idnits 2.17.1 draft-fujiwara-dnsop-nsec-aggressiveuse-03.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 (March 18, 2016) is 2960 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- ** Obsolete normative reference: RFC 7719 (Obsoleted by RFC 8499) Summary: 1 error (**), 0 flaws (~~), 1 warning (==), 2 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 Intended status: Informational A. Kato 5 Expires: September 19, 2016 Keio/WIDE 6 March 18, 2016 8 Aggressive use of NSEC/NSEC3 9 draft-fujiwara-dnsop-nsec-aggressiveuse-03 11 Abstract 13 While DNS highly depends on cache, its cache usage of non-existence 14 information has been limited to exact matching. This draft proposes 15 the aggressive use of a NSEC/NSEC3 resource record, which is able to 16 express non-existence of a range of names authoritatively. With this 17 proposal, it is expected that shorter latency to many of negative 18 responses as well as some level of mitigation of random sub-domain 19 attacks (referred to as "Water Torture" attacks). It is also 20 expected that non-existent TLD queries to Root DNS servers will 21 decrease. 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 September 19, 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. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 3. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 4 60 4. Proposed Solution . . . . . . . . . . . . . . . . . . . . . . 4 61 4.1. Aggressive Negative Caching . . . . . . . . . . . . . . . 4 62 4.2. NSEC . . . . . . . . . . . . . . . . . . . . . . . . . . 5 63 4.3. NSEC3 . . . . . . . . . . . . . . . . . . . . . . . . . . 5 64 4.4. NSEC3 Opt-Out . . . . . . . . . . . . . . . . . . . . . . 6 65 4.5. Wildcard . . . . . . . . . . . . . . . . . . . . . . . . 6 66 4.6. Consideration on TTL . . . . . . . . . . . . . . . . . . 6 67 5. Additional Considerations . . . . . . . . . . . . . . . . . . 6 68 5.1. The CD Bit . . . . . . . . . . . . . . . . . . . . . . . 6 69 5.2. Detecting random subdomain attacks . . . . . . . . . . . 7 70 6. Possible side effect . . . . . . . . . . . . . . . . . . . . 7 71 7. Additional proposals . . . . . . . . . . . . . . . . . . . . 7 72 7.1. Partial implementation . . . . . . . . . . . . . . . . . 7 73 7.2. Aggressive negative caching without DNSSEC validation . . 8 74 7.3. Aggressive negative caching flag idea . . . . . . . . . . 8 75 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 76 9. Security Considerations . . . . . . . . . . . . . . . . . . . 8 77 10. Implementation Status . . . . . . . . . . . . . . . . . . . . 9 78 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 9 79 12. Change History . . . . . . . . . . . . . . . . . . . . . . . 9 80 12.1. Version 01 . . . . . . . . . . . . . . . . . . . . . . . 9 81 12.2. Version 02 . . . . . . . . . . . . . . . . . . . . . . . 9 82 12.3. Version 03 . . . . . . . . . . . . . . . . . . . . . . . 9 83 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 84 13.1. Normative References . . . . . . . . . . . . . . . . . . 10 85 13.2. Informative References . . . . . . . . . . . . . . . . . 10 86 Appendix A. Aggressive negative caching from RFC 5074 . . . . . 11 87 Appendix B. Detailed implementation idea . . . . . . . . . . . . 11 88 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 14 90 1. Introduction 92 While negative (non-existence) information of DNS caching mechanism 93 has been known as DNS negative cache [RFC2308], it requires exact 94 matching in most cases. Assume that "example.com" zone doesn't have 95 names such as "a.example.com" and "b.example.com". When a full- 96 service resolver receives a query "a.example.com" , it performs a DNS 97 resolution process, and eventually gets NXDOMAIN and stores it into 98 its negative cache. When the full-service resolver receives another 99 query "b.example.com", it doesn't match with "a.example.com". So it 100 will send a query to one of the authoritative servers of 101 "example.com". This was because the NXDOMAIN response just says 102 there is no such name "a.example.com" and it doesn't tell anything 103 for "b.example.com". 