idnits 2.17.1 draft-ietf-dnsop-rfc7816bis-00.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 seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) == There are 9 instances of lines with non-RFC2606-compliant FQDNs in the document. -- 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 -- The document date (September 14, 2018) is 2051 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reference to an Informational RFC: RFC 6973 ** Obsolete normative reference: RFC 7816 (Obsoleted by RFC 9156) == Outdated reference: A later version (-03) exists of draft-wkumari-dnsop-hammer-01 == Outdated reference: A later version (-02) exists of draft-bortzmeyer-dprive-rfc7626-bis-01 Summary: 3 errors (**), 0 flaws (~~), 4 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group S. Bortzmeyer 3 Internet-Draft AFNIC 4 Obsoletes: 7816 (if approved) P. Hoffman 5 Intended status: Standards Track ICANN 6 Expires: March 18, 2019 September 14, 2018 8 DNS Query Name Minimisation to Improve Privacy 9 draft-ietf-dnsop-rfc7816bis-00 11 Abstract 13 This document describes a technique to improve DNS privacy, a 14 technique called "QNAME minimisation", where the DNS resolver no 15 longer sends the full original QNAME to the upstream name server. It 16 obsoletes RFC 7816. 18 This document is part of the IETF DNSOP (DNS Operations) Working 19 Group. The source of the document, as well as a list of open issues, 20 is at 22 Status of This Memo 24 This Internet-Draft is submitted in full conformance with the 25 provisions of BCP 78 and BCP 79. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF). Note that other groups may also distribute 29 working documents as Internet-Drafts. The list of current Internet- 30 Drafts is at https://datatracker.ietf.org/drafts/current/. 32 Internet-Drafts are draft documents valid for a maximum of six months 33 and may be updated, replaced, or obsoleted by other documents at any 34 time. It is inappropriate to use Internet-Drafts as reference 35 material or to cite them other than as "work in progress." 37 This Internet-Draft will expire on March 18, 2019. 39 Copyright Notice 41 Copyright (c) 2018 IETF Trust and the persons identified as the 42 document authors. All rights reserved. 44 This document is subject to BCP 78 and the IETF Trust's Legal 45 Provisions Relating to IETF Documents 46 (https://trustee.ietf.org/license-info) in effect on the date of 47 publication of this document. Please review these documents 48 carefully, as they describe your rights and restrictions with respect 49 to this document. Code Components extracted from this document must 50 include Simplified BSD License text as described in Section 4.e of 51 the Trust Legal Provisions and are provided without warranty as 52 described in the Simplified BSD License. 54 Table of Contents 56 1. Introduction and Background . . . . . . . . . . . . . . . . . 2 57 2. QNAME Minimisation . . . . . . . . . . . . . . . . . . . . . 3 58 3. Possible Issues . . . . . . . . . . . . . . . . . . . . . . . 4 59 4. Protocol and Compatibility Discussion . . . . . . . . . . . . 6 60 5. Operational Considerations . . . . . . . . . . . . . . . . . 6 61 6. Performance Considerations . . . . . . . . . . . . . . . . . 7 62 7. Results of the Experimentation . . . . . . . . . . . . . . . 7 63 8. Security Considerations . . . . . . . . . . . . . . . . . . . 8 64 9. Implementation status - RFC EDITOR: PLEASE REMOVE BEFORE 65 PUBLICATION . . . . . . . . . . . . . . . . . . . . . . . . . 8 66 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 67 10.1. Normative References . . . . . . . . . . . . . . . . . . 9 68 10.2. Informative References . . . . . . . . . . . . . . . . . 9 69 Appendix A. An Algorithm to Perform QNAME Minimisation . . . . . 10 70 Appendix B. Alternatives . . . . . . . . . . . . . . . . . . . . 11 71 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 12 72 Changes from RFC 7816 . . . . . . . . . . . . . . . . . . . . . . 