idnits 2.17.1 draft-ietf-dnsop-nsec-aggressiveuse-07.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 has examples using IPv4 documentation addresses according to RFC6890, but does not use any IPv6 documentation addresses. Maybe there should be IPv6 examples, too? Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year (Using the creation date from RFC4035, updated by this document, for RFC5378 checks: 2002-10-29) -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (December 13, 2016) is 2691 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) ** Downref: Normative reference to an Historic RFC: RFC 5074 ** Downref: Normative reference to an Informational RFC: RFC 7129 ** Obsolete normative reference: RFC 7719 (Obsoleted by RFC 8499) Summary: 3 errors (**), 0 flaws (~~), 1 warning (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group K. Fujiwara 3 Internet-Draft JPRS 4 Updates: 4035 (if approved) A. Kato 5 Intended status: Standards Track Keio/WIDE 6 Expires: June 16, 2017 W. Kumari 7 Google 8 December 13, 2016 10 Aggressive use of NSEC/NSEC3 11 draft-ietf-dnsop-nsec-aggressiveuse-07 13 Abstract 15 The DNS relies upon caching to scale; however, the cache lookup 16 generally requires an exact match. This document specifies the use 17 of NSEC/NSEC3 resource records to allow DNSSEC validating resolvers 18 to generate negative answers within a range, and positive answers 19 from wildcards. This increases performance / decreases latency, 20 decreases resource utilization on both authoritative and recursive 21 servers, and also increases privacy. It may also help increase 22 resilience to certain DoS attacks in some circumstances. 24 This document updates RFC4035 by allowing validating resolvers to 25 generate negative answers based upon NSEC/NSEC3 records (and positive 26 answers in the presence of wildcards). 28 [ Ed note: Text inside square brackets ([]) is additional background 29 information, answers to frequently asked questions, general musings, 30 etc. They will be removed before publication.This document is being 31 collaborated on in Github at: https://github.com/wkumari/draft-ietf- 32 dnsop-nsec-aggressiveuse. The most recent version of the document, 33 open issues, etc should all be available here. The authors 34 (gratefully) accept pull requests.] 36 Status of This Memo 38 This Internet-Draft is submitted in full conformance with the 39 provisions of BCP 78 and BCP 79. 41 Internet-Drafts are working documents of the Internet Engineering 42 Task Force (IETF). Note that other groups may also distribute 43 working documents as Internet-Drafts. The list of current Internet- 44 Drafts is at http://datatracker.ietf.org/drafts/current/. 46 Internet-Drafts are draft documents valid for a maximum of six months 47 and may be updated, replaced, or obsoleted by other documents at any 48 time. It is inappropriate to use Internet-Drafts as reference 49 material or to cite them other than as "work in progress." 51 This Internet-Draft will expire on June 16, 2017. 53 Copyright Notice 55 Copyright (c) 2016 IETF Trust and the persons identified as the 56 document authors. All rights reserved. 58 This document is subject to BCP 78 and the IETF Trust's Legal 59 Provisions Relating to IETF Documents 60 (http://trustee.ietf.org/license-info) in effect on the date of 61 publication of this document. Please review these documents 62 carefully, as they describe your rights and restrictions with respect 63 to this document. Code Components extracted from this document must 64 include Simplified BSD License text as described in Section 4.e of 65 the Trust Legal Provisions and are provided without warranty as 66 described in the Simplified BSD License. 68 Table of Contents 70 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 71 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 72 3. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 3 73 4. Background . . . . . . . . . . . . . . . . . . . . . . . . . 4 74 5. Aggressive use of Cache . . . . . . . . . . . . . . . . . . . 