105 Section 5 of [RFC2308] seems to show that negative answers should be 106 cached only for the exact query name, and not (necessarily) for 107 anything below it. 109 Recently, DNSSEC [RFC4035] [RFC5155] has been practically deployed. 110 Two types of resource record (NSEC and NSEC3) along with their RRSIG 111 records represent authentic non-existence. For a zone signed with 112 NSEC, it would be possible to use the information carried in NSEC 113 resource records to indicate that a range of names where no valid 114 name exists. Such use is discouraged by Section 4.5 of RFC 4035, 115 however. 117 This document proposes to make a minor change to RFC 4035 and a full- 118 service resolver can use NSEC/NSEC3 resource records aggressively so 119 that the resolver responds with NXDOMAIN immediately if the name in 120 question falls into a range expressed by a NSEC/NSEC3 resource 121 record. 123 Aggressive Negative Caching was first proposed in Section 6 of DNSSEC 124 Lookaside Validation (DLV) [RFC5074] in order to find covering NSEC 125 records efficiently. Unbound [UNBOUND] has aggressive negative 126 caching code in its DLV validator. Unbound TODO file contains "NSEC/ 127 NSEC3 aggressive negative caching". 129 Section 3 of [I-D.vixie-dnsext-resimprove] ("Stopping Downward Cache 130 Search on NXDOMAIN") proposed another approach to use NXDOMAIN 131 information effectively. 133 2. Terminology 135 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 136 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 137 document are to be interpreted as described in RFC 2119 [RFC2119]. 139 Many of the specialized terms used in this specification are defined 140 in DNS Terminology [RFC7719]. 142 3. Problem Statement 144 Random sub-domain attacks (referred to as "Water Torture" attacks or 145 NXDomain attacks) send many non-existent queries to full-service 146 resolvers. Their query names consist of random prefixes and a target 147 domain name. The negative cache does not work well and target full- 148 service resolvers result in sending queries to authoritative DNS 149 servers of the target domain name. 151 When number of queries is large, the full-service resolvers drop 152 queries from both legitimate users and attackers as their outstanding 153 queues are filled up. 155 For example, BIND 9.10.2 [BIND9] full-service resolvers answer 156 SERVFAIL and Unbound 1.5.2 full-service resolvers drop most of 157 queries under 10,000 queries per second attack. 159 The countermeasures implemented at this moment are rate limiting and 160 disabling name resolution of target domain names in ad-hoc manner. 162 4. Proposed Solution 164 4.1. Aggressive Negative Caching 166 If the target domain names are DNSSEC signed, aggressive use of NSEC/ 167 NSEC3 resource records mitigates the problem. 169 Section 4.5 of [RFC4035] shows that "In theory, a resolver could use 170 wildcards or NSEC RRs to generate positive and negative responses 171 (respectively) until the TTL or signatures on the records in question 172 expire. However, it seems prudent for resolvers to avoid blocking 173 new authoritative data or synthesizing new data on their own. 174 Resolvers that follow this recommendation will have a more consistent 175 view of the namespace". 177 To reduce non-existent queries sent to authoritative DNS servers, it 178 is suggested to relax this restriction as follows: 180 +--------------------------------------------------------------+ 181 | DNSSEC enabled full-service resolvers MAY use | 182 | NSEC/NSEC3 resource records to generate negative responses | 183 | until their effective TTLs or signatures on the records | 184 | in question expire. | 185 +--------------------------------------------------------------+ 187 If the full-service resolver's cache have enough information to 188 validate the query, the full-service resolver MAY use NSEC/NSEC3/ 189 wildcard records aggressively. Otherwise, the full-service resolver 190 MUST fall back to send the query to the authoritative DNS servers. 