dnsop | P. Wouters |
Internet-Draft | Red Hat |
Obsoletes: 6944 (if approved) | O. Sury |
Intended status: Standards Track | Internet Systems Consortium |
Expires: December 7, 2018 | June 5, 2018 |
Algorithm Implementation Requirements and Usage Guidance for DNSSEC
draft-ietf-dnsop-algorithm-update-01
The DNSSEC protocol makes use of various cryptographic algorithms in order to provide authentication of DNS data and proof of non-existence. To ensure interoperability between DNS resolvers and DNS authoritative servers, it is necessary to specify a set of algorithm implementation requirements and usage guidance to ensure that there is at least one algorithm that all implementations support. This document defines the current algorithm implementation requirements and usage guidance for DNSSEC. This document obsoletes [RFC6944].
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
This Internet-Draft will expire on December 7, 2018.
Copyright (c) 2018 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
The DNSSEC signing algorithms are defined by various RFCs, including [RFC4034], [RFC5155], [RFC5702], [RFC5933], [RFC6605], [RFC8080]. DNSSEC is used to provide authentication of data. To ensure interoperability, a set of "mandatory-to-implement" DNSKEY algorithms are defined. This document obsoletes [RFC6944].
The field of cryptography evolves continuously. New stronger algorithms appear and existing algorithms are found to be less secure then originally thought. Therefore, algorithm implementation requirements and usage guidance need to be updated from time to time to reflect the new reality. The choices for algorithms must be conservative to minimize the risk of algorithm compromise.
The mandatory-to-implement algorithm of tomorrow should already be available in most implementations of DNSSEC by the time it is made mandatory. This document attempts to identify and introduce those algorithms for future mandatory-to-implement status. There is no guarantee that the algorithms in use today may become mandatory in the future. Published algorithms are continuously subjected to cryptographic attack and may become too weak or could become completely broken before this document is updated.
This document only provides recommendations for the mandatory-to-implement algorithms or algorithms too weak that are recommended not to be implemented. As a result, any algorithm listed at the [DNSKEY-IANA] and [DS-IANA] registries not mentioned in this document MAY be implemented. For clarification and consistency, an algorithm will be set to MAY only when it has been downgraded.
Although this document updates the algorithms to keep the DNSSEC authentication secure over time, it also aims at providing recommendations so that DNSSEC implementations remain interoperable. DNSSEC interoperability is addressed by an incremental introduction or deprecation of algorithms.
While [RFC2119] consider term SHOULD equivalent to RECOMMENDED, and term SHOULD NOT to NOT RECOMMENDED, the authors of this document has chosen to use terms RECOMMENDED and NOT RECOMMENDED, as it better reflects the recommendations for implementations.
It is expected that deprecation of an algorithm is performed gradually. This provides time for various implementations to update their implemented algorithms while remaining interoperable. Unless there are strong security reasons, an algorithm is expected to be downgraded from MUST to NOT RECOMMENDED or MAY, instead of MUST NOT. Similarly, an algorithm that has not been mentioned as mandatory-to-implement is expected to be introduced with a RECOMMENDED instead of a MUST.
Since the effects of using an unknown DNSKEY algorithm is for the zone to be treated as insecure, it is recommended that algorithms downgraded to NOT RECOMMENDED or below are no longer used by authoritative nameservers and DNSSEC signers to create new DNSKEY's. This will allow for algorithms to slowly become more unused over time. Once deployment has reached a sufficiently low point these algorithms can finally be marked as MUST NOT so that recursive nameservers can remove support for these algorithms.
Recursive nameservers are encouraged to keep support for all algorithms not marked as MUST NOT.
The recommendations of this document mostly target DNSSEC implementers as implementations need to meet both high security expectations as well as high interoperability between various vendors and with different versions. Interoperability requires a smooth move to more secure algorithms. This may differ from a user point of view that may deploy and configure DNSSEC with only the safest algorithm. On the other hand, comments and recommendations from this document are also expected to be useful for such users.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119].
Implemenation recommendations for DNSKEY algorithms [DNSKEY-IANA].
Number | Mnemonics | DNSSEC Signing | DNSSEC Validation |
---|---|---|---|
1 | RSAMD5 | MUST NOT | MUST NOT |
3 | DSA | MUST NOT | MUST NOT |
5 | RSASHA1 | NOT RECOMMENDED | MUST |
6 | DSA-NSEC3-SHA1 | MUST NOT | MUST NOT |
7 | RSASHA1-NSEC3-SHA1 | NOT RECOMMENDED | MUST |
8 | RSASHA256 | MUST | MUST |
10 | RSASHA512 | NOT RECOMMENDED | MUST |
12 | ECC-GOST | MUST NOT | MAY |
13 | ECDSAP256SHA256 | MUST | MUST |
14 | ECDSAP384SHA384 | MAY | RECOMMENDED |
15 | ED25519 | RECOMMENDED | RECOMMENDED |
16 | ED448 | MAY | RECOMMENDED |
RSAMD5 is not widely deployed and there is an industry-wide trend to deprecate MD5 usage.
RSASHA1 and RSASHA1-NSEC3-SHA1 are widely deployed, although zones deploying it are recommended to switch to ECDSAP256SHA256 as there is an industry-wide trend to move to elliptic curve cryptography. RSASHA1 does not support NSEC3. RSASHA1-NSEC3-SHA1 can be used with or without NSEC3.
DSA and DSA-NSEC3-SHA1 are not widely deployed and vulnerable to private key compromise when generating signatures using a weak or compromised random number generator.
RSASHA256 is in wide use and considered strong.