12 73 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 12 75 1. Introduction and Background 77 The problem statement for this document and its predecessor [RFC7816] 78 is described in [I-D.bortzmeyer-dprive-rfc7626-bis]. The terminology 79 ("QNAME", "resolver", etc.) is defined in 80 [I-D.ietf-dnsop-terminology-bis]. This specific solution is not 81 intended to fully solve the DNS privacy problem; instead, it should 82 be viewed as one tool amongst many. 84 QNAME minimisation follows the principle explained in Section 6.1 of 85 [RFC6973]: the less data you send out, the fewer privacy problems 86 you have. 88 Before QNAME minimisation, when a resolver received the query "What 89 is the AAAA record for www.example.com?", it sent to the root 90 (assuming a cold resolver, whose cache is empty) the very same 91 question. Sending the full QNAME to the authoritative name server 92 was a tradition, not a protocol requirement. In a conversation with 93 the author in January 2015, Paul Mockapetris explained that this 94 tradition comes from a desire to optimise the number of requests, 95 when the same name server is authoritative for many zones in a given 96 name (something that was more common in the old days, where the same 97 name servers served .com and the root) or when the same name server 98 is both recursive and authoritative (something that is strongly 99 discouraged now). Whatever the merits of this choice at this time, 100 the DNS is quite different now. 102 2. QNAME Minimisation 104 The idea is to minimise the amount of data sent from the DNS resolver 105 to the authoritative name server. In the example in the previous 106 section, sending "What are the NS records for .com?" would have been 107 sufficient (since it will be the answer from the root anyway). The 108 rest of this section describes the recommended way to do QNAME 109 minimisation -- the way that maximises privacy benefits (other 110 alternatives are discussed in the appendices). 112 Instead of sending the full QNAME and the original QTYPE upstream, a 113 resolver that implements QNAME minimisation and does not already have 114 the answer in its cache sends a request to the name server 115 authoritative for the closest known ancestor of the original QNAME. 116 The request is done with: 118 o the QTYPE NS 120 o the QNAME that is the original QNAME, stripped to just one label 121 more than the zone for which the server is authoritative 123 For example, a resolver receives a request to resolve 124 foo.bar.baz.example. Let's assume that it already knows that 125 ns1.nic.example is authoritative for .example and the resolver does 126 not know a more specific authoritative name server. It will send the 127 query QTYPE=NS,QNAME=baz.example to ns1.nic.example. 129 The minimising resolver works perfectly when it knows the zone cut 130 (zone cuts are described in Section 6 of [RFC2181]). But zone cuts 131 do not necessarily exist at every label boundary. If we take the 132 name www.foo.bar.example, it is possible that there is a zone cut 133 between "foo" and "bar" but not between "bar" and "example". So, 134 assuming that the resolver already knows the name servers of 135 .example, when it receives the query "What is the AAAA record of 136 www.foo.bar.example?", it does not always know where the zone cut 137 will be. To find the zone cut, it will query the .example 138 name servers for the NS records for bar.example. It will get a 139 NODATA response, indicating that there is no zone cut at that point, 140 so it has to query the .example name servers again with one more 141 label, and so on. (Appendix A describes this algorithm in deeper 142 detail.) 144 Here are more detailed examples of queries with QNAME minimisation: 146 www.isc.org, cold cache, aggressive algorithm: 148 QTYPE QNAME TARGET NOTE 149 NS org root nameserver 150 NS isc.org Afilias nameserver 151 NS www.isc.org ISC nameserver "www" may be delegated 152 A www.isc.org ISC nameserver 154 www.isc.org, cold cache, lazy algorithm (for a cold cache, it is the 155 same algorithm as now): 157 QTYPE QNAME TARGET NOTE 158 A www.isc.org root nameserver 159 A www.isc.org Afilias nameserver 160 A www.isc.org ISC nameserver 162 www.isc.org, warm cache (all NS RRsets are known), both algorithms: 164 QTYPE QNAME TARGET NOTE 165 A www.isc.org ISC nameserver 167 www.example.org, warm cache (but for isc.org only, example.org's 168 NS RRset is not known), aggressive algorithm 170 QTYPE QNAME TARGET NOTE 171 NS example.org Afilias nameserver 172 NS www.example.org Example nameserver 173 A www.example.org Example nameserver 175 Since the information about the zone cuts will be stored in the 176 resolver's cache, the performance cost is probably reasonable. 