Network Working Group                                         S. Leonard
Internet-Draft                                             Penango, Inc.
Intended status: Standards Track                            July 8, 2012
Expires: January 9, 2013

        A Uniform Resource Name (URN) Namespace for Certificates


   Digital certificates are used in many systems and protocols to
   identify and authenticate parties.  This document describes a Uniform
   Resource Name (URN) namespace that identifies certificates.  These
   URNs can be used when certificates need to be identified by value or
   reference.  Applications can also use this specification in non-URI

Status of this Memo

   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

   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 January 9, 2013.

Copyright Notice

   Copyright (c) 2012 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
   ( 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

Leonard                  Expires January 9, 2013                [Page 1]

Internet-Draft                  certspec                       July 2012

   described in the Simplified BSD License.

1.  Introduction

   Digital certificates are used in many systems and protocols to
   identify and authenticate parties.  Security considerations
   frequently require that the certificate must be identified with
   certainty, because selecting the wrong certificate will lead to
   validation errors (resulting in denial of service) or improper
   credential selection (resulting in unwanted disclosure or
   substitution attacks).  The goal of this namespace is to provide a
   uniform syntax for identifying certificates with precision in Uniform
   Resource Identifiers (URIs), specifically Uniform Resource Names

   Using this syntax, any protocol or system that refers to a
   certificate in a textual format can unambiguously identify that
   certificate by value or reference.  Implementers that parse these
   URNs can resolve them into actual certificates.  Examples:


1.1.  Requirements Language

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   document are to be interpreted as described in RFC 2119.

2.  Motivation and Purpose

   Although certificates have diverse applications, there has been
   traditionally no way to refer to a certificate uniformly and
   unambiguously by reference in text.  Certificates that identify long
   public keys (e.g., 2048-bit RSA keys) and that contain required and
   recommended PKIX extensions can easily exceed many kilobytes in
   length, which are impractical for certain applications to store by

   The purpose of this specification is to provide a uniform textual
   format for identifying individual certficates.  When a resolver
   resolves a certificate specification, the resolver's output is either
   a single certficiate or nothing.  This specification is not designed
   or intended to provide a search tool or query language to match
   multiple certificates.  Identifying a specific certificate by
   reference or value allows diverse applications to have a common

Leonard                  Expires January 9, 2013                [Page 2]

Internet-Draft                  certspec                       July 2012

   syntax.  For example, applications can store certspecs as local or
   shared preferences, so that users can edit them without resorting to
   application-specific storage formats.  When conveyed in protocol, a
   certspec can identify a specific certificate to a client or server
   using text-based formats such as YAML, XML, JSON, and others.

3.  certspec Syntax

   The Namespace Specific String (NSS) portion of a certspec has the
   following ABNF specification:

    NSS = spec-type ":" spec-value ( '?' certattrs)
    spec-type = scheme
    certattrs = <URN chars>
    hexOctet  = hexDigit hexDigit
    hexDigit  =
             "0" / "1" / "2" / "3" / "4" / "5" / "6" / "7" / "8" / "9" /
             "a" / "b" / "c" / "d" / "e" / "f" /
             "A" / "B" / "C" / "D" / "E" / "F"

3.1.  spec-type and spec-value

   The spec-type identifies the certificate specification type.  The
   acceptable characters for spec-type are the same as an URI scheme
   name Section 3.1 of [RFC3986]. spec-types are compared case-
   insensitively.  The spec-value identifies the certificate
   specification value.  The acceptable characters for spec-value depend
   on the spec-type.

3.2.  certattrs

   A certspec can include attributes that are associated with the
   identified certificate.  These attributes do NOT affect certificate
   identification; the syntax is intended primarily to convey
   certificate metadata such as attributes found in PKCS #9, PKCS #11,
   PKCS #12, and particular implementations of cryptographic libraries.
   The characters of certattrs can be any valid URN character from
   [RFC2141] (effectively, any URI character from [RFC3986] except "["
   and "]").  This Internet-Draft does not further define certattrs.

4.  Certificate Specifications

   A certificate specification (certspec) unambiguously identifies a
   single certificate.  A certificate has exactly one canonical certspec
   per applicable definition (irrespective of case changes or other
   differences that do not affect lexical equivalence).  This Internet-

Leonard                  Expires January 9, 2013                [Page 3]

Internet-Draft                  certspec                       July 2012

   Draft provides four cryptographic hash-based specs, two value-based
   specs, and two data-based specs.

