INTERNET-DRAFT                                                  P. Leach
Obsoletes: 2831                                                Microsoft
Intended category: Standards track                             C. Newman
                                                        Sun Microsystems
                                                             A. Melnikov
                                                              Isode Ltd.
                                                            October 2005

            Using Digest Authentication as a SASL Mechanism
                   draft-ietf-sasl-rfc2831bis-07.txt

Status of this Memo

   By submitting this Internet-Draft, each author represents that any
   applicable patent or other IPR claims of which he or she is aware
   have been or will be disclosed, and any of which he or she becomes
   aware will be disclosed, in accordance with Section 6 of BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   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".

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.

Copyright Notice

   Copyright (C) The Internet Society (2005).

Abstract

   This specification defines how HTTP Digest Authentication [Digest]
   can be used as a SASL [RFC 2222] mechanism for any protocol that has
   a SASL profile. It is intended both as an improvement over CRAM-MD5
   [RFC 2195] and as a convenient way to support a single authentication
   mechanism for web, mail, LDAP, and other protocols.






Leach & Newman             Expires: April 2006                  [Page 1]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


Table of Contents

   1 INTRODUCTION.....................................................3
    1.1 CONVENTIONS AND NOTATION......................................3
    1.2 CHANNEL BINDINGS..............................................4
   2 AUTHENTICATION...................................................5
    2.1 INITIAL AUTHENTICATION........................................5
     2.1.1 Step One...................................................5
     2.1.2 Step Two...................................................9
     2.1.3 Step Three................................................16
    2.2 SUBSEQUENT AUTHENTICATION....................................17
     2.2.1 Step one..................................................17
     2.2.2 Step Two..................................................17
    2.3 INTEGRITY PROTECTION.........................................18
    2.4 CONFIDENTIALITY PROTECTION...................................18
   3 SECURITY CONSIDERATIONS.........................................21
    3.1 AUTHENTICATION OF CLIENTS USING DIGEST AUTHENTICATION........21
    3.2 COMPARISON OF DIGEST WITH PLAINTEXT PASSWORDS................21
    3.3 REPLAY ATTACKS...............................................21
    3.4 ONLINE DICTIONARY ATTACKS....................................22
    3.5 OFFLINE DICTIONARY ATTACKS...................................22
    3.6 MAN IN THE MIDDLE............................................22
    3.7 CHOSEN PLAINTEXT ATTACKS.....................................22
    3.8 CBC MODE ATTACKS.............................................
    3.9 SPOOFING BY COUNTERFEIT SERVERS..............................23
    3.10 STORING PASSWORDS...........................................23
    3.11 MULTIPLE REALMS.............................................24
    3.12 SUMMARY.....................................................24
   4 EXAMPLE.........................................................24
   5 REFERENCES......................................................26
    5.1 NORMATIVE REFERENCES.........................................26
    5.2 INFORMATIVE REFERENCES.......................................27
   6 IANA CONSIDERATIONS.............................................28
   7 ABNF............................................................29
    7.1 AUGMENTED BNF................................................29
    7.2 BASIC RULES..................................................31
   8 SAMPLE CODE.....................................................33
   9 AUTHORS' ADDRESSES..............................................XX
   10  ACKNOWLEDGEMENTS..............................................34
   11 FULL COPYRIGHT STATEMENT.......................................35
   Appendix A: Changes from 2831.....................................36
   Appendix B: Open Issues...........................................37

   <<Page numbers are all wrong, sorry.
   Section ordering should be changed too>>






Leach & Newman             Expires: April 2006                  [Page 2]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


1  Introduction

   This specification describes the use of HTTP Digest Access
   Authentication as a SASL mechanism. The authentication type
   associated with the Digest SASL mechanism is "DIGEST-MD5".

   This specification is intended to be upward compatible with the
   "md5-sess" algorithm of HTTP/1.1 Digest Access Authentication
   specified in [Digest]. The only difference in the "md5-sess"
   algorithm is that some directives not needed in a SASL mechanism have
   had their values defaulted.

   There is one new feature for use as a SASL mechanism: integrity
   protection on application protocol messages after an authentication
   exchange.

   Also, compared to CRAM-MD5, DIGEST-MD5 prevents chosen plaintext
   attacks, and permits the use of third party authentication servers,
   mutual authentication, and optimized reauthentication if a client has
   recently authenticated to a server.

1.1  Conventions and Notation

   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 RFC 2119 [RFC 2119].

   <<This specification uses the same ABNF notation and lexical
   conventions as HTTP/1.1 specification; see section 7>>.

   Let { a, b, ... } be the concatenation of the octet strings a, b, ...

   Let ** denote the power operation.