177 Section 6 discusses this performance discrepancy further. 179 Note that DNSSEC-validating resolvers already have access to this 180 information, since they have to know the zone cut (the DNSKEY record 181 set is just below; the DS record set is just above). 183 3. Possible Issues 185 TODO may be remove the whole section now that it is no longer 186 experimental? 188 QNAME minimisation is legal, since the original DNS RFCs do not 189 mandate sending the full QNAME. So, in theory, it should work 190 without any problems. However, in practice, some problems may occur 191 (see [Huque-QNAME-Min] for an analysis and [Huque-QNAME-Discuss] for 192 an interesting discussion on this topic). 194 Some broken name servers do not react properly to QTYPE=NS requests. 195 For instance, some authoritative name servers embedded in load 196 balancers reply properly to A queries but send REFUSED to NS queries. 197 This behaviour is a protocol violation, and there is no need to stop 198 improving the DNS because of such behaviour. However, QNAME 199 minimisation may still work with such domains, since they are only 200 leaf domains (no need to send them NS requests). Such a setup breaks 201 more than just QNAME minimisation. It breaks negative answers, since 202 the servers don't return the correct SOA, and it also breaks anything 203 dependent upon NS and SOA records existing at the top of the zone. 205 Another way to deal with such incorrect name servers would be to try 206 with QTYPE=A requests (A being chosen because it is the most common 207 and hence a QTYPE that will always be accepted, while a QTYPE NS may 208 ruffle the feathers of some middleboxes). Instead of querying 209 name servers with a query "NS example.com", we could use 210 "A _.example.com" and see if we get a referral. TODO this is what 211 Unbound does 213 A problem can also appear when a name server does not react properly 214 to ENTs (Empty Non-Terminals). If ent.example.com has no resource 215 records but foobar.ent.example.com does, then ent.example.com is an 216 ENT. Whatever the QTYPE, a query for ent.example.com must return 217 NODATA (NOERROR / ANSWER: 0). However, some name servers incorrectly 218 return NXDOMAIN for ENTs. If a resolver queries only 219 foobar.ent.example.com, everything will be OK, but if it implements 220 QNAME minimisation, it may query ent.example.com and get an NXDOMAIN. 221 See also Section 3 of [DNS-Res-Improve] for the other bad 222 consequences of this bad behaviour. 224 A possible solution, currently implemented in Knot or Unbound, is to 225 retry with the full query when you receive an NXDOMAIN. It works, 226 but it is not ideal for privacy. 228 Other practices that do not conform to the DNS protocol standards may 229 pose a problem: there is a common DNS trick used by some web hosters 230 that also do DNS hosting that exploits the fact that the DNS protocol 231 (pre-DNSSEC) allows certain serious misconfigurations, such as parent 232 and child zones disagreeing on the location of a zone cut. 233 Basically, they have a single zone with wildcards for each TLD, like: 235 *.example. 60 IN A 192.0.2.6 237 (They could just wildcard all of "*.", which would be sufficient. We 238 don't know why they don't do it.) 240 This lets them have many web-hosting customers without having to 241 configure thousands of individual zones on their name servers. They 242 just tell the prospective customer to point their NS records at the 243 hoster's name servers, and the web hoster doesn't have to provision 244 anything in order to make the customer's domain resolve. NS queries 245 to the hoster will therefore not give the right result, which may 246 endanger QNAME minimisation (it will be a problem for DNSSEC, too). 248 TODO report by Akamai about why they return erroneous responses 249 https://mailarchive.ietf.org/arch/msg/dnsop/ 250 XIX16DCe2ln3ZnZai723v32ZIjE 252 4. Protocol and Compatibility Discussion 254 QNAME minimisation is compatible with the current DNS system and 255 therefore can easily be deployed; since it is a unilateral change to 256 the resolver, it does not change the protocol. (Because it is a 257 unilateral change, resolver implementers may do QNAME minimisation in 258 slightly different ways; see the appendices for examples.) 