6 75 5.1. NSEC . . . . . . . . . . . . . . . . . . . . . . . . . . 6 76 5.2. NSEC3 . . . . . . . . . . . . . . . . . . . . . . . . . . 6 77 5.3. Wildcards . . . . . . . . . . . . . . . . . . . . . . . . 6 78 5.4. Consideration on TTL . . . . . . . . . . . . . . . . . . 7 79 6. Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . 7 80 7. Update to RFC 4035 . . . . . . . . . . . . . . . . . . . . . 8 81 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 82 9. Security Considerations . . . . . . . . . . . . . . . . . . . 9 83 10. Implementation Status . . . . . . . . . . . . . . . . . . . . 9 84 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 9 85 11.1. Change History . . . . . . . . . . . . . . . . . . . . . 10 86 11.1.1. Version draft-fujiwara-dnsop-nsec-aggressiveuse-01 . 13 87 11.1.2. Version draft-fujiwara-dnsop-nsec-aggressiveuse-02 . 13 88 11.1.3. Version draft-fujiwara-dnsop-nsec-aggressiveuse-03 . 13 89 11.2. new section . . . . . . . . . . . . . . . . . . . . . . 13 90 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 13 91 12.1. Normative References . . . . . . . . . . . . . . . . . . 13 92 12.2. Informative References . . . . . . . . . . . . . . . . . 14 93 Appendix A. Detailed implementation notes . . . . . . . . . . . 14 94 Appendix B. Procedure for determining ENT vs NXDOMAN with NSEC . 15 95 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 16 97 1. Introduction 99 A DNS negative cache exists, and is used to cache the fact that an 100 RRset does not exist. This method of negative caching requires exact 101 matching; this leads to unnecessary additional lookups, increases 102 latency, leads to extra resource utilization on both authoritative 103 and recursive servers, and decreases privacy by leaking queries. 105 This document updates RFC 4035 to allow recursive resolvers to use 106 NSEC/NSEC3 resource records to synthesize negative answers from the 107 information they have in the cache. This allows validating resolvers 108 to respond with a negative answer immediately if the name in question 109 falls into a range expressed by a NSEC/NSEC3 resource record already 110 in the cache. It also allows the synthesis of positive answers in 111 the presence of wildcard records. 113 Aggressive Negative Caching was first proposed in Section 6 of DNSSEC 114 Lookaside Validation (DLV) [RFC5074] in order to find covering NSEC 115 records efficiently. 117 [RFC8020], and [I-D.vixie-dnsext-resimprove] proposes first steps to 118 using NXDOMAIN information for more effective caching. This takes 119 this technique further. 121 2. Terminology 123 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 124 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 125 document are to be interpreted as described in RFC 2119 [RFC2119]. 127 Many of the specialized terms used in this document are defined in 128 DNS Terminology [RFC7719]. 130 The key words "Source of Synthesis" in this document are to be 131 interpreted as described in [RFC4592]. 133 3. Problem Statement 135 The DNS negative cache caches negative (non-existent) information, 136 and requires an exact match in most instances [RFC2308]. 138 Assume that the (DNSSEC signed) "example.com" zone contains: 140 albatross.example.com IN A 192.0.2.1 141 elephant.example.com IN A 192.0.2.2 142 zebra.example.com IN A 192.0.2.3 143 If a validating resolver receives a query for cat.example.com, it 144 contacts its resolver (which may be itself) to query the example.com 145 servers and will get back an NSEC record stating that there are no 146 records (alphabetically) between albatross and elephant, or an NSEC3 147 record stating there is nothing between two hashed names. The 148 resolver then knows that cat.example.com does not exist; however, it 149 does not use the fact that the proof covers a range (albatross to 150 elephant) to suppress queries for other labels that fall within this 151 range. This means that if the validating resolver gets a query for 152 ball.example.com (or dog.example.com) it will once again go off and 153 query the example.com servers for these names. 