   Let H(s) be the 16 octet MD5 hash [RFC 1321] of the octet string s.

   Let KD(k, s) be H({k, ":", s}), i.e., the 16 octet hash of the string
   k, a colon and the string s.

   Let HEX(n) be the representation of the 16 octet MD5 hash n as a
   string of 32 hex digits (with alphabetic characters always in lower
   case, since MD5 is case sensitive).

   Let HMAC(k, s) be the 16 octet HMAC-MD5 [RFC 2104] of the octet
   string s using the octet string k as a key.






Leach & Newman             Expires: April 2006                  [Page 3]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   Let unq(X) be the value of the quoted-string X without the
   surrounding quotes and with all escape characters "\\" removed. For
   example for the quoted-string "Babylon" the value of unq("Babylon")
   is Babylon; for the quoted string "ABC\"123\\" the value of
   unq("ABC\"123\\") is ABC"123\.

   The value of a quoted string constant as an octet string does not
   include any terminating null character.

   <<Other terms like "protocol profile" are defined in RFC2222.>>

1.2  Channel Bindings


   "Channel binding" is a concept described in [RFC2743] and which
   refers to the act of cryptographically binding authentication at one
   network layer to a secure channel at another layer and where the end-
   points at both layers are the same entities.  In the context of the
   DIGEST-MD5 SASL mechanism this means ensuring that the challenge and
   response messages include the "channel bindings" of any cryptographic
   channel (e.g. TLS) over which the DIGEST-MD5 exchange is transported,
   and that the inputs to the digest function include the same as well.
   The "channel bindings" of a channel here refer to information which
   securely identifies one instance of such a channel to both endpoints
   such that MITM attacks are detectable.   For TLS, the channel
   bindings are the TLS client and server finished messages.

   Channel bindings are herein added to DIGEST-MD5 by overloading the
   nonce and cnonce fields of the digest-challenge and digest-response
   messages, respectively.  Because these nonces are treated as opaque
   octet strings in previous versions of this mechanism such overloading
   is backwards compatible.  Because these nonces are used in the
   construction of the response-value (i.e., as input to the digest
   function) using these fields for carrying channel bindings data makes
   the channel binding operation possible without requiring incompatible
   changes to the message formats.  The fact that the odds that older
   implementations may select random nonces that resemble actual channel
   bindings data are so low allows new implementations to detect old
   peers and to decide whether to allow such peers or reject them
   according to local policy.











Leach & Newman             Expires: April 2006                  [Page 4]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


2  Authentication

   DIGEST-MD5 can operate in two modes. Initial authentication (section
   2.1) is usually used when a client authenticates to a server for the
   first time.  If protocol profile supports initial client response
   (see "Protocol profile requirements" in [RFC 2222]) and the client
   supports reauthentication and it has successfully authenticated to
   the server before, the client may be able to use the more efficient
   fast reauthentication mode as described in section 2.2.

   The following sections describe these two modes in details.

2.1  Initial Authentication

   If the client has not recently authenticated to the server, then it
   must perform "initial authentication", as defined in this section. If
   it has recently authenticated, then a more efficient form is
   available, defined in the next section.

2.1.1  Step One

   The server starts by sending a challenge. The data encoded in the
   challenge is formatted according to the rules for the "digest-
   challenge" defined as follows:



























Leach & Newman             Expires: April 2006                  [Page 5]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   digest-challenge  =
         1#( realm | nonce | qop-options | stale | server_maxbuf | charset
               algorithm | cipher-opts | auth-param )

        realm             = "realm" "=" realm-value
        realm-value       = quoted-string
        nonce             = "nonce" "=" nonce-value
        nonce-value       = quoted-string
                            ;; contains data described by "nonce-data"
        qop-options       = "qop" "=" <"> qop-list <">
        qop-list          = 1#qop-value
        qop-value         = "auth" | "auth-int" | "auth-conf" |
                             qop-token
                             ;; qop-token is reserved for identifying future
                             ;; extensions to DIGEST-MD5
        qop-token         = token
        stale             = "stale" "=" "true"
        server_maxbuf     = "maxbuf" "=" maxbuf-value
        maxbuf-value      = 1*DIGIT
        charset           = "charset" "=" "utf-8"
        algorithm         = "algorithm" "=" "md5-sess"
        cipher-opts       = "cipher" "=" <"> 1#cipher-value <">
        cipher-value      = "rc4-40" | "rc4" | "rc4-56" |
                            "aes-cbc" | cipher-token
                             ;; cipher-token is reserved for new ciphersuites
        cipher-token      = token
        auth-param        = token "=" ( token | quoted-string )
        nonce-data        = new-nonce-data | obs-nonce-data
        new-nonce-data    = "CB-" channel-type ":" channel-bindings ":" nonce-octets
        obs-nonce-data    = nonce-octets
                            ;; nonce value as defined in RFC 2831. Should be accepted
                            ;; must not be generated.
        channel-type      = "TLS" / channel-type-ext
        channel-type-ext  = 1*(ALPHA | DIGIT)
                            ;; for future channel bindings
        channel-bindings  = 1*TEXTCHAR
                            ;; channel binding data as defined by the channel type
        nonce-octets      = 1*TEXTCHAR