260 One should note that the behaviour suggested here (minimising the 261 amount of data sent in QNAMEs from the resolver) is NOT forbidden by 262 Section 5.3.3 of [RFC1034] or Section 7.2 of [RFC1035]. As stated in 263 Section 1, the current method, sending the full QNAME, is not 264 mandated by the DNS protocol. 266 One may notice that many documents that explain the DNS and that are 267 intended for a wide audience incorrectly describe the resolution 268 process as using QNAME minimisation (e.g., by showing a request going 269 to the root, with just the TLD in the query). As a result, these 270 documents may confuse readers that use them for privacy analysis. 272 5. Operational Considerations 274 TODO what to do if the resolver forwards? Unbound disables QNAME 275 minimisation in that case, since the forwarder will see everything, 276 anyway. What should a minimising resolver do when forwading the 277 request to a forwarder, not to an authoritative name server? Send 278 the full qname? Minimises? (But how since we do not know the zone 279 cut?) 281 The administrators of the forwarders, and of the authoritative 282 name servers, will get less data, which will reduce the utility of 283 the statistics they can produce (such as the percentage of the 284 various QTYPEs). 286 DNS administrators are reminded that the data on DNS requests that 287 they store may have legal consequences, depending on your 288 jurisdiction (check with your local lawyer). 290 6. Performance Considerations 292 The main goal of QNAME minimisation is to improve privacy by sending 293 less data. However, it may have other advantages. For instance, if 294 a root name server receives a query from some resolver for A.example 295 followed by B.example followed by C.example, the result will be three 296 NXDOMAINs, since .example does not exist in the root zone. Under 297 query name minimisation, the root name servers would hear only one 298 question (for .example itself) to which they could answer NXDOMAIN, 299 thus opening up a negative caching opportunity in which the full 300 resolver could know a priori that neither B.example nor C.example 301 could exist. Thus, in this common case the total number of upstream 302 queries under QNAME minimisation would be counterintuitively less 303 than the number of queries under the traditional iteration (as 304 described in the DNS standard). TODO mention [RFC8020]? And 305 [RFC8198], the latter depending on DNSSEC? 307 QNAME minimisation may also improve lookup performance for TLD 308 operators. For a typical TLD, delegation-only, and with delegations 309 just under the TLD, a two-label QNAME query is optimal for finding 310 the delegation owner name. 312 QNAME minimisation can decrease performance in some cases -- for 313 instance, for a deep domain name (like 314 www.host.group.department.example.com, where 315 host.group.department.example.com is hosted on example.com's 316 name servers). Let's assume a resolver that knows only the 317 name servers of example.com. Without QNAME minimisation, it would 318 send these example.com name servers a query for 319 www.host.group.department.example.com and immediately get a specific 320 referral or an answer, without the need for more queries to probe for 321 the zone cut. For such a name, a cold resolver with QNAME 322 minimisation will, depending on how QNAME minimisation is 323 implemented, send more queries, one per label. Once the cache is 324 warm, there will be no difference with a traditional resolver. 325 Actual testing is described in [Huque-QNAME-Min]. Such deep domains 326 are especially common under ip6.arpa. 328 7. Results of the Experimentation 330 TODO various experiences from actual deployments, problems heard. 331 TODO the Knot bug #339 https://gitlab.labs.nic.cz/knot/knot-resolver/ 332 issues/339? TODO Problems with AWS https://forums.aws.amazon.com/ 333 thread.jspa?threadID=269116? 335 8. Security Considerations 337 QNAME minimisation's benefits are clear in the case where you want to 338 decrease exposure to the authoritative name server. But minimising 339 the amount of data sent also, in part, addresses the case of a wire 340 sniffer as well as the case of privacy invasion by the servers. 