 Email: gert@space.net 681 Will (Shucheng) Liu 682 Huawei Technologies 683 Bantian, Longgang District 684 Shenzhen 518129 685 P.R. China 687 Email: liushucheng@huawei.com 689 Warren Kumari 690 Google 691 1600 Amphitheatre Parkway 692 Mountain View, CA 94043 693 US 695 Email: warren@kumari.net