   The meanings of the values of the directives used above are as
   follows:

   realm
      Mechanistically, a string which enables users to decide which
      username and password to use, in case they have different ones for
      different servers.  Conceptually, it is the name of a collection
      of accounts that might include the user's account. This string
      should contain the name of the host performing the authentication



Leach & Newman             Expires: April 2006                  [Page 6]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      and might additionally indicate the collection of users who might
      have access. An example might be
      "registered_users@gotham.news.example.com".  Note that the server
      MAY not advertise (hide) some or all realms it supports.

      Other examples:

      1) "dc=gotham, dc=news, dc=example, dc=com".

      2) If there are two servers (e.g. server1.example.com and
         server2.example.com) that share authentication database, they
         both may advertise "example.com" as the realm.

      A server implementation that uses a fixed string as the advertised
      realm is compliant with this specification, however this is not
      recommended.  See also sections 3.10 "Storing passwords" and 3.11
      "Multiple realms" for discussion.

      The value of this directive is case-sensitive. This directive is
      optional; if not present, the client SHOULD solicit it from the
      user or be able to compute a default; a plausible default might be
      the realm supplied by the user when they logged in to the client
      system.  Multiple realm directives are allowed, in which case the
      user or client must choose one as the realm for which to supply
      username and password.

      Requirements on UIs: UIs MUST allow users to enter arbitrary user
      names and realm names. In order to achieve this, UIs MAY present
      two separate edit boxes. Alternatively, UIs MAY present a single
      edit box and allow user to enter a special character that
      separates user name from the realm name. In the latter case, UIs
      MUST be able to escape the special character and they need to
      present their escape rules to the user.  UIs MUST also present the
      list of realms advertised by the server.

   nonce
      A server-specified string erstwhile intended to add entropy to the
      challenge.  The nonce field may be used to exchange channel
      binding data.

      This directive is required and MUST appear exactly once; if not
      present, or if multiple instances are present, the client should
      abort the authentication exchange.

      Older implementations typically generate some random or pseudo-
      random data and base64 [RFC 3548] or hexadecimally encode it.
      When channel binding is not used the nonce string MUST be
      different each time a digest-challenge is sent as part of initial



Leach & Newman             Expires: April 2006                  [Page 7]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      authentication.  It is RECOMMENDED that the random data contain at
      least 64 bits of entropy.

      When channel binding is performed, the nonce must be generated
      from the channel type, the bindings to the channel being bound to
      and an actual nonce consisting of 64-bits or more of entropy and
      base64-encoded, and formatted as follows:

      "CB-" <channel type> ":" <channel bindings> ":" <nonce octets>

      The only channel binding currently defined is to TLS channels.
      The channel type for TLS is "TLS" and the channel bindings for TLS
      channels consist of the TLS client and server finished messages
      concatenated in that order and base64-encoded.

      <<Do we need an IANA registry?>>

      An actual nonce is included in order to allow for channel bindings
      to possible future channels with channel bindings data which is
      not necessarily unique for each instance.

      When channel bindings are in use, clients MUST reject challenges
      that contain server nonce values of this form and whose channel
      bindings do not match those of the actual underlying channel as
      observed by the client.

   qop-options
      A quoted string of one or more comma-separated tokens indicating
      the "quality of protection" values supported by the server.  The
      value "auth" indicates authentication; the value "auth-int"
      indicates authentication with integrity protection; the value
      "auth-conf" indicates authentication with integrity protection and
      encryption.  This directive is optional; if not present it
      defaults to "auth". The client MUST ignore unrecognized options;
      if the client recognizes no option, it MUST abort the
      authentication exchange.

      <<What if this directive is present multiple times? Error, or take
      the union of all values?>>

   stale
      The "stale" directive is not used in initial authentication. See
      the next section for its use in subsequent authentications. This
      directive may appear at most once; if multiple instances are
      present, the client MUST abort the authentication exchange.

   server_maxbuf ("maximal ciphertext buffer size")
      A number indicating the size of the largest buffer (in bytes) the



Leach & Newman             Expires: April 2006                  [Page 8]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      server is able to receive when using "auth-int" or "auth-conf".
      The value MUST be bigger than 16 (32 for Confidentiality
      protection with the "aes-cbc" cipher) and smaller or equal to
      16777215 (i.e. 2**24-1). If this directive is missing, the default
      value is 65536. This directive may appear at most once; if
      multiple instances are present, or the value is out of range the
      client MUST abort the authentication exchange.