155 Apart from wasting bandwidth, this also wastes resources on the 156 recursive server (it needs to keep state for outstanding queries), 157 wastes resources on the authoritative server (it has to answer 158 additional questions), increases latency (the end user has to wait 159 longer than necessary to get back an NXDOMAIN answer), can be used by 160 attackers to cause a DoS (see additional resources), and also has 161 privacy implications (e.g: typos leak out further than necessary). 163 Another example: assume that the (DNSSEC signed) "example.org" zone 164 contains: 166 avocado.example.org IN A 192.0.2.1 167 *.example.org IN A 192.0.2.2 168 zucchini.example.org IN A 192.0.2.3 170 If a query is received for leek.example.org, it contacts its resolver 171 (which may be itself) to query the example.org servers and will get 172 back an NSEC record stating that there are no records 173 (alphabetically) between avocado and zucchini (or an NSEC3 record 174 stating there is nothing between two hashed names), as well as an 175 answer for leek.example.org, with the label count of the signature 176 set to two (see [RFC7129], section 5.3 for more details). 178 If the validating resolver gets a query for banana.example.org it 179 will once again go off and query the example.org servers for 180 banana.example.org (even though it already has proof that there is a 181 wildcard record) - just like above, this has privacy implications, 182 wastes resources, can be used to contribute to a DoS, etc. 184 4. Background 186 DNSSEC [RFC4035] and [RFC5155] both provide "authenticated denial of 187 existence"; this is a cryptographic proof that the queried for name 188 does not exist or type does not exist. Proof that a name does not 189 exist is accomplished by providing a (DNSSEC secured) record 190 containing the names which appear alphabetically before and after the 191 queried for name. In the first example above, if the (DNSSEC 192 validating) recursive server were to query for dog.example.com it 193 would receive a (signed) NSEC record stating that there are no labels 194 between "albatross" and "elephant" (or, for NSEC3, a similar pair of 195 hashed names). This is a signed, cryptographic proof that these 196 names are the ones before and after the queried for label. As 197 dog.example.com falls within this range, the recursive server knows 198 that dog.example.com really does not exist. Proof that a type does 199 not exist is accomplished by providing a (DNSSEC secured) record 200 containing the queried for name, and a type bitmap which does not 201 include the requested type. 203 This document specifies that this NSEC/NSEC3 record should be used to 204 generate negative answers for any queries that the validating server 205 receives that fall within the range covered by the record (for the 206 TTL for the record). This document also specifies that a positive 207 answer should be generated for any queries that the validating server 208 receives that are proven to be covered by a wildcard record. 210 Section 4.5 of [RFC4035] says: 212 "In theory, a resolver could use wildcards or NSEC RRs to generate 213 positive and negative responses (respectively) until the TTL or 214 signatures on the records in question expire. However, it seems 215 prudent for resolvers to avoid blocking new authoritative data or 216 synthesizing new data on their own. Resolvers that follow this 217 recommendation will have a more consistent view of the namespace." 218 and "The reason for these recommendations is that, between the 219 initial query and the expiration of the data from the cache, the 220 authoritative data might have been changed (for example, via dynamic 221 update).". In other words, if a resolver generates negative answers 222 from an NSEC record, it will not send any queries for names within 223 that NSEC range (for the TTL). If a new name is added to the zone 224 during this interval the resolver will not know this. Similarly, if 225 the resolver is generating responses from a wildcard record, it will 226 continue to do so (for the TTL). 