341 (Encryption is of course a better defense against wire sniffers, but, 342 unlike QNAME minimisation, it changes the protocol and cannot be 343 deployed unilaterally. Also, the effect of QNAME minimisation on 344 wire sniffers depends on whether the sniffer is on the DNS path.) 346 QNAME minimisation offers zero protection against the recursive 347 resolver, which still sees the full request coming from the stub 348 resolver. 350 All the alternatives mentioned in Appendix B decrease privacy in the 351 hope of improving performance. They must not be used if you want 352 maximum privacy. 354 9. Implementation status - RFC EDITOR: PLEASE REMOVE BEFORE PUBLICATION 356 This section records the status of known implementations of the 357 protocol defined by this specification at the time of posting of this 358 Internet-Draft, and is based on a proposal described in [RFC7942]. 359 The description of implementations in this section is intended to 360 assist the IETF in its decision processes in progressing drafts to 361 RFCs. Please note that the listing of any individual implementation 362 here does not imply endorsement by the IETF. Furthermore, no effort 363 has been spent to verify the information presented here that was 364 supplied by IETF contributors. This is not intended as, and must not 365 be construed to be, a catalog of available implementations or their 366 features. Readers are advised to note that other implementations may 367 exist. 369 According to [RFC7942], "this will allow reviewers and working groups 370 to assign due consideration to documents that have the benefit of 371 running code, which may serve as evidence of valuable experimentation 372 and feedback that have made the implemented protocols more mature. 373 It is up to the individual working groups to use this information as 374 they see fit". 376 Unbound has QNAME minimisation for several years, and it is now the 377 default. It has two modes, strict (no workaround for broken 378 authoritative name servers) and "lax" (retries when there is a 379 NXDOMAIN). TODO Ralph Dolmans talk at OARC https://indico.dns- 380 oarc.net/event/22/contributions/332/attachments/310/542/ 381 unbound_qnamemin_oarc24.pdf 382 Knot resolver also has QNAME minimisation since 2016, and it is 383 activated by default. 385 BIND has QNAME minimisation since BIND 9.13.2, released in july 2018. 386 Like Unbound, it has several modes, with or without workarounds for 387 broken authoritative name servers. 389 PowerDNS does not have QNAME minimisation. TODO 390 https://github.com/PowerDNS/pdns/issues/2311 392 The public DNS resolver at Cloudflare ("1.1.1.1") has QNAME 393 minimisation (it uses Knot). 395 10. References 397 10.1. Normative References 399 [RFC1034] Mockapetris, P., "Domain names - concepts and facilities", 400 STD 13, RFC 1034, DOI 10.17487/RFC1034, November 1987, 401 . 403 [RFC1035] Mockapetris, P., "Domain names - implementation and 404 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 405 November 1987, . 407 [RFC6973] Cooper, A., Tschofenig, H., Aboba, B., Peterson, J., 408 Morris, J., Hansen, M., and R. Smith, "Privacy 409 Considerations for Internet Protocols", RFC 6973, 410 DOI 10.17487/RFC6973, July 2013, 411 . 413 [RFC7816] Bortzmeyer, S., "DNS Query Name Minimisation to Improve 414 Privacy", RFC 7816, DOI 10.17487/RFC7816, March 2016, 415 . 417 10.2. Informative References 419 [DNS-Res-Improve] 420 Vixie, P., Joffe, R., and F. Neves, "Improvements to DNS 421 Resolvers for Resiliency, Robustness, and Responsiveness", 422 Work in Progress, draft-vixie-dnsext-resimprove-00, June 423 2010. 425 [HAMMER] Kumari, W., Arends, R., Woolf, S., and D. Migault, "Highly 426 Automated Method for Maintaining Expiring Records", Work 427 in Progress, draft-wkumari-dnsop-hammer-01, July 2014. 429 [Huque-QNAME-Discuss] 430 Huque, S., "Qname Minimization @ DNS-OARC", May 2015, 431 . 433 [Huque-QNAME-Min] 434 Huque, S., "Query name minimization and authoritative 435 server behavior", May 2015, 436 . 438 [I-D.bortzmeyer-dprive-rfc7626-bis] 439 Bortzmeyer, S. and S. Dickinson, "DNS Privacy 440 Considerations", draft-bortzmeyer-dprive-rfc7626-bis-01 441 (work in progress), July 2018. 443 [I-D.ietf-dnsop-terminology-bis] 444 Hoffman, P., Sullivan, A., and K. Fujiwara, "DNS 445 Terminology", draft-ietf-dnsop-terminology-bis-14 (work in 446 progress), September 2018. 448 [RFC2181] Elz, R. and R. Bush, "Clarifications to the DNS 449 Specification", RFC 2181, DOI 10.17487/RFC2181, July 1997, 450 . 