      Let "maximal cleartext buffer size" (or "maximal sender size") be
      the maximal size of a cleartext buffer that, after being
      transformed by integrity (section 2.3) or confidentiality (section
      2.4) protection function, will produce a SASL block of the maxbuf
      size.  As it should be clear from the name, the sender MUST never
      pass a block of data bigger than the "maximal sender size" through
      the selected protection function.  This will guarantee that the
      receiver will never get a block bigger than the maxbuf.



































Leach & Newman             Expires: April 2006                  [Page 9]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   charset
      This directive, if present, specifies that the server supports
      UTF-8 [UTF-8] encoding for the username, realm and password. If
      present, the username, realm and password are encoded as UTF-8
      [UTF-8].  If not present, the username, realm and password used by
      the client in Step 2 MUST be encoded in ISO 8859-1 [ISO-8859] (of
      which US-ASCII [USASCII] is a subset). The directive is needed for
      backwards compatibility with HTTP Digest, which only supports ISO
      8859-1.  This directive may appear at most once; if multiple
      instances are present, the client MUST abort the authentication
      exchange.

      Note, that this directive doesn't affect authorization id
      ("authzid").

   algorithm
      This directive is required for backwards compatibility with HTTP
      Digest, which supports other algorithms. This directive is
      required and MUST appear exactly once; if not present, or if
      multiple instances are present, the client SHOULD abort the
      authentication exchange.

   cipher-opts
      A list of ciphers that the server supports. This directive must be
      present exactly once if "auth-conf" is offered in the
      "qop-options" directive, in which case the "rc4" cipher is
      mandatory-to-implement. The client MUST ignore unrecognized
      ciphers; if the client recognizes no cipher, it MUST behave as if
      "auth-conf" qop option wasn't provided by the server. If the
      client recognizes no cipher and the server only advertised "auth-
      conf" in the qop option, the client MUST abort the authentication
      exchange.  See section 2.4 for more detailed description of the
      ciphers.

      rc4, rc4-40, rc4-56
         the RC4 cipher with a 128 bit, 40 bit, and 56 bit key,
         respectively.

      aes-cbc
         the Advanced Encryption Standard (AES) cipher [AES] in cipher
         block chaining (CBC) mode with a 128 bit key and explicit
         Initialization Vector (IV). This mode requires an IV that has
         the same size as the block size.

   auth-param
      This construct allows for future extensions; it may appear more
      than once. The client MUST ignore any unrecognized directives.




Leach & Newman             Expires: April 2006                 [Page 10]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   For use as a SASL mechanism, note that the following changes are made
   to "digest-challenge" from HTTP: the following Digest options (called
   "directives" in HTTP terminology) are unused (i.e., MUST NOT be sent,
   and MUST be ignored if received):

    opaque
    domain

   The size of a "digest-challenge" MUST be less than 2048 bytes.

2.1.2  Step Two

   The client validates "digest-challenge" as described in the previous
   section. In particular, when channel bindings are in use, client MUST
   reject "digest-challenge" that contain server nonce whose channel
   bindings do not match those of the actual underlying channel as
   observed by the client.

   The client makes note of the "digest-challenge" and then responds
   with a string formatted and computed according to the rules for a
   "digest-response" defined as follows:

   digest-response  = 1#( username | realm | nonce | cnonce |
                          nonce-count | qop | digest-uri | response |
                          client_maxbuf | charset | cipher | authzid |
                          auth-param )

       username         = "username" "=" username-value
       username-value   = quoted-string
       cnonce           = "cnonce" "=" cnonce-value
       cnonce-value     = nonce-value
       nonce-count      = "nc" "=" nc-value
       nc-value         = 8LHEX
       client_maxbuf    = "maxbuf" "=" maxbuf-value
       qop              = "qop" "=" qop-value
       digest-uri       = "digest-uri" "=" <"> digest-uri-value <">
       digest-uri-value  = serv-type "/" host [ "/" serv-name ]
       serv-type        = 1*ALPHA
       serv-name        = host
       response         = "response" "=" response-value
       response-value   = 32LHEX
       LHEX             = "0" | "1" | "2" | "3" |
                          "4" | "5" | "6" | "7" |
                          "8" | "9" | "a" | "b" |
                          "c" | "d" | "e" | "f"
       cipher           = "cipher" "=" cipher-value
       authzid          = "authzid" "=" authzid-value
       authzid-value    = quoted-string



Leach & Newman             Expires: April 2006                 [Page 11]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   The 'host' non-terminal is defined in [RFC 3986] as

       host          = IP-literal / IPv4address / reg-name

   username
      The user's name in the specified realm, encoded according to the
      value of the "charset" directive. This directive is REQUIRED and
      MUST be present exactly once; otherwise, authentication fails.