4.1.  Cryptographic Hash-Based Specifications

   A cryptographic hash of a certificate uniquely identifies that
   certificate.  Such a hash may also be called a "certificate
   fingerprint".  In all certspecs in this specification *or* derived
   from this specification, the hash is computed over the octets of the
   DER encoding of the certificate, namely, the Certificate type of
   Section 4.1 of [RFC5280].  The DER encoding includes tag and length
   octets, so the first octet is always 30h (the tag for SEQUENCE), and
   the second octet is never 80h (the octet for indefinite length).

   In certspecs in this specification, the spec-value is the hexadecimal
   encoding of the hash value octets.  For example, a 256-bit SHA-256
   hash is represented by exactly 32 hex octets, or 64 hex characters.
   The following ABNF defines proper spec-values:

      spec-value-sha-1   = 20hexOctet
      spec-value-sha-256 = 32hexOctet
      spec-value-SHA-384 = 48hexOctet
      spec-value-SHA-512 = 64hexOctet

   Lexical equivalence of two certspecs that have the same spec-type
   SHALL be determined by converting the hexadecimal spec-values to
   octets and comparing exact equivalence of the octets.  A strict
   implementation would reject values that contain non-hex digits, such
   as spaces, tabs, or hyphens.  However, a lenient implementation MAY
   ignore non-hex characters.  In any event, lexical equivalence is
   determined by converting the hex to octets, and then comparing the
   octets.  If there are too few or too many hex characters, a
   conforming implementation MUST reject the certspec.

   Conforming implementations to this Internet-Draft MUST recognize
   these hash-based certspecs, unless security considerations dictate
   otherwise.  Acceptable reasons for refusing to process a certspec
   include a) the local policy prohibits use of the hash, or b) the hash
   has known cryptographic weaknesses, such as a preimage attacks, which
   weaken the cryptographic uniqueness guarantees of the hash.

4.1.1.  SHA-1

   The spec-type is "SHA-1".  The hash is computed using SHA-1 [SHS].

Leonard                  Expires January 9, 2013                [Page 4]

Internet-Draft                  certspec                       July 2012

4.1.2.  SHA-256

   The spec-type is "SHA-256".  The hash is computed using SHA-256

4.1.3.  SHA-384

   The spec-type is "SHA-384".  The hash is computed using SHA-384

4.1.4.  SHA-512

   The spec-type is "SHA-512".  The hash is computed using SHA-512

4.1.5.  [Discussion]

   [[DP1: Hashes could be grouped into the spec-type "fp", such as:
   urn:cert:fp:HASHNAME:HASHVALUE.  In such a case, a registry, such as
   one maintained by IANA, could be used to establish valid HASHNAMEs.
   In the formulation above, three characters ("fp:") are omitted, and
   the ability to use "arbitrary" hashes is curtailed.  A new hash can
   be used, but must be procured through the IETF consensus process,
   which is a higher barrier for establishing new hashes than a registry
   entry, but offer less oversight.  It is believed that omitting "fp:"
   fosters interoperability because it is better to identify
   certificates by a small number of widely-recognized, standard
   algorithms, especially if human consumption or production are
   foreseen. --S.L.]]

4.2.  Value-Based Specifications

   A certificate may be identified reflexively, by its constitutent
   octets.  For small-to-medium certificates, identifying the
   certificate by embedding it in the certspec will be computationally
   efficient and resistant to denial-of-service attacks (by being always

   The octets of a certificate are the octets of the DER encoding of the
   certificate, namely, the Certificate type of Section 4.1 of RFC 5280.

   Lexical equivalence of two certspecs that are value-based SHALL be
   determined by converting the spec-value to certificate octets, and
   comparing the octets for strict equivalence.  Accordingly, it is
   possible for a base64 and a hex certspec to be lexically equivalent
   to each other.

   A conforming implementation MUST implement base64 and hex specs.

Leonard                  Expires January 9, 2013                [Page 5]

Internet-Draft                  certspec                       July 2012

4.2.1.  base64

   The spec-type is "base64".  The spec-value is the base64 encoding
   (Section 4 of [RFC4648]) of the certificate octets.  Like the data:
   URL [RFC2397], the characters '+' and '/' refer to values 62 and 63,
   respectively.  Additionally, if the length of certificate octets is
   not a multiple of 3, it is expected that one or two trailing equal
   signs '=' will be present.

   '+', '/', and '=' have no reserved meaning in this spec-type.  While
   the URN syntax rules [RFC2141] state that '/' should not be used in
   unencoded form, in this specification, '/' MAY be present in
   unencoded form in the base64 spec-type.  In any case, a conforming
   implementation MUST be able to process "%"-encoded characters.