228 We believe this recommendation can be relaxed because, in the absence 229 of this technique, a lookup for the exact name could have come in 230 during this interval, and so a negative answer could already be 231 cached (see [RFC2308] for more background). This means that zone 232 operators should have no expectation that an added name would work 233 immediately. With DNSSEC and Aggressive NSEC, the TTL of the NSEC/ 234 NSEC3 record and the SOA.MINIMUM field are the authoritative 235 statement of how quickly a name can start working within a zone. 237 5. Aggressive use of Cache 239 Section 4.5 of [RFC4035] says that "In theory, a resolver could use 240 wildcards or NSEC RRs to generate positive and negative responses 241 (respectively) until the TTL or signatures on the records in question 242 expire. However, it seems prudent for resolvers to avoid blocking 243 new authoritative data or synthesizing new data on their own. 244 Resolvers that follow this recommendation will have a more consistent 245 view of the namespace". This document relaxes this this restriction, 246 see Section 7 for more detail. 248 If the negative cache of the validating resolver has sufficient 249 information to validate the query, the resolver SHOULD use NSEC, 250 NSEC3 and wildcard records aggressively. Otherwise, it MUST fall 251 back to send the query to the authoritative DNS servers. 253 5.1. NSEC 255 The validating resolver needs to check the existence of an NSEC RR 256 matching/covering the source of synthesis and an NSEC RR covering the 257 query name. 259 If denial of existence can be determined according to the rules set 260 out in Section 5.4 of [RFC4035], using NSEC records in the cache, 261 then the resolver can immediately return an NXDOMAIN or NODATA (as 262 appropriate) response. 264 5.2. NSEC3 266 NSEC3 aggressive negative caching is more difficult than NSEC 267 aggressive caching. If the zone is signed with NSEC3, the validating 268 resolver needs to check the existence of non-terminals and wildcards 269 which derive from query names. 271 If denial of existence can be determined according to the rules set 272 out in [RFC5155] Sections 8.4, 8.5, 8.6, 8.7, using NSEC3 records in 273 the cache, then the resolver can immediately return an NXDOMAIN or 274 NODATA response (as appropriate). 276 If a covering NSEC3 RR has Opt-Out flag, the covering NSEC3 RR does 277 not prove the non-existence of the domain name and the aggressive 278 negative caching is not possible for the domain name. 280 5.3. Wildcards 282 The last paragraph of [RFC4035] Section 4.5 also discusses the use of 283 wildcards and NSEC RRs to generate positive responses and recommends 284 that it not be relied upon. Just like the case for the aggressive 285 use of NSEC/NSEC3 for negative answers, we revise this 286 recommendation. 288 As long as the validating resolver can determine that a name would 289 not exist without the wildcard match, determined according to the 290 rules set out in Section 5.3.4 of [RFC4035] (NSEC), or in Section 8.8 291 of [RFC5155], it SHOULD synthesize an answer (or NODATA response) for 292 that name using the cached deduced wildcard. If the corresponding 293 wildcard record is not in the cache, it MUST fall back to send the 294 query to the authoritative DNS servers. 296 5.4. Consideration on TTL 298 The TTL value of negative information is especially important, 299 because newly added domain names cannot be used while the negative 300 information is effective. 302 Section 5 of [RFC2308] states that the maximum number of negative 303 cache TTL value is 3 hours (10800). It is RECOMMENDED that 304 validating resolvers limit the maximum effective TTL value of 305 negative responses (NSEC/NSEC3 RRs) to this same value. 