idnits 2.17.1 draft-hardaker-dnsop-nsec3-guidance-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 (6 May 2021) is 1079 days in the past. Is this intentional? Checking references for intended status: Best Current Practice ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'TBD' is mentioned on line 237, but not defined Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group W. Hardaker 3 Internet-Draft USC/ISI 4 Intended status: Best Current Practice V. Dukhovni 5 Expires: 7 November 2021 Bloomberg, L.P. 6 6 May 2021 8 Guidance for NSEC3 parameter settings 9 draft-hardaker-dnsop-nsec3-guidance-03 11 Abstract 13 NSEC3 is a DNSSEC mechanism providing proof of non-existence by 14 promising there are no names that exist between two domainnames 15 within a zone. Unlike its counterpart NSEC, NSEC3 avoids directly 16 disclosing the bounding domainname pairs. This document provides 17 guidance on setting NSEC3 parameters based on recent operational 18 deployment experience. 20 Status of This Memo 22 This Internet-Draft is submitted in full conformance with the 23 provisions of BCP 78 and BCP 79. 25 Internet-Drafts are working documents of the Internet Engineering 26 Task Force (IETF). Note that other groups may also distribute 27 working documents as Internet-Drafts. The list of current Internet- 28 Drafts is at https://datatracker.ietf.org/drafts/current/. 30 Internet-Drafts are draft documents valid for a maximum of six months 31 and may be updated, replaced, or obsoleted by other documents at any 32 time. It is inappropriate to use Internet-Drafts as reference 33 material or to cite them other than as "work in progress." 35 This Internet-Draft will expire on 7 November 2021. 37 Copyright Notice 39 Copyright (c) 2021 IETF Trust and the persons identified as the 40 document authors. All rights reserved. 42 This document is subject to BCP 78 and the IETF Trust's Legal 43 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 44 license-info) in effect on the date of publication of this document. 45 Please review these documents carefully, as they describe your rights 46 and restrictions with respect to this document. Code Components 47 extracted from this document must include Simplified BSD License text 48 as described in Section 4.e of the Trust Legal Provisions and are 49 provided without warranty as described in the Simplified BSD License. 51 Table of Contents 53 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 54 1.1. Requirements notation . . . . . . . . . . . . . . . . . . 3 55 2. Recommendation for zone publishers . . . . . . . . . . . . . 3 56 2.1. Algorithms . . . . . . . . . . . . . . . . . . . . . . . 3 57 2.2. Flags . . . . . . . . . . . . . . . . . . . . . . . . . . 3 58 2.3. Iterations . . . . . . . . . . . . . . . . . . . . . . . 3 59 2.4. Salt . . . . . . . . . . . . . . . . . . . . . . . . . . 4 60 3. Best-practice for zone publishers . . . . . . . . . . . . . . 5 61 4. Recommendation for validating resolvers . . . . . . . . . . . 5 62 5. Security Considerations . . . . . . . . . . . . . . . . . . . 6 63 6. Operational Considerations . . . . . . . . . . . . . . . . . 6 64 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6 65 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 6 66 8.1. Normative References . . . . . . . . . . . . . . . . . . 6 67 8.2. Informative References . . . . . . . . . . . . . . . . . 7 68 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . 7 69 Appendix B. Github Version of this document . . . . . . . . . . 7 70 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 7 72 1. Introduction 74 As with NSEC [RFC4035], NSEC3 [RFC5155] provides proof of non- 75 existence that consists of signed DNS records establishing the non- 76 existence of a given name or associated Resource Record Type (RRTYPE) 77 in a DNSSEC [RFC4035] signed zone. In the case of NSEC3, however, 78 the names of valid nodes in the zone are obfuscated through (possibly 79 multiple iterations of) hashing via SHA-1. (currently only SHA-1 is 80 in use within the Internet). 82 NSEC3 also provides "opt-out support", allowing for blocks of 83 unsigned delegations to be covered by a single NSEC3 record. Use of 84 the opt-out feature allow large registries to only sign as many NSEC3 85 records as there are signed DS or other RRsets in the zone - with 86 opt-out, unsigned delegations don't require additional NSEC3 records. 87 This sacrifices the tamper-resistance proof of non-existence offered 88 by NSEC3 in order to reduce memory and CPU overheads. 