192 Necessary information to validate are matching/covering NSEC/NSEC3 of 193 the wildcards which may match the query name, matching/covering NSEC/ 194 NSEC3 of non-terminals which derive from the query name and matching/ 195 covering NSEC/NSEC3 of the query name. 197 If the query name has the matching NSEC/NSEC3 RR and it shows the 198 query type does not exist, the full-service resolver is possible to 199 respond with NODATA (empty) answer. 201 4.2. NSEC 203 A full-service resolver implementation SHOULD support aggressive use 204 of NSEC and enable it by default. It SHOULD provide a configuration 205 knob to disable aggressive use of NSEC. 207 The validating resolver need to check the existence of matching 208 wildcards which derive from the query name, covering NSEC RRs of the 209 matching wildcards and covering NSEC RR of the query name. 211 If the full-service resolver's cache contains covering NSEC RRs of 212 matching wildcards and the covering NSEC RR of the query name, the 213 full-service resolver is possible to respond with NXDOMAIN error 214 immediately. 216 4.3. NSEC3 218 NSEC3 aggressive negative caching is more difficult. If the zone is 219 signed with NSEC3, the validating resolver need to check the 220 existence of non-terminals and wildcards which derive from query 221 names. 223 If the full-service resolver's cache contains covering NSEC3 RRs of 224 matching wildcards, the covering NSEC3 RRs of the non-terminals and 225 the covering NSEC3 RR of the query name, the full-service resolver is 226 possible to respond with NXDOMAIN error immediately. 228 If the validating resolver proves the non-exisence of the non- 229 terminal domain name of the query name, the query name does not 230 exist. 232 To identify signing types of the zone, validating resolvers need to 233 build separated cache of NSEC and NSEC3 resource records for each 234 signer domain name. 236 When a query name is not in the regular cache, find closest enclosing 237 NS RRset in the regular cache. The owner of the closest enclosing NS 238 RRset may be the longest signer domain name of the query name. If 239 there is no entry in the NSEC/NSEC3 cache of the signer domain name, 240 aggressive negative caching is not possible at this moment. 241 Otherwise, there is at least one NSEC or NSEC3 resource records. The 242 record shows the signing type. 244 A full-service resolver implementation MAY support aggressive use of 245 NSEC3. It SHOULD provide a configuration knob to disable aggressive 246 use NSEC3 in this case. 248 4.4. NSEC3 Opt-Out 250 If the zone is signed with NSEC3 and with Opt-Out flag set to 1, the 251 aggressive negative caching is not possible at the zone. 253 4.5. Wildcard 255 Even if a wildcard is cached, it is necessary to send a query to an 256 authoritative server to ensure that the name in question doesn't 257 exist as long as the name is not in the negative cache. 259 When aggressive use is enabled, regardless of description of 260 Section 4.5 of [RFC4035], it is possible to send a positive response 261 immediately when the name in question matches a NSEC/NSEC3 RRs in the 262 negative cache. 264 4.6. Consideration on TTL 266 This function needs care on the TTL value of negative information 267 because newly added domain names cannot be used while the negative 268 information is effective. RFC 2308 states the maximum number of 269 negative cache TTL value is 10800 (3 hours). So the full-service 270 resolver SHOULD limit the maximum effective TTL value of negative 271 responses (NSEC/NSEC3 RRs) to 10800 (3 hours). It is reasonably 272 small but still effective for the purpose of this document as it can 273 eliminate significant amount of DNS attacks with randomly generated 274 names. 276 5. Additional Considerations 278 5.1. The CD Bit 280 The CD bit disables signature validation. It is one of the basic 281 functions of DNSSEC protocol and it SHOULD NOT be changed. However, 282 attackers may set the CD bit to their attack queries and the 283 aggressive negative caching will be of no use. 285 Ignoring the CD bit function may break the DNSSEC protocol. 