   While a strict implementation would reject non-base64 characters, a
   lenient implementation MAY ignore non-base64 characters, such as CR,
   LF, whitespace, or the absence of trailing '='.  As a result, two
   certspecs that have the same base64-encoded data but different stray
   non-base64 characters MAY be judged lexically equivalent.  Similarly,
   [RFC2141] requires that non-reserved characters (in this case,
   alphanumerics) must not be "%"-encoded, but a lenient implementation
   MAY decode these "%"-encoded characters anyway.  This specification
   neither recommends nor discourages such leniency, but implementors
   should weigh the benefits and risks as discussed further in the
   Security Considerations section.

4.2.2.  hex

   The spec-type is "hex".  The spec-value is the hexadecimal encoding
   (Section 8 of [RFC4648]) of the certificate octets.  Whether an
   implementation should process "%"-encoded characters or non-hex
   characters is subject to the same considerations as the equivalent
   characters in the base64 spec.

4.3.  Data-Based Specifications

   A certificate may be identified by data contained within it.  The
   following specs reflect the traditional reliance of PKIX [RFC5280]
   and CMS [RFC5652] on a certficiate's issuer name and serial number,
   or a certificate's subject key identifier.  These specs provide
   textual representations for these identifiers.

4.3.1.  issuersn: Issuer Name and Serial Number

   The spec-type is "issuersn".  The spec-value is given by the
   following ABNF:

Leonard                  Expires January 9, 2013                [Page 6]

Internet-Draft                  certspec                       July 2012

     spec-value-issuersn = dn SEMI serialNumber
     serialNumber        = 1*hexOctet

   <dn> is defined in [RFC4516], which is based on <distinguishedName>
   from [RFC4514] with the addition of "%"-encoding. <SEMI> is defined
   in [RFC4512].  RFC 4514 no longer separates relative distinguished
   names (RDNs) by semicolons, as required by its predecessor, RFC 2253.
   Accordingly, ';' is used to separate the issuer's DN from the
   subject's serial number.  If ';' is present in the dn, it MUST be
   preceded by a backslash '\', which MUST be "%"-encoded.

   Care should be taken in escaping and "%"-encoding the relevant
   characters.  In particular, "?" is permitted in a distinguishedName,
   but is RESERVED by this specification and [RFC2141].  Any question
   marks in distinguished names MUST be "%"-encoded when placed in the

   <serialNumber> is the hexadecimal encoding of the certificate's
   serial number, with the exact same (DER encoded) contents octets of a
   CertificateSerialNumber ::= INTEGER as specified in Section 4.1 of
   [RFC5280].  If the serial number hex octets are malformed, the
   certspec is invalid.

   A conforming implementation SHOULD implement this issuersn spec.  If
   the implementation implements it, the implementation MUST process
   serial numbers up to the same length as required by Section
   of [RFC5280] (20 octets), and MUST process distinguished name strings
   as required by [RFC4514], including the table of minimum
   AttributeType name strings that MUST be recognized.

   Lexical equivalence of two issuersn certspecs SHALL be determined by
   comparing the integer values of the serialNumbers for exact
   equivalence, and comparing the distinguished names for a match.
   Distinguished names match if they satisfy the name matching
   requirements of [RFC5280].  An implementation MAY use attribute
   matching rules that are more restrictive or comprehensive than
   [RFC5280], provided that the applied rules do not contradict the LDAP
   definitions of the attributes.

4.3.2.  ski: Subject Key Identifier

   The spec-type is "ski".  The spec-value is given by the following

     spec-value-ski = keyIdentifier
     keyIdentifier  = 1*hexOctet

   <keyIdentifier> is the hexadecimal encoding of the certificate's

Leonard                  Expires January 9, 2013                [Page 7]

Internet-Draft                  certspec                       July 2012

   subject key identifier, which is recorded in the certificate's
   Subject Key Identifier extension (Section of [RFC5280]).  A
   certificate that lacks a subject key identifier cannot be identified
   using this spec.

   Lexical equivalence of two ski certspecs SHALL be determined by
   converting the hexadecimal spec-values to octets and comparing the
   exact equivalence of the octets.

   A conforming implementation MAY implement this ski spec.

5.  Registration Template

   Namespace ID:

   Registration Information:
      Version: 1
      Date: 2012-07-08

   Declared registrant of the namespace:

   Declaration of syntactic structures:
      The structure of the Namespace Specific String is provided

   Relevant ancillary documentation:
      Certificates are defined by [RFC5280] and [X.509].