      If the charset directive is also specified (which means that the
      username is encoded as UTF-8) the client MUST first check if all
      the characters of the username are in the ISO 8859-1 character
      set. If they are, no further changes are performed. Otherwise, the
      client SHOULD prepare <<SHOULD prepare or MUST prepare, but SHOULD
      use SASLPrep. The same issue elsewhere>> the username using the
      "SASLPrep" profile [SASLPrep] of the "stringprep" algorithm [RFC
      3454]. If the preparation of the username fails or results in an
      empty string, the client SHOULD abort the authentication exchange
      (*).  If the preparation succeeds, the prepared value will be sent
      to the server and used in hash computations described in section
      2.1.2.1.

      (*) An interactive client can request a repeated entry of username
      value.

   realm
      The realm containing the user's account, encoded according to the
      value of the "charset" directive. This directive MUST appear at
      most once and SHOULD contain one of the realms provided by the
      server in the "digest-challenge". If the directive is missing,
      "realm-value" will set to the empty string when computing A1 (see
      below for details).

      If the realm value was provided by the client and if the charset
      directive is also specified in "digest-response" (which means that
      the realm is encoded as UTF-8), the client SHOULD first prepare it
      using the "SASLPrep" profile [SASLPrep] of the "stringprep"
      algorithm [RFC 3454]. If the preparation of the realm fails, the
      client SHOULD abort the authentication exchange(*).  If the
      preparation succeeds, the prepared version is sent to the server
      and used in hash computations described in section 2.1.2.1.

      (*) An interactive client can request a repeated entry of the
      realm value.

   nonce
      The server-specified data string received in the preceding digest-
      challenge.  This directive is required and MUST be present exactly



Leach & Newman             Expires: April 2006                 [Page 12]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      once; otherwise, authentication fails.

   cnonce
      A client-specified string erstwhile intended to add entropy to the
      challenge.  The cnonce field may be used to exchange channel
      binding data.

      This directive is required and MUST be present exactly once;
      otherwise, authentication fails.

      Older implementations typically generate some random or pseudo-
      random data and base64 [RFC 3548] or hexadecimally encode it.
      When channel binding is not used the cnonce string MUST be
      different each time a digest-challenge is sent as part of initial
      authentication.  It is RECOMMENDED that the random data contain at
      least 64 bits of entropy.

      When channel binding is performed, the cnonce must be generated
      from the channel type, the bindings to the channel being bound to
      and an actual nonce consisting of 64-bits or more of entropy and
      base64-encoded, and formatted as follows:

      "CB-" <channel type> ":" <channel bindings> ":" <nonce octets>

      The only channel binding currently defined is to TLS channels.
      The channel type for TLS is "TLS" and the channel bindings for TLS
      channels consist of the TLS client and server finished messages
      concatenated in that order and base64-encoded.

      An actual nonce is included in order to allow for channel bindings
      to possible future channels with channel bindings data which is
      not necessarily unique for each instance.  It is used by both
      client and server to avoid chosen plaintext attacks, and to
      provide mutual authentication.

   nonce-count
      The nc-value is the hexadecimal count of the number of requests
      (including the current request) that the client has sent with the
      nonce value in this request.  For example, in the first request
      sent in response to a given nonce value, the client sends
      "nc=00000001".  The purpose of this directive is to allow the
      server to detect request replays by maintaining its own copy of
      this count - if the same nc-value is seen twice, then the request
      is a replay. See the description below of the construction of the
      response value. This directive is required and MUST be present
      exactly once; otherwise, or if the value is 0, authentication
      fails.




Leach & Newman             Expires: April 2006                 [Page 13]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   qop
      Indicates what "quality of protection" the client accepted. If
      present, it may appear exactly once and  its value MUST be one of
      the alternatives in qop-options. If not present, it defaults to
      "auth".  These values affect the computation of the response. Note
      that this is a single token, not a quoted list of alternatives.

   serv-type
      Indicates the type of service, such as "http" for web service,
      "ftp" for FTP service, "smtp" for mail delivery service, etc. The
      service name as defined in the SASL profile for the protocol see
      section 4 of [RFC 2222], registered in the IANA registry of
      "service" elements for the GSSAPI host-based service name form
      [RFC 2078].