307 Section 5 of [RFC2308] also states that a negative cache entry TTL is 308 taken from the minimum of the SOA.MINIMUM field and SOA's TTL. This 309 can be less than the TTL of an NSEC or NSEC3 record, since their TTL 310 is equal to the SOA.MINIMUM field (see [RFC4035]section 2.3 and 311 [RFC5155] section 3.) 313 A resolver that supports aggressive use of NSEC and NSEC3 SHOULD 314 reduce the TTL of NSEC and NSEC3 records to match the SOA.MINIMUM 315 field in the authority section of a negative response, if SOA.MINIMUM 316 is smaller. 318 6. Benefits 320 The techniques described in this document provide a number of 321 benefits, including (in no specific order): 323 Reduced latency: By answering directly from cache, validating 324 resolvers can immediately inform clients that the name they are 325 looking for does not exist, improving the user experience. 327 Decreased recursive server load: By answering queries from the cache 328 by synthesizing answers, validating servers avoid having to send a 329 query and wait for a response. In addition to decreasing the 330 bandwidth used, it also means that the server does not need to 331 allocate and maintain state, thereby decreasing memory and CPU 332 load. 334 Decreased authoritative server load: Because recursive servers can 335 answer queries without asking the authoritative server, the 336 authoritative servers receive fewer queries. This decreases the 337 authoritative server bandwidth, queries per second and CPU 338 utilization. 340 The scale of the benefit depends upon multiple factors, including the 341 query distribution. For example, at the time of this writing, around 342 65% of queries to Root Name servers result in NXDOMAIN responses (see 343 statistics from [root-servers.org]); this technique will eliminate a 344 sizable quantity of these. 346 The technique described in this document may also mitigate so-called 347 "random QNAME attacks", in which attackers send many queries for 348 random sub-domains to resolvers. As the resolver will not have the 349 answers cached, it has to ask external servers for each random query, 350 leading to a DoS on the authoritative servers (and often resolvers). 351 Aggressive NSEC may help mitigate these attacks by allowing the 352 resolver to answer directly from cache for any random queries which 353 fall within already requested ranges. It will not always work as an 354 effective defense, not least because not many zones are DNSSEC signed 355 at all -- but it will still provide an additional layer of defense. 357 As these benefits are only accrued by those using DNSSEC, it is hoped 358 that these techniques will lead to more DNSSEC deployment. 360 7. Update to RFC 4035 362 Section 4.5 of [RFC4035] shows that "In theory, a resolver could use 363 wildcards or NSEC RRs to generate positive and negative responses 364 (respectively) until the TTL or signatures on the records in question 365 expire. However, it seems prudent for resolvers to avoid blocking 366 new authoritative data or synthesizing new data on their own. 367 Resolvers that follow this recommendation will have a more consistent 368 view of the namespace". 370 The paragraph is updated as follows: 372 +-----------------------------------------------------------------+ 373 | Once the records are validated, DNSSEC enabled validating | 374 | resolvers SHOULD use wildcards and NSEC/NSEC3 resource records | 375 | to generate positive and negative responses until the | 376 | effective TTLs or signatures for those records expire. | 377 +-----------------------------------------------------------------+ 379 8. IANA Considerations 381 This document has no IANA actions. 383 9. Security Considerations 385 Use of NSEC / NSEC3 resource records without DNSSEC validation may 386 create serious security issues, and so this technique requires DNSSEC 387 validation. 389 Newly registered resource records may not be used immediately. 390 However, choosing suitable TTL value and negative cache TTL value 391 (SOA MINIMUM field) will mitigate the delay concern, and it is not a 392 security problem. 