RSASHA512 is NOT RECOMMENDED for DNSSEC Signing because it has not seen wide deployment, but there are some deployments hence DNSSEC Validation MUST implement RSASHA512 to ensure interoperability. There's isn't significant difference in cryptographics strength between RSASHA512 and RSASHA256, therefore it is discouraged to use RSASHA512, as it will only make deprecation of older algorithms harder. People that want to use cryptographically stronger algorithm should switch to elliptic curve cryptography algorithms.
ECC-GOST (GOST R 34.11-94) has been superseded by GOST R 34.11-2012 in [RFC6986]. The GOST R 34.11-2012 hasn't been standardized for use in DNSSEC.
ECDSAP256SHA256 provide more strength for signature size than RSASHA256 and RSASHA512 variants. ECDSAP256SHA256 has been widely deployed and therefore it is now at MUST level for both validation and signing. It is RECOMMENDED to use deterministic digital signature generation procedure of the ECDSA ([RFC6979]) when implementing ECDSAP256SHA256 (and ECDSAP384SHA384).
ECDSAP384SHA384 share the same properties as ECDSAP256SHA256, but offers a modest security advantage over ECDSAP256SHA256 (192-bits of strength versus 128-bits). For most applications of DNSSEC, ECDSAP256SHA256 should be satisfactory and robust for the foreseeable future, and is therefore recommended for signing. While it is unlikely for a DNSSEC use case which requires 192-bit security strength to arise, ECDSA384SHA384 is provided for such applications and it MAY be used for signing in these cases.
ED25519 and ED448 uses Edwards-curve Digital Security Algorithm (EdDSA). There are three main advantages of the EdDSA algorithm: It does not require the use of a unique random number for each signature, there are no padding or truncation issues as with ECDSA, and it is more resilient to side-channel attacks. Furthermore, EdDSA cryptography is less prone to implementation errors ([RFC8032], [RFC8080]). It is expected that ED25519 will become the future RECOMMENDED default algorithm once there's enough support for this algorithm in the deployed DNSSEC validators.
Operation recommendation for new and existing deployments.
Due to industry-wide trend to move to elliptic curve cryptography, the ECDSAP256SHA256 is RECOMMENDED to be used by new DNSSEC deployments, and users of RSA based algorithms SHOULD upgrade to ECDSAP256SHA256.
Recommendations for Delegation Signer Digest Algorithms [DNSKEY-IANA] These also apply to the CDS RRTYPE as specified in [RFC7344]
Number | Mnemonics | DNSSEC Delegation | DNSSEC Validation |
---|---|---|---|
0 | NULL (CDS only) | MUST NOT [*] | MUST NOT [*] |
1 | SHA-1 | MUST NOT | MUST |
2 | SHA-256 | MUST | MUST |
3 | GOST R 34.11-94 | MUST NOT | MAY |
4 | SHA-384 | MAY | RECOMMENDED |
[*] - This is a special type of CDS record signaling removal of DS at the parent in [RFC8078]
NULL is a special case, see [RFC8078]
SHA-1 is still in wide use for DS records, so validators MUST implement the validation, but it is disallowed to use SHA-1 to generate new DS records. (See Operational Considerations for caveats when upgrading from SHA-1 to SHA-256 DS Algorithm.)
SHA-256 is in wide use and considered strong.
GOST R 34.11-94 has been deprecated by [RFC6986].
SHA-384 share the same properties as SHA-256, but offers a modest security advantage over SHA-384 (384-bits of strength versus 256-bits). For most applications of DNSSEC, SHA-256 should be satisfactory and robust for the foreseeable future, and is therefore recommended for DS/CDS records. While it is unlikely for a DNSSEC use case which requires 384-bit security strength to arise, SHA-384 is provided for such applications and it MAY be used for generating DS/CDS records in these cases.
The security of cryptographic-based systems depends on both the strength of the cryptographic algorithms chosen and the strength of the keys used with those algorithms. The security also depends on the engineering of the protocol used by the system to ensure that there are no non-cryptographic ways to bypass the security of the overall system.
This document concerns itself with the selection of cryptographic algorithms for the use of DNSSEC, specifically with the selection of "mandatory-to-implement" algorithms. The algorithms identified in this document as MUST or RECOMMENDED to implement are not known to be broken at the current time, and cryptographic research so far leads us to believe that they will likely remain secure into the foreseeable future. However, this isn't necessarily forever and it is expected that new revisions of this document will be issued from time to time to reflect the current best practice in this area.
Retiring an algorithm too soon would result in a signed zone with such an algorithm to be downgraded to the equivalent of an unsigned zone. Therefore, algorithm deprecation must be done very slowly and only after careful consideration and measurements of its use.
DNSKEY algorithm rollover in a live zone is a complex process. See [RFC6781] and [RFC7583] for guidelines on how to perform algorithm rollovers.
DS algorithm rollover in a live zone is also a complex process. Upgrading algorithm at the same time as rolling the new KSK key will lead to DNSSEC validation failures, and users MUST upgrade the DS algorithm first before rolling the Key Signing Key.
This document makes no requests of IANA.
This document borrows text from RFC 4307 by Jeffrey I. Schiller of the Massachusetts Institute of Technology (MIT) and the 4307bis document by Yoav Nir, Tero Kivinen, Paul Wouters and Daniel Migault. Much of the original text has been copied verbatim.
We wish to thank Michael Sinatra, Roland van Rijswijk-Deij, Olafur Gudmundsson and Paul Hoffman for their imminent feedback.
Kudos to Roy Arends for bringing the DS rollover issue to the daylight.
[RFC2119] | Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997. |