   host
      The DNS host name or IP (IPv4 or IPv6) address for the service
      requested.  The DNS host name must be the fully-qualified
      canonical name of the host.  The DNS host name is the preferred
      form; see notes on server processing of the digest-uri.

   serv-name
      Indicates the name of the service if it is replicated. The service
      is considered to be replicated if the client's service-location
      process involves resolution using standard DNS lookup operations,
      and if these operations involve DNS records (such as SRV [RFC
      2052], or MX) which resolve one DNS name into a set of other DNS
      names. In this case, the initial name used by the client is the
      "serv-name", and the final name is the "host" component. For
      example, the incoming mail service for "example.com" may be
      replicated through the use of MX records stored in the DNS, one of
      which points at an SMTP server called "mail3.example.com"; it's
      "serv-name" would be "example.com", it's "host" would be
      "mail3.example.com". If the service is not replicated, or the
      serv-name is identical to the host, then the serv-name component
      MUST be omitted.

   digest-uri
      Indicates the principal name of the service with which the client
      wishes to connect, formed from the serv-type, host, and serv-name.
      For example, the FTP service on "ftp.example.com" would have a
      "digest-uri" value of "ftp/ftp.example.com"; the SMTP server from
      the example above would have a "digest-uri" value of
      "SMTP/mail3.example.com/example.com".

   Servers SHOULD check that the supplied value is correct. This will
   detect accidental connection to the incorrect server, as well as some
   redirection attacks. It is also so that clients will be trained to



Leach & Newman             Expires: April 2006                 [Page 14]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   provide values that will work with implementations that use a shared
   back-end authentication service that can provide server
   authentication.

   The serv-type component should match the service being offered. The
   host component should match one of the host names of the host on
   which the service is running, or it's IP address. Servers SHOULD NOT
   normally support the IP address form, because server authentication
   by IP address is not very useful; they should only do so if the DNS
   is unavailable or unreliable. The serv-name component should match
   one of the service's configured service names.

   This directive is required and MUST be present exactly once; if
   multiple instances are present, the server MUST abort the
   authentication exchange.

   Note: In the HTTP use of Digest authentication, the digest-uri is the
   URI (usually a URL) of the resource requested -- hence the name of
   the directive.

   response
      A string of 32 hex digits computed as defined below, which proves
      that the user knows a password. This directive is required and
      MUST be present exactly once; otherwise, authentication fails.

   client_maxbuf
      A number indicating the size of the largest ciphertext buffer the
      client is able to receive when using "auth-int" or "auth-conf". If
      this directive is missing, the default value is 65536. This
      directive may appear at most once; if multiple instances are
      present, the server MUST abort the authentication exchange. If the
      value is less or equal to 32 (if "aes-cbc" cipher was selected) or
      16 (if one of the rc4 ciphers was selected), or bigger than
      16777215 (i.e. 2**24-1), the server MUST abort the authentication
      exchange.

      Upon processing/sending of the client_maxbuf value both the server
      and the client calculate their "maximal ciphertext buffer size" as
      the minimum of the server_maxbuf (Step One) and the client_maxbuf
      (Step Two).  The "maximal sender size" can be calculated by
      subtracting 32 (if "aes-cbc" cipher was selected) or 16 (for any
      of the rc4 ciphers) from the calculated "maximal ciphertext buffer
      size".

      When sending a block of data the client/server MUST NOT pass more
      than the "maximal sender size" bytes of data to the selected
      protection function (2.3 or 2.4).




Leach & Newman             Expires: April 2006                 [Page 15]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   charset
      This directive, if present, specifies that the client has used
      UTF-8 [UTF-8] encoding for the username, realm and password. If
      present, the username, realm and password are encoded as UTF-8
      [UTF-8].  If not present, the username, realm and password MUST be
      encoded in ISO 8859-1 [ISO-8859] (of which US-ASCII [USASCII] is a
      subset). The client should send this directive only if the server
      has indicated that it supports UTF-8 [UTF-8].  The directive is
      needed for backwards compatibility with HTTP Digest, which only
      supports ISO 8859-1.

      Note, that this directive doesn't affect the authorization
      identity ("authzid").

   LHEX
      32 hex digits, where the alphabetic characters MUST be lower case,
      because MD5 is not case insensitive.

   cipher
      The cipher chosen by the client. This directive MUST appear
      exactly once if "auth-conf" is negotiated; if required and not
      present, authentication fails.  If the cipher chosen by the client
      is not one of the ciphers advertised by the server, authentication
      fails.

   authzid
      The "authorization ID" (authzid) directive may appear at most
      once; if multiple instances are present, the server MUST abort the
      authentication exchange. If present, and the authenticating user
      has sufficient privilege, and the server supports it, then after
      authentication the server will use this identity for making all
      accesses and access checks. If the client specifies it, and the
      server does not support it, then the response-value calculated on
      the server will not match the one calculated on the client and
      authentication will fail.