   Identifier uniqueness considerations:
      The spec type is assigned by IANA through the IETF consensus
      process, so this process guarantees uniqueness of these
      identifiers. The uniqueness of the spec value depends on the
      spec type. For specs that identify cryptographic hashes, the
      cryptographic hash algorithm itself guarantees uniquess.
      For specs that identify certificates by value, the inclusion
      of the certificate in the URN itself guarantees uniqueness.
      For specs that identify certificates by certificate data,
      the resolver's database of certificates and implementation
      of certification path validation Section 6 of [RFC5280] ensure

   Identifier persistence considerations:
      A certificate is a permanent digital artifact, irrespective of
      its origin. As the assignment process records mathematical or
      existential facts about the certificate, such as one of its

Leonard                  Expires January 9, 2013                [Page 8]

Internet-Draft                  certspec                       July 2012

      cryptographic hashes, the binding between the URN and
      the certificate resource is permanent. Changing even one bit
      of the certificate will alter its URN, will make the
      certificate unusable, or both.

   Process of identifiers assignment:
      Generating a certspec (cert URN) does not require that
      a registration authority be contacted.

   Process for identifier resolution:
      This Internet Draft does not specify a resolution service
      for certspecs. However, resolving certificate references
      to actual certificates is a common practice with a wide number
      of offline and online implementations.

   Rules for Lexical Equivalence:
      Certspecs (cert URNs) are lexically equivalent if they both
      have the same spec type (compared case-insensitively)
      and the same space value, and therefore impliedly point
      to the same certificate.
      Comparison of spec values depends on the rules of the spec.
      Although extensions may be appended to a certspec,
      these extensions are guaranteed not to affect lexical

      Certspecs are semantically equivalent if they both resolve
      to the same certificate.

   Conformance with URN Syntax:
      The character '?' is reserved for future extensions to this
      specification. The URN of this namespace conforms to URN Syntax
      [RFC2141] and Uniform Resource Identifier (URI): Generic Syntax

   Validation mechanism:
      Each spec defines the validation mechanism for its respective
      value. It may be appreciated that validation of the URN is a
      completely different process from the Certification Path
      Validation Algorithm (Section 6 of [RFC5280]), which determines
      whether the *certificate* is valid.


6.  Use of certspec outside URN

   certspec is useful wherever a system may need to refer to a

Leonard                  Expires January 9, 2013                [Page 9]

Internet-Draft                  certspec                       July 2012

   certificate by value or by reference.  Some implementations may wish
   to refer to a certificate without enabling all of the expressive
   power (and security considerations) of URIs.  Accordingly, this
   section provides a uniform method for using a certspec outside of a
   URN.  Examples:


   To use certspec outside of a URI (URN) context, the prefix
   "urn:cert:" MAY be omitted.  All other lexical rules remain in
   effect, including "%"-encoding.  Care should be taken to process '?'
   in particular, since '?' separates the certspec from appended
   attributes.  A conforming implementation of raw certspecs MUST permit
   the prefix "urn:cert:" in addition to the raw certspec, which starts
   with the spec-type.  This specification guarantees that the the cert-
   type "urn" is RESERVED and will never be used.  However, implementors
   must take note that a raw certspec is not a valid URI, because cert-
   types are not registered URI schemes and do not have the same
   semantics as a URI.

7.  IANA Considerations

   This document requests the assignment of formal URN namespace ID

   This document requests the creation of a registry to record specs.
   New specs shall be ratified by the IETF consensus process.

8.  Best Practices

   When producing a hash-based certspec, the producer has a wide choice
   of hashes.  Nevertheless, this Internet-Draft RECOMMENDS that SHA-1
   or SHA-256 be used to foster the greatest interoperability and human
   recognition, provided that the Security Considerations are heeded.

9.  Security Considerations

   Digital certificates are important building blocks for
   authentication, integrity, authorization, and (occasionally)
   confidentiality services.  Accordingly, identifying digital
   certificates incorrectly can have significant security ramifications.

   When using specs that are cryptographic hashes (fingerprints) for
   lookups and comparisons, the cryptographic hash algorithm MUST be

Leonard                  Expires January 9, 2013               [Page 10]

Internet-Draft                  certspec                       July 2012

   implemented properly and SHOULD have no known attack vectors.  The
   registration of a particular algorithm spec in this namespace does
   NOT mean that it is acceptable or safe for every usage, even though
   this Internet-Draft requires that a conforming implementation MUST
   implement certain specs.