287 This draft proposes that the CD bit may be ignored to support 288 aggressive negative caching when the full-service resolver is under 289 attacks with CD bit set. 291 5.2. Detecting random subdomain attacks 293 Full-service resolvers should detect conditions under random 294 subdomain attacks. When they are under attacks, their outstanding 295 queries increase. If there are some destination addresses whose 296 outstanding queries are many, they may contain attack target domain 297 names. Existing countermeasures may implement attack detection. 299 6. Possible side effect 301 Aggressive use of NSEC/NSEC3 resource records may decrease queries to 302 Root DNS servers. 304 People may generate many typos in TLD, and they will result in 305 unnecessary DNS queries. Some implementations leak non-existent TLD 306 queries whose second level domain are different each other. Well 307 observed TLDs are ".local" and ".belkin". With this proposal, it is 308 possible to return NXDOMAIN immediately to such queries without 309 further DNS recursive resolution process. It may reduces round trip 310 time, as well as reduces the DNS queries to corresponding 311 authoritative servers, including Root DNS servers. 313 7. Additional proposals 315 There are additional proposals to the aggressive negative caching. 317 7.1. Partial implementation 319 It is possible to implement aggressive negative caching partially. 321 DLV aggressive negative caching [RFC5074] is an implementation of 322 NSEC aggressive negative caching which targets DLV domain names. 324 NSEC only aggressive negative caching is easier to implement NSEC/ 325 NSEC3 aggressive negative caching (full implantation) because NSEC3 326 handling is hard to implement. 328 Root only aggressive negative caching is possible. It uses NSEC and 329 RRSIG resource records whose signer domain name is root. 331 An implementation without detecting attacks is possible. It cannot 332 ignore the CD bit and the effectiveness may be limited. 334 7.2. Aggressive negative caching without DNSSEC validation 336 Aggressive negative caching may be applicable to full-service 337 resolvers without DNSSEC validation. They can set DNSSEC OK bit in 338 query packets to obtain corresponding NSEC/NSEC3 resource records. 339 While the full-service resolvers SHOULD validate the NSEC/NSEC3 340 resource records, they MAY use the records to respond NXDOMAIN error 341 immediately without DNSSEC validation. 343 However, it is highly recommended to apply DNSSEC validation. 345 7.3. Aggressive negative caching flag idea 347 Authoritative DNS servers that dynamically generate NSEC records 348 normally generate minimally covering NSEC Records [RFC4470]. 349 Aggressive negative caching does not work with minimally covering 350 NSEC records. Most of DNS operators don't want zone enumeration and 351 zone information leaks. They prefer NSEC resource records with 352 narrow ranges. When there is a flag that show a full-service 353 resolver support the aggressive negative caching and a query have the 354 aggressive negative caching flag, authoritative DNS servers can 355 generate NSEC resource records with wider range under random 356 subdomain attacks. 358 However, changing range of minimally covering NSEC Records may be 359 implemented by detecting attacks. Authoritative DNS servers can 360 answer any range of minimally covering NSEC Records. 362 8. IANA Considerations 364 This document has no IANA actions. 366 9. Security Considerations 368 Newly registered resource records may not be used immediately. 369 However, choosing suitable TTL value will mitigate the problem and it 370 is not a security problem. 372 It is also suggested to limit the maximum TTL value of NSEC resource 373 records in the negative cache to, for example, 10800 seconds (3hrs), 374 to mitigate the issue. Implementations which comply with this 375 proposal is suggested to have a configurable maximum value of NSEC 376 RRs in the negative cache. 378 Aggressive use of NSEC/NSEC3 resource records without DNSSEC 379 validation may cause security problems. 381 10. Implementation Status 383 Unbound has aggressive negative caching code in its DLV validator. 