452 [RFC7942] Sheffer, Y. and A. Farrel, "Improving Awareness of Running 453 Code: The Implementation Status Section", BCP 205, 454 RFC 7942, DOI 10.17487/RFC7942, July 2016, 455 . 457 [RFC8020] Bortzmeyer, S. and S. Huque, "NXDOMAIN: There Really Is 458 Nothing Underneath", RFC 8020, DOI 10.17487/RFC8020, 459 November 2016, . 461 [RFC8198] Fujiwara, K., Kato, A., and W. Kumari, "Aggressive Use of 462 DNSSEC-Validated Cache", RFC 8198, DOI 10.17487/RFC8198, 463 July 2017, . 465 Appendix A. An Algorithm to Perform QNAME Minimisation 467 This algorithm performs name resolution with QNAME minimisation in 468 the presence of zone cuts that are not yet known. 470 Although a validating resolver already has the logic to find the 471 zone cuts, implementers of other resolvers may want to use this 472 algorithm to locate the cuts. This is just a possible aid for 473 implementers; it is not intended to be normative: 475 (0) If the query can be answered from the cache, do so; otherwise, 476 iterate as follows: 478 (1) Find the closest enclosing NS RRset in your cache. The owner of 479 this NS RRset will be a suffix of the QNAME -- the longest suffix 480 of any NS RRset in the cache. Call this ANCESTOR. 482 (2) Initialise CHILD to the same as ANCESTOR. 484 (3) If CHILD is the same as the QNAME, resolve the original query 485 using ANCESTOR's name servers, and finish. 487 (4) Otherwise, add a label from the QNAME to the start of CHILD. 489 (5) If you have a negative cache entry for the NS RRset at CHILD, go 490 back to step 3. 492 (6) Query for CHILD IN NS using ANCESTOR's name servers. The 493 response can be: 495 (6a) A referral. Cache the NS RRset from the authority section, 496 and go back to step 1. 498 (6b) An authoritative answer. Cache the NS RRset from the 499 answer section, and go back to step 1. 501 (6c) An NXDOMAIN answer. Return an NXDOMAIN answer in response 502 to the original query, and stop. 504 (6d) A NOERROR/NODATA answer. Cache this negative answer, and 505 go back to step 3. 507 Appendix B. Alternatives 509 Remember that QNAME minimisation is unilateral, so a resolver is not 510 forced to implement it exactly as described here. 512 There are several ways to perform QNAME minimisation. See Section 2 513 for the suggested way. It can be called the aggressive algorithm, 514 since the resolver only sends NS queries as long as it does not know 515 the zone cuts. This is the safest, from a privacy point of view. 516 Another possible algorithm, not fully studied at this time, could be 517 to "piggyback" on the traditional resolution code. At startup, it 518 sends traditional full QNAMEs and learns the zone cuts from the 519 referrals received, then switches to NS queries asking only for the 520 minimum domain name. This leaks more data but could require fewer 521 changes in the existing resolver codebase. 523 In the above specification, the original QTYPE is replaced by NS (or 524 may be A, if too many servers react incorrectly to NS requests); this 525 is the best approach to preserve privacy. But this erases 526 information about the relative use of the various QTYPEs, which may 527 be interesting for researchers (for instance, if they try to follow 528 IPv6 deployment by counting the percentage of AAAA vs. A queries). A 529 variant of QNAME minimisation would be to keep the original QTYPE. 531 Another useful optimisation may be, in the spirit of the HAMMER idea 532 [HAMMER], to probe in advance for the introduction of zone cuts where 533 none previously existed (i.e., confirm their continued absence, or 534 discover them). 536 To address the "number of queries" issue described in Section 6, a 537 possible solution is to always use the traditional algorithm when the 538 cache is cold and then to move to QNAME minimisation (precisely 539 defining what is "hot" or "cold" is left to the implementer). This 540 will decrease the privacy but will guarantee no degradation of 541 performance. 543 Acknowledgments 545 TODO (refer to 7816) 547 Changes from RFC 7816 549 o Made changes to deal with Errata #4644 551 o Changed status to be on standards track 553 Authors' Addresses 555 Stephane Bortzmeyer 556 AFNIC 557 1, rue Stephenson 558 Montigny-le-Bretonneux 78180 559 France 561 Phone: +33 1 39 30 83 46 562 Email: bortzmeyer+ietf@nic.fr 563 URI: https://www.afnic.fr/ 565 Paul Hoffman 566 ICANN 568 Email: paul.hoffman@icann.org