      The authzid MUST NOT be an empty string.

      The authorization identifier MUST NOT be converted to ISO 8859-1
      even if the authentication identifier ("username") is converted
      for compatibility as directed by "charset" directive.

      The server SHOULD verify the correctness of an authzid as
      specified by the corresponding SASL protocol profile.

   The size of a digest-response MUST be less than 4096 bytes.

2.1.2.1   Response-value



Leach & Newman             Expires: April 2006                 [Page 16]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   The definition of "response-value" above indicates the encoding for
   its value -- 32 lower case hex characters. The following definitions
   show how the value is computed.

   Although qop-value and components of digest-uri-value may be
   case-insensitive, the case which the client supplies in step two is
   preserved for the purpose of computing and verifying the
   response-value.

      response-value  =
         HEX( KD ( HEX(H(A1)),
                 { unq(nonce-value), ":" nc-value, ":",
                   unq(cnonce-value), ":", qop-value, ":", HEX(H(A2)) }))

   If authzid is specified, then A1 is

      A1 = { SS, ":", unq(nonce-value), ":",
             unq(cnonce-value), ":", unq(authzid-value) }

   If authzid is not specified, then A1 is

      A1 = { SS, ":", unq(nonce-value), ":", unq(cnonce-value) }

   where

         password   = *OCTET

         SS = H( { unq(username-value), ":", unq(realm-value), ":", password } )



   The "username-value", "realm-value" and "password" are encoded
   according to the value of the "charset" directive. If "charset=UTF-8"
   is present, and all the characters of "username-value" are in the ISO
   8859-1 character set, then it MUST be converted to ISO 8859-1 before
   being hashed. The same transformation has to be done for "realm-
   value" and "password".  This is so that authentication databases that
   store the  hashed username, realm and password (which is common) can
   be shared  compatibly with HTTP, which specifies ISO 8859-1. A sample
   implementation of this conversion is in section 8.

   Note that on the client side the "username-value", "realm-value"(*)
   and "password" MUST be prepared before being encoded as described in
   the previous paragraph.  The [SASLPrep] profile of the [StringPrep]
   algorithm is the RECOMMENDED preparation algorithm.  The SASLprep
   preparation algorithm is recommended to improve the likelihood that
   comparisons behave in an expected manner.  The SASLprep preparation
   algorithm is not mandatory.  This is done to allow, when appropriate,



Leach & Newman             Expires: April 2006                 [Page 17]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   the server to employ other preparation algorithms (including none).
   For instance, use of different preparation algorithm may be necessary
   for the server to interoperate with an external system.

   (*) - Note that if the "realm-value" was provided by the client (i.e.
   not chosen from the list of realms returned by the server), it MUST
   NOT be prepared.

   On the server side the preparation of the "username-value", "realm-
   value" and "password" MUST NOT be performed.

   If the "qop" directive's value is "auth", then A2 is:

      A2       = { "AUTHENTICATE:", digest-uri-value }





































Leach & Newman             Expires: April 2006                 [Page 18]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   If the "qop" value is "auth-int" or "auth-conf" then A2 is:

      A2       = { "AUTHENTICATE:", digest-uri-value,
               ":00000000000000000000000000000000" }

   Note that "AUTHENTICATE:" must be in upper case, and the second
   string constant is a string with a colon followed by 32 zeros.

   These apparently strange values of A2 are for compatibility with
   HTTP; they were arrived at by setting "Method" to "AUTHENTICATE" and
   the hash of the entity body to zero in the HTTP digest calculation of
   A2.

   Also, in the HTTP usage of Digest, several directives in the
   "digest-challenge" sent by the server have to be returned by the
   client in the "digest-response". These are:

    opaque
    algorithm

   These directives are not needed when Digest is used as a SASL
   mechanism (i.e., MUST NOT be sent, and MUST be ignored if received).

2.1.3  Step Three

   The server receives and validates the "digest-response".  In
   particular, when channel bindings are in use, server MUST reject
   "digest-response" that contain client nonce whose channel bindings do
   not match those of the actual underlying channel as observed by the
   server.

   The server also checks that the nonce-count is "00000001". If it
   supports subsequent authentication (see section 2.2), it saves the
   value of the nonce <<(do channel bindings affect reauthentication?)>>
   and the nonce-count. It sends a message formatted as follows:

    auth-info      = 1#( response-auth | auth-param )

     response-auth = "rspauth" "=" response-value

   where response-value is calculated as above, using the values sent in
   step two, except that if qop is "auth", then A2 is

       A2 = { ":", digest-uri-value }

   And if qop is "auth-int" or "auth-conf" then A2 is

       A2 = { ":", digest-uri-value, ":00000000000000000000000000000000" }



Leach & Newman             Expires: April 2006                 [Page 19]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   Note that only one occurance of the "response-auth" is allowed. If
   more than one is found, the client MUST treat this as an
   authentication error.