384 The author implemented NSEC aggressive caching using Unbound and its 385 DLV validator code. 387 11. Acknowledgments 389 The authors gratefully acknowledge DLV [RFC5074] author Samuel Weiler 390 and Unbound developers. Olafur Gudmundsson and Pieter Lexis proposed 391 aggressive negative caching flag idea. Valuable comments were 392 provided by Bob Harold, Tatuya JINMEI, Shumon Huque, Mark Andrews, 393 and Casey Deccio. 395 12. Change History 397 This section is used for tracking the update of this document. Will 398 be removed after finalize. 400 12.1. Version 01 402 o Added reference to DLV [RFC5074] and imported some sentences. 404 o Added Aggressive Negative Caching Flag idea. 406 o Added detailed algorithms. 408 12.2. Version 02 410 o Added reference to [I-D.vixie-dnsext-resimprove] 412 o Added considerations for the CD bit 414 o Updated detailed algorithms. 416 o Moved Aggressive Negative Caching Flag idea into Additional 417 Proposals 419 12.3. Version 03 421 o Added "Partial implementation" 423 o Section 4,5,6 reorganized for better representation 425 o Added NODATA answer in Section 4 427 o Trivial updates 428 o Updated pseudo code 430 13. References 432 13.1. Normative References 434 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 435 Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/ 436 RFC2119, March 1997, 437 . 439 [RFC2308] Andrews, M., "Negative Caching of DNS Queries (DNS 440 NCACHE)", RFC 2308, DOI 10.17487/RFC2308, March 1998, 441 . 443 [RFC4035] Arends, R., Austein, R., Larson, M., Massey, D., and S. 444 Rose, "Protocol Modifications for the DNS Security 445 Extensions", RFC 4035, DOI 10.17487/RFC4035, March 2005, 446 . 448 [RFC4470] Weiler, S. and J. Ihren, "Minimally Covering NSEC Records 449 and DNSSEC On-line Signing", RFC 4470, DOI 10.17487/ 450 RFC4470, April 2006, 451 . 453 [RFC5074] Weiler, S., "DNSSEC Lookaside Validation (DLV)", RFC 5074, 454 DOI 10.17487/RFC5074, November 2007, 455 . 457 [RFC5155] Laurie, B., Sisson, G., Arends, R., and D. Blacka, "DNS 458 Security (DNSSEC) Hashed Authenticated Denial of 459 Existence", RFC 5155, DOI 10.17487/RFC5155, March 2008, 460 . 462 [RFC7719] Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS 463 Terminology", RFC 7719, DOI 10.17487/RFC7719, December 464 2015, . 466 13.2. Informative References 468 [BIND9] Internet Systems Consortium, Inc., "Name Server Software", 469 2000, . 471 [I-D.vixie-dnsext-resimprove] 472 Vixie, P., Joffe, R., and F. Neves, "Improvements to DNS 473 Resolvers for Resiliency, Robustness, and Responsiveness", 474 draft-vixie-dnsext-resimprove-00 (work in progress), June 475 2010. 477 [UNBOUND] NLnet Labs, "Unbound DNS validating resolver", 2006, 478 . 480 Appendix A. Aggressive negative caching from RFC 5074 482 Imported from Section 6 of [RFC5074]. 484 Previously, cached negative responses were indexed by QNAME, QCLASS, 485 QTYPE, and the setting of the CD bit (see RFC 4035, Section 4.7), and 486 only queries matching the index key would be answered from the cache. 487 With aggressive negative caching, the validator, in addition to 488 checking to see if the answer is in its cache before sending a query, 489 checks to see whether any cached and validated NSEC record denies the 490 existence of the sought record(s). 492 Using aggressive negative caching, a validator will not make queries 493 for any name covered by a cached and validated NSEC record. 494 Furthermore, a validator answering queries from clients will 495 synthesize a negative answer whenever it has an applicable validated 496 NSEC in its cache unless the CD bit was set on the incoming query. 498 Imported from Section 6.1 of [RFC5074]. 500 Implementing aggressive negative caching suggests that a validator 501 will need to build an ordered data structure of NSEC records in order 502 to efficiently find covering NSEC records. Only NSEC records from 503 DLV domains need to be included in this data structure. 505 Appendix B. Detailed implementation idea 507 Section 6.1 of [RFC5074] is expanded as follows. 509 Implementing aggressive negative caching suggests that a validator 510 will need to build an ordered data structure of NSEC and NSEC3 511 records for each signer domain name of NSEC / NSEC3 records in order 512 to efficiently find covering NSEC / NSEC3 records. Call the table as 513 NSEC_TABLE. 