90 NSEC3 records have a number of tunable parameters that are specified 91 via an NSEC3PARAM record at the zone apex. These parameters are the 92 Hash Algorithm, processing Flags, the number of hash Iterations and 93 the Salt. Each of these has security and operational considerations 94 that impact both zone owners and validating resolvers. This document 95 provides some best-practice recommendations for setting the NSEC3 96 parameters. 98 1.1. Requirements notation 100 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 101 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 102 "OPTIONAL" in this document are to be interpreted as described in BCP 103 14 [RFC2119] [RFC8174] when, and only when, they appear in all 104 capitals, as shown here. 106 2. Recommendation for zone publishers 108 The following sections describe recommendations for setting 109 parameters for NSEC3 and NSEC3PARAM. 111 2.1. Algorithms 113 The algorithm field is not discussed by this document. 115 2.2. Flags 117 The NSEC3PARAM flags field currently contains no flags, but 118 individual NSEC3 records contain the "Opt-Out" flag [RFC5155], which 119 specifies whether or not that NSEC3 record provides proof of non- 120 existence or not. In general, NSEC3 with the Opt-Out flag enabled 121 should only be used in large, highly dynamic zones with a small 122 percentage of signed delegations. Operationally, this allows for 123 fewer signature creations when new delegations are inserted into a 124 zone. This is typically only necessary for extremely large 125 registration points providing zone updates faster than real-time 126 signing allows. Smaller zones, or large but relatively static zones, 127 are encouraged to use a Flags value of 0 (zero) and take advantage of 128 DNSSEC's proof-of-non-existence support. 130 2.3. Iterations 132 NSEC3 records are created by first hashing the input domain and then 133 repeating that hashing algorithm a number of times based on the 134 iterations parameter in the NSEC3PARM and NSEC3 records. The first 135 hash is typically sufficient to discourage zone enumeration performed 136 by "zone walking" an NSEC or NSEC3 chain. Only determined parties 137 with significant resources are likely to try and uncover hashed 138 values, regardless of the number of additional iterations performed. 139 If an adversary really wants to expend significant CPU resources to 140 mount an offline dictionary attack on a zone's NSEC3 chain, they'll 141 likely be able to find most of the "guessable" names despite any 142 level of additional hashing iterations. 144 Most names published in the DNS are rarely secret or unpredictable. 145 They are published to be memorable, used and consumed by humans. 146 They are often recorded in many other network logs such as email 147 logs, certificate transparency logs, web page links, intrusion 148 detection systems, malware scanners, email archives, etc. Many times 149 a simple dictionary of commonly used domain names prefixes (www, ftp, 150 mail, imap, login, database, etc) can be used to quickly reveal a 151 large number of labels within a zone. Because of this, there are 152 increasing performance costs yet diminishing returns associated with 153 applying additional hash iterations beyond the first. 155 Although Section 10.3 of [RFC5155] specifies upper bounds for the 156 number of hash iterations to use, there is no published guidance for 157 zone owners about good values to select. Because hashing provides 158 only moderate protection, as shown recently in academic studies of 159 NSEC3 protected zones [GPUNSEC3][ZONEENUM], this document recommends 160 that zone owners SHOULD use an iteration value of 0 (zero), 161 indicating that only the initial hash value should be placed into a 162 DNS zone's NSEC3 records. 164 TODO: discuss the authoritative overhead of needing to find the right 165 range for new random strings coming in. Note white-lies online 166 signing differences. 168 2.4. Salt 170 Salts add yet another layer of protection against offline, stored 171 dictionary attacks by combining the value to be hashed (in our case, 172 a DNS domainname) with a randomly generated value. This prevents 173 adversaries from building up and remembering a dictionary of values 174 that can translate a hash output back to the value that it derived 175 from. 177 In the case of DNS, it should be noted the hashed names placed in 178 NSEC3 records already include the fully-qualified domain name from 179 each zone. Thus, no single pre-computed table works to speed up 180 dictionary attacks against multiple target zones. An attacker is 181 required to compute a complete dictionary per zone, which is 182 expensive in both storage and CPU time. 184 To protect against a dictionary being built and used for a target 185 zone, an additional salt field can be included and changed on a 186 regular basis, forcing a would-be attacker to repeatedly compute a 187 new dictionary (or just do trial and error without the benefits of 188 precomputation). 