   Compared to its use in HTTP, the following Digest directives in the
   "auth-info" are unused:

       nextnonce
       qop
       cnonce
       nonce-count

   The size of an auth-info MUST be less than 2048 bytes.

2.2  Subsequent Authentication

   If the client has previously authenticated to the server, and
   remembers the values of username, realm, nonce, nonce-count, cnonce,
   and qop that it used in that authentication, and the SASL profile for
   a protocol permits an initial client response, then it MAY perform
   "subsequent authentication" or "fast reauthentication", as defined in
   this section.  Note, that a subsequent authentication can be done on
   a different connection, or on the same connection, if the protocol
   profile also permits multiple authentications.

2.2.1  Step one

   The client uses the values from the previous authentication and sends
   an initial response with a string formatted and computed according to
   the rules for a "digest-response", as defined above, but with a
   nonce-count one greater than used in the last "digest-response".

2.2.2  Step Two

   The server receives the "digest-response". If the server does not
   support subsequent authentication, then it sends a
   "digest-challenge", and authentication proceeds as in initial
   authentication. If the server has no saved nonce and nonce-count from
   a previous authentication, then it sends a "digest-challenge", and
   authentication proceeds as in initial authentication. Otherwise, the
   server validates the "digest-response", checks that the nonce-count
   is one greater than that used in the previous authentication using
   that nonce, and saves the new value of nonce-count.

   If the response is invalid, then the server sends a
   "digest-challenge", and authentication proceeds as in initial
   authentication (and should be configurable to log an authentication
   failure in some sort of security audit log, since the failure may be



Leach & Newman             Expires: April 2006                 [Page 20]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   a symptom of an attack). The nonce-count MUST NOT be incremented in
   this case: to do so would allow a denial of service attack by sending
   an out-of-order nonce-count.

   If the response is valid, the server MAY choose to deem that
   authentication has succeeded. However, if it has been too long since
   the previous authentication, or for any other reason, the server MAY
   send a new "digest-challenge" with a new value for nonce. The
   challenge MAY contain a "stale" directive with value "true", which
   says that the client may respond to the challenge using the password
   it used in the previous response; otherwise, the client must solicit
   the password anew from the user. This permits the server to make sure
   that the user has presented their password recently. (The directive
   name refers to the previous nonce being stale, not to the last use of
   the password.) Except for the handling of "stale", after sending the
   "digest-challenge" authentication proceeds as in the case of initial
   authentication.

2.3   Integrity Protection

   If the server offered "qop=auth-int" and the client responded
   "qop=auth-int", then subsequent messages, up to but not including the
   next subsequent authentication, between the client and the server
   MUST be integrity protected. Using as a base session key the value of
   H(A1), as defined above the client and server calculate a pair of
   message integrity keys as follows.

   The key for integrity protecting messages from client to server is:

   Kic = MD5({H(A1),
   "Digest session key to client-to-server signing key magic constant"})

   The key for integrity protecting messages from server to client is:

   Kis = MD5({H(A1),
   "Digest session key to server-to-client signing key magic constant"})

   where MD5 is as specified in [RFC 1321]. If message integrity is
   negotiated, a MAC block for each message is appended to the message.
   The MAC block is 16 bytes: the first 10 bytes of the HMAC-MD5 [RFC
   2104] of the message, a 2-byte message type number in network byte
   order with value 1, and the 4-byte sequence number in network byte
   order. The message type is to allow for future extensions such as
   rekeying.

   MAC(Ki, SeqNum, msg) = (HMAC(Ki, {SeqNum, msg})[0..9], 0x0001,
   SeqNum)




Leach & Newman             Expires: April 2006                 [Page 21]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   where Ki is Kic for messages sent by the client and Kis for those
   sent by the server. The sequence number (SeqNum) is an unsigned
   number initialized to zero after initial or subsequent
   authentication, and incremented by one for each message
   sent/successfully verified. (Note, that there are two independent
   counters for sending and receiving.) The sequence number wraps around
   to 0 after 2**32-1.

   Upon receipt, MAC(Ki, SeqNum, msg) is computed and compared with the
   received value; the message is discarded if they differ and as the
   result the connection being used MUST be dropped. The receiver's
   sequence counter is incremented if they match.

2.4   Confidentiality Protection

   If the server sent a "cipher