394 It is also suggested to limit the maximum TTL value of NSEC / NSEC3 395 resource records in the negative cache to, for example, 10800 seconds 396 (3hrs), to mitigate this issue. 398 Although the TTL of NSEC/NSEC3 records is typically fairly short 399 (minutes or hours), their RRSIG expiration time can be much further 400 in the future (weeks). An attacker who is able to successfully spoof 401 responses might poison a cache with old NSEC/NSEC3 records. If the 402 resolver is NOT making aggressive use of NSEC/NSEC3, the attacker has 403 to repeat the attack for every query. If the resolver IS making 404 aggressive use of NSEC/NSEC3, one successful attack would be able to 405 suppress many queries for new names, up to the negative TTL. 407 10. Implementation Status 409 [ Editor note: RFC Editor, please remove this entire section. 410 RFC6982 says: "Since this information is necessarily time dependent, 411 it is inappropriate for inclusion in a published RFC." ] 413 Unbound currently implements aggressive negative caching, as does 414 Google Public DNS. 416 11. Acknowledgments 418 The authors gratefully acknowledge DLV [RFC5074] author Samuel Weiler 419 and the Unbound developers. 421 Thanks to Mark Andrews for providing the helpful notes for 422 implementors provided in Appendix B. 424 The authors would like to specifically thank Stephane Bortzmeyer (for 425 standing next to and helping edit), Ralph Dolmans, Tony Finch, Tatuya 426 JINMEI for extensive review and comments, and also Mark Andrews, 427 Casey Deccio, Alexander Dupuy, Olafur Gudmundsson, Bob Harold, Shumon 428 Huque, John Levine, Pieter Lexis, Matthijs Mekking (who even sent 429 pull requests!) and Ondrej Sury. Mark Andrews also provided the 430 helpful notes for implementors (https://www.ietf.org/mail- 431 archive/web/dnsop/current/msg18332.html) which we made into 432 Appendix B. 434 11.1. Change History 436 RFC Editor: Please remove this section prior to publication. 438 -05 to -06: 440 o Moved some dangling text around - when the examples were added 441 some text added in the wrong place. 443 o There were some bits which mentioned "negative" in the title. 445 o We had the cut-and-paste of what changed in 4035 twice. 447 o Clarified that this also allows NODATA responses to be 448 synthesized. 450 -04 to -05: 452 o Bob pointed out that I did a stupid - when I added the wildcard to 453 'example.com' I made the example wrong / confusing. I have 454 attempted to fix this by adding a second example zone 455 (example.org) with the wildcard instead. 457 o More helpful changes (in a pull request, thanks!) from Matthijs 459 o Included Mark Andrew's useful explanation of how to tell ENT from 460 NXD as an Appendix. 462 -03 to -04: 464 o Working group does want the "positive" answers, not just negative 465 ones. This requires reading what used to be Section 7, and a 466 bunch of cleanup, including: 468 * Additional text in the Problem Statement 470 * Added a wildcard record to the zone. 472 * Added "or positive answers from wildcards" type text (where 473 appropriate) to explain that this isn't just for negative 474 answers. 476 * Reworded much of the Wildcard text. 478 o Incorporated pull request from Tony Finch (thanks!): 479 https://github.com/wkumari/draft-ietf-dnsop-nsec-aggressiveuse/ 480 pull/1 482 o More fixups from Tony (including text): https://www.ietf.org/mail- 483 archive/web/dnsop/current/msg18271.html. This included much 484 clearer text on TTL, references to the NSEC / NSEC3 RFCs (instead 485 of my clumsy summary), good text on replays, etc. 487 o Converted the "zone file" to a figure to make it more readable. 489 o Text from Tim W: "If a validating resolver receives a query for 490 cat.example.com, it contacts its resolver (which may be itself) to 491 query..." - which satisfies Jinmei's concern (which I was too 492 dense to grock). 494 o Fixup of the "validation required" in security considerations. 