190 Changing a zone's salt value requires the construction of a complete 191 new NSEC3 chain. This is true both when resigning the entire zone at 192 once, or incrementally signing it in the background where the new 193 salt is only activated once every name in the chain has been 194 completed. 196 Most users of NSEC3 publish static salt values that never change. 197 This provides no added security benefit (because the complete fully 198 qualified domain name is already unique). If no rotation is planned, 199 operators are encouraged to forgo the salt entirely by using a zero- 200 length salt value instead (represented as a "-" in the presentation 201 format). 203 3. Best-practice for zone publishers 205 In short, for all zones, the recommended NSEC3 parameters are as 206 shown below: 208 ; SHA-1, no extra iterations, empty salt: 209 ; 210 bcp.example. IN NSEC3PARAM 1 0 0 - 212 For small zones, the use of opt-out based NSEC3 records is NOT 213 RECOMMENDED. 215 For very large and sparsely signed zones, where the majority of the 216 records are insecure delegations, opt-out MAY be used. 218 4. Recommendation for validating resolvers 220 Because there has been a large growth of open (public) DNSSEC 221 validating resolvers that are subject to compute resource constraints 222 when handling requests from anonymous clients, this document 223 recommends that validating resolvers should change their behavior 224 with respect to large iteration values. Validating resolvers SHOULD 225 return an insecure response when processing NSEC3 records with 226 iterations larger than 100. Validating resolvers MAY return SERVFAIL 227 when processing NSEC3 records with iterations larger than 500. Note 228 that this significantly decreases the requirements originally 229 specified in Section 10.3 of [RFC5155]. 231 Note that a validating resolver MUST still validate the signature 232 over the NSEC3 record to ensure the iteration count was not altered 233 since record publication (see [RFC5155] section 10.3). 235 Validating resolvers returning an insecure or SERVFAIL answer in this 236 situation SHOULD return an Extended DNS Error (EDE) {RFC8914} EDNS0 237 option of value [TBD]. 239 5. Security Considerations 241 This entire document discusses security considerations with various 242 parameters selections of NSEC3 and NSEC3PARAM fields. 244 6. Operational Considerations 246 This entire document discusses operational considerations with 247 various parameters selections of NSEC3 and NSEC3PARAM fields. 249 7. IANA Considerations 251 This document requests a new allocation in the "Extended DNS Error 252 Codes" of the "Domain Name System (DNS) Parameters" registration 253 table with the following characteristics: 255 * INFO-CODE: TBD 257 * Purpose: Unsupported NSEC3 iterations value 259 * Reference: this document 261 8. References 263 8.1. Normative References 265 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 266 Requirement Levels", BCP 14, RFC 2119, 267 DOI 10.17487/RFC2119, March 1997, 268 . 270 [RFC4035] Arends, R., Austein, R., Larson, M., Massey, D., and S. 271 Rose, "Protocol Modifications for the DNS Security 272 Extensions", RFC 4035, DOI 10.17487/RFC4035, March 2005, 273 . 275 [RFC5155] Laurie, B., Sisson, G., Arends, R., and D. Blacka, "DNS 276 Security (DNSSEC) Hashed Authenticated Denial of 277 Existence", RFC 5155, DOI 10.17487/RFC5155, March 2008, 278 . 280 8.2. Informative References 282 [GPUNSEC3] Wander, M., Schwittmann, L., Boelmann, C., and T. Weis, 283 "GPU-Based NSEC3 Hash Breaking", DOI 10.1109/NCA.2014.27, 284 2014, . 286 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 287 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 288 May 2017, . 290 [ZONEENUM] Wang, Z., Xiao, L., and R. Wang, "An efficient DNSSEC zone 291 enumeration algorithm", n.d.. 293 Appendix A. Acknowledgments 295 dns-operations discussion participants 297 Appendix B. Github Version of this document 299 While this document is under development, it can be viewed, tracked, 300 issued, pushed with PRs, ... here: 302 https://github.com/hardaker/draft-hardaker-dnsop-nsec3-guidance 304 Authors' Addresses 306 Wes Hardaker 307 USC/ISI 309 Email: ietf@hardakers.net 311 Viktor Dukhovni 312 Bloomberg, L.P. 314 Email: ietf-dane@dukhovni.org