   When using by-value specs, the implementation MUST be prepared to
   process URNs of arbitrary length.  As of this writing, useful
   certificates rarely exceed 10KB, and most implementations are
   concerned with keeping certificate sizes down rather than up (for
   example, to fit in a single TCP packet).  However, a pathological or
   malicious certificate could easily exceed these metrics.  If an URN
   resolver cannot process a URN's full length, it MUST reject the

   When using specs that depend on certificate data, the implementation
   MUST be prepared to deal with multiple found certificates that
   contain the same certificate data, but are not the same certificate.
   In such a case, the implementation MUST segregate these certificates
   so that it only resolves the URN to certificates that it considers
   valid or trustworthy (as discussed further below).  If, despite this
   segregation, multiple valid or trustworthy certificates match the
   certspec, the certspec MUST be rejected, because a certspec is meant
   to identify exactly one certificate (not a family of certificates).

   Apart from the mechanics of certspecs (cert URNs), certificates
   should not be used unless they have passed the Certification Path
   Validation Algorithm (Section 6 of [RFC5280]), or another algorithm
   that provides some guarantee of validity.  For example, if a
   certificate database contains a set of certificates that it considers
   inherently trustworthy, then the inclusion of a certificate in that
   set makes it trustworthy, regardless of the results of the
   Certification Path Validation Algorithm.  Such a database is
   frequently used for "Root CA" lists.

10.  References

10.1.  Normative References

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

   [RFC2141]  Moats, R., "URN Syntax", RFC 2141, May 1997.

   [RFC3406]  Daigle, L., van Gulik, D., Iannella, R., and P. Faltstrom,
              "Uniform Resource Names (URN) Namespace Definition
              Mechanisms", BCP 66, RFC 3406, October 2002.

Leonard                  Expires January 9, 2013               [Page 11]

Internet-Draft                  certspec                       July 2012

   [RFC3986]  Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform
              Resource Identifier (URI): Generic Syntax", STD 66,
              RFC 3986, January 2005.

   [RFC4512]  Zeilenga, K., "Lightweight Directory Access Protocol
              (LDAP): Directory Information Models", RFC 4512,
              June 2006.

   [RFC4514]  Zeilenga, K., "Lightweight Directory Access Protocol
              (LDAP): String Representation of Distinguished Names",
              RFC 4514, June 2006.

   [RFC4516]  Smith, M. and T. Howes, "Lightweight Directory Access
              Protocol (LDAP): Uniform Resource Locator", RFC 4516,
              June 2006.

   [RFC4648]  Josefsson, S., "The Base16, Base32, and Base64 Data
              Encodings", RFC 4648, October 2006.

   [RFC5280]  Cooper, D., Santesson, S., Farrell, S., Boeyen, S.,
              Housley, R., and W. Polk, "Internet X.509 Public Key
              Infrastructure Certificate and Certificate Revocation List
              (CRL) Profile", RFC 5280, May 2008.

   [SHS]      National Institute of Standards and Technology, "Secure
              Hash Standard", FIPS PUB 180-4, March 2012, <http://

10.2.  Informative References

   [P11]      RSA Laboratories, "PKCS #11 v2.20: Cryptographic Token
              Interface Standard", PKCS #11, June 2004.

   [P12]      RSA Laboratories, "PKCS #12 v1.0: Personal Information
              Exchange Syntax Standard", PKCS #12, June 1999.

   [P9]       RSA Laboratories, "PKCS #9 v2.0: Selected Object Classes
              and Attribute Types", PKCS #9, February 2000.

   [RFC2253]  Wahl, M., Kille, S., and T. Howes, "Lightweight Directory
              Access Protocol (v3): UTF-8 String Representation of
              Distinguished Names", RFC 2253, December 1997.

   [RFC2397]  Masinter, L., "The "data" URL scheme", RFC 2397,
              August 1998.

   [RFC5652]  Housley, R., "Cryptographic Message Syntax (CMS)", STD 70,
              RFC 5652, September 2009.

Leonard                  Expires January 9, 2013               [Page 12]

Internet-Draft                  certspec                       July 2012

   [X.509]    International Telecommunications Union, "Information
              technology - Open Systems Interconnection - The Directory:
              Public-key and attribute certificate frameworks", ITU-
              T Recommendation X.509, ISO Standard 9594-8,
              November 2008.

Author's Address

   Sean Leonard
   Penango, Inc.
   1215 K Street
   17th Floor
   Sacramento, CA  95814


Leonard                  Expires January 9, 2013               [Page 13]