496 -02 to -03: 498 o Integrated a bunch of comments from Matthijs Mekking - details in: 499 https://github.com/wkumari/draft-ietf-dnsop-nsec-aggressiveuse/ 500 pull/1. I decided to keep "Aggressive Negative Caching" instead 501 of "Aggressive USE OF Negative Caching" for readability. 503 o Attempted to address Bob Harold's comment on the readability 504 issues with "But, it will be more effective when both are 505 enabled..." in Section 5.4 - https://www.ietf.org/mail- 506 archive/web/dnsop/current/msg17997.html 508 o MAYs and SHOULD drifted in the text block. Fixed - thanks to 509 https://mailarchive.ietf.org/arch/msg/ 510 dnsop/2ljmmzxtIMCFMLOZmWcSbTYVOy4 512 o A number of good edits from Stephane in: https://www.ietf.org/ 513 mail-archive/web/dnsop/current/msg18109.html 515 o A bunch more edits from Jinmei, as in: https://www.ietf.org/mail- 516 archive/web/dnsop/current/msg18206.html 518 -01 to -02: 520 o Added Section 6 - Benefits (as suggested by Jinmei). 522 o Removed Appendix B (Jinmei) 523 o Replaced "full-service" with "validating" (where applicable) 525 o Integrated other comments from Jinmei from https://www.ietf.org/ 526 mail-archive/web/dnsop/current/msg17875.html 528 o Integrated comment from co-authors, including re-adding parts of 529 Appendix B, terminology, typos. 531 o Tried to explain under what conditions this may actually mitigate 532 attacks. 534 -00 to -01: 536 o Comments from DNSOP meeting in Berlin. 538 o Changed intended status to Standards Track (updates RFC 4035) 540 o Added a section "Updates to RFC 4035" 542 o Some language clarification / typo / cleanup 544 o Cleaned up the TTL section a bit. 546 o Removed Effects section, Additional proposal section, and pseudo 547 code. 549 o Moved "mitigation of random subdomain attacks" to Appendix. 551 From draft-fujiwara-dnsop-nsec-aggressiveuse-03 -> draft-ietf-dnsop- 552 nsec-aggressiveuse 554 o Document adopted by DNSOP WG. 556 o Adoption comments 558 o Changed main purpose to performance 560 o Use NSEC3/Wildcard keywords 562 o Improved wordings (from good comments) 564 o Simplified pseudo code for NSEC3 566 o Added Warren as co-author. 568 o Reworded much of the problem statement 570 o Reworked examples to better explain the problem / solution. 572 11.1.1. Version draft-fujiwara-dnsop-nsec-aggressiveuse-01 574 o Added reference to DLV [RFC5074] and imported some sentences. 576 o Added Aggressive Negative Caching Flag idea. 578 o Added detailed algorithms. 580 11.1.2. Version draft-fujiwara-dnsop-nsec-aggressiveuse-02 582 o Added reference to [I-D.vixie-dnsext-resimprove] 584 o Added considerations for the CD bit 586 o Updated detailed algorithms. 588 o Moved Aggressive Negative Caching Flag idea into Additional 589 Proposals 591 11.1.3. Version draft-fujiwara-dnsop-nsec-aggressiveuse-03 593 o Added "Partial implementation" 595 o Section 4,5,6 reorganized for better representation 597 o Added NODATA answer in Section 4 599 o Trivial updates 601 o Updated pseudo code 603 11.2. new section 605 12. References 607 12.1. Normative References 609 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 610 Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/ 611 RFC2119, March 1997, 612 . 614 [RFC2308] Andrews, M., "Negative Caching of DNS Queries (DNS 615 NCACHE)", RFC 2308, DOI 10.17487/RFC2308, March 1998, 616 . 618 [RFC4035] Arends, R., Austein, R., Larson, M., Massey, D., and S. 619 Rose, "Protocol Modifications for the DNS Security 620 Extensions", RFC 4035, DOI 10.17487/RFC4035, March 2005, 621 . 623 [RFC4592] Lewis, E., "The Role of Wildcards in the Domain Name 624 System", RFC 4592, DOI 10.17487/RFC4592, July 2006, 625 . 627 [RFC5074] Weiler, S., "DNSSEC Lookaside Validation (DLV)", RFC 5074, 628 DOI 10.17487/RFC5074, November 2007, 629 . 631 [RFC5155] Laurie, B., Sisson, G., Arends, R., and D. Blacka, "DNS 632 Security (DNSSEC) Hashed Authenticated Denial of 633 Existence", RFC 5155, DOI 10.17487/RFC5155, March 2008, 634 . 636 [RFC7129] Gieben, R. and W. Mekking, "Authenticated Denial of 637 Existence in the DNS", RFC 7129, DOI 10.17487/RFC7129, 638 February 2014, . 