515 The aggressive negative caching may be inserted at the cache lookup 516 part of the full-service resolvers. 518 If errors happen in aggressive negative caching algorithm, resolvers 519 MUST fall back to resolve the query as usual. "Resolve the query as 520 usual" means that the full-resolver resolve the query in Recursive- 521 mode as if the full-service resolver does not implement aggressive 522 negative caching. 524 To implement aggressive negative caching, resolver algorithm near 525 cache lookup will be changed as follows: 527 QNAME = the query name; 528 QTYPE = the query type; 529 if ({QNAME,QTYPE} entry exists in the cache) { 530 // the resolver responds the RRSet from the cache 531 resolve the query as usual; 532 } 534 // if NSEC* exists, QTYPE existence is proved by type bitmap 535 if (matching NSEC/NSEC3 of QNAME exists in the cache) { 536 if (QTYPE exists in type bitmap of NSEC/NSEC3 of QNAME) { 537 // the entry exists, however, it is not in the cache. 538 // need to iterate QNAME/QTYPE. 539 resolve the query as usual; 540 } else { 541 // QNAME exists, QTYPE does not exist. 542 the resolver can generate NODATA response; 543 } 544 } 546 // Find closest enclosing NS RRset in the cache. 547 // The owner of this NS RRset will be a suffix of the QNAME 548 // - the longest suffix of any NS RRset in the cache. 549 SIGNER = closest enclosing NS RRSet of QNAME in the cache; 551 // Check the SOA RR of the SIGNER 552 if (SOA RR of SIGNER does not exist in the cache 553 or SIGNER zone is not signed or not validated) { 554 Resolve the query as usual; 555 } 557 if (SIGNER zone does not have NSEC_TABLE) { 558 Resolve the query as usual; 559 } 561 if (SIGNER zone is signed with NSEC) { // NSEC mode 563 // Check the non-existence of QNAME 564 CoveringNSEC = Find the covering NSEC of QNAME; 565 if (Covering NSEC doesn't exist in the cache) { 566 Resolve the query as usual. 567 } 569 // Select the longest existing name of QNAME from covering NSEC 570 LongestExistName = common part of both owner name and 571 next domain name of CoveringNSEC; 573 if (*.LongestExistName entry exists in the cache) { 574 the resolver can generate positive response 575 // synthesize the wildcard *.TEST 576 } 577 if covering NSEC RR of "*.LongestExistName" at SIGNER zone exists 578 in the cache { 579 the resolver can generate negative response; 580 } 581 //*.LongestExistName may exist. cannot generate negative response 582 Resolve the query as usual. 584 } else 585 if (SIGNER zone is signed with NSEC3 and does not use Opt-Out) { 586 // NSEC3 mode 588 TEST = SIGNER; 589 while (TEST != QNAME) { 590 // if any error happens in this loop, break this loop 591 UPPER = TEST; 592 add a label from the QNAME to the start of TEST; 593 // TEST = label.UPPER 594 if (TEST name entry exist in the cache 595 || matching NSEC3 of TEST exist in the cache) { 596 // TEST exist 597 continue; // need to check rest of QNAME 598 } 599 if (covering NSEC3 of TEST exist in the cache) { 600 // (non-)terminal name TEST does not exist 601 if (*.UPPER name entry exist in the cache) { 602 // TEST does not exist and *.UPPER exist 603 the resolver can generate positive response; 604 } else 605 if (covering NSEC3 of *.UPPER exist in the cache) { 606 // TEST does not exist and *.UPPER does not exist 607 the resolver can generate negative response; 608 } 609 break; // Lack of information (No *.UPPER information) 610 } 611 break; // Lack of information (No TEST information) 612 } 613 // no matching/covering NSEC3 of QNAME information 614 Resolve the query as usual 615 } 617 Authors' Addresses 619 Kazunori Fujiwara 620 Japan Registry Services Co., Ltd. 621 Chiyoda First Bldg. East 13F, 3-8-1 Nishi-Kanda 622 Chiyoda-ku, Tokyo 101-0065 623 Japan 625 Phone: +81 3 5215 8451 626 Email: fujiwara@jprs.co.jp 628 Akira Kato 629 Keio University/WIDE Project 630 Graduate School of Media Design, 4-1-1 Hiyoshi 631 Kohoku, Yokohama 223-8526 632 Japan 634 Phone: +81 45 564 2490 635 Email: kato@wide.ad.jp