640 [RFC7719] Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS 641 Terminology", RFC 7719, DOI 10.17487/RFC7719, December 642 2015, . 644 12.2. Informative References 646 [I-D.vixie-dnsext-resimprove] 647 Vixie, P., Joffe, R., and F. Neves, "Improvements to DNS 648 Resolvers for Resiliency, Robustness, and Responsiveness", 649 draft-vixie-dnsext-resimprove-00 (work in progress), June 650 2010. 652 [RFC8020] Bortzmeyer, S. and S. Huque, "NXDOMAIN: There Really Is 653 Nothing Underneath", RFC 8020, DOI 10.17487/RFC8020, 654 November 2016, . 656 [root-servers.org] 657 IANA, "Root Server Technical Operations Assn", 658 . 660 Appendix A. Detailed implementation notes 662 o Previously, cached negative responses were indexed by QNAME, 663 QCLASS, QTYPE, and the setting of the CD bit (see RFC 4035, 664 Section 4.7), and only queries matching the index key would be 665 answered from the cache. With aggressive negative caching, the 666 validator, in addition to checking to see if the answer is in its 667 cache before sending a query, checks to see whether any cached and 668 validated NSEC record denies the existence of the sought 669 record(s). Using aggressive negative caching, a validator will 670 not make queries for any name covered by a cached and validated 671 NSEC record. Furthermore, a validator answering queries from 672 clients will synthesize a negative answer (or NODATA response) 673 whenever it has an applicable validated NSEC in its cache unless 674 the CD bit was set on the incoming query. (Imported from 675 Section 6 of [RFC5074]). 677 o Implementing aggressive negative caching suggests that a validator 678 will need to build an ordered data structure of NSEC and NSEC3 679 records for each signer domain name of NSEC / NSEC3 records in 680 order to efficiently find covering NSEC / NSEC3 records. Call the 681 table as NSEC_TABLE. (Imported from Section 6.1 of [RFC5074] and 682 expanded.) 684 o The aggressive negative caching may be inserted at the cache 685 lookup part of the recursive resolvers. 687 o If errors happen in aggressive negative caching algorithm, 688 resolvers MUST fall back to resolve the query as usual. "Resolve 689 the query as usual" means that the resolver must process the query 690 as though it does not implement aggressive negative caching. 692 Appendix B. Procedure for determining ENT vs NXDOMAN with NSEC 694 This procedure outlines how to determine if a given name does not 695 exist, or is an ENT (Empty Non-Terminal, see [RFC5155] Section 1.3) 696 with NSEC. 698 If the NSEC record has not been verified as secure discard it. 700 If the given name sorts before or matches the NSEC owner name discard 701 it as it does not prove the NXDOMAIN or ENT. 703 If the given name is a subdomain of the NSEC owner name and the NS 704 bit is present and the SOA bit is absent then discard the NSEC as it 705 is from a parent zone. 707 If the next domain name sorts after the NSEC owner name and the given 708 name sorts after or matches next domain name then discard the NSEC 709 record as it does not prove the NXDOMAIN or ENT. 711 If the next domain name sorts before or matches the NSEC owner name 712 and the given name is not a subdomain of the next domain name then 713 discard the NSEC as it does not prove the NXDOMAIN or ENT. 715 You now have a NSEC record that proves the NXDOMAIN or ENT. 717 If the next domain name is a subdomain of the given name you have a 718 ENT otherwise you have a NXDOMAIN. 720 Authors' Addresses 722 Kazunori Fujiwara 723 Japan Registry Services Co., Ltd. 724 Chiyoda First Bldg. East 13F, 3-8-1 Nishi-Kanda 725 Chiyoda-ku, Tokyo 101-0065 726 Japan 728 Phone: +81 3 5215 8451 729 Email: fujiwara@jprs.co.jp 731 Akira Kato 732 Keio University/WIDE Project 733 Graduate School of Media Design, 4-1-1 Hiyoshi 734 Kohoku, Yokohama 223-8526 735 Japan 737 Phone: +81 45 564 2490 738 Email: kato@wide.ad.jp 740 Warren Kumari 741 Google 742 1600 Amphitheatre Parkway 743 Mountain View, CA 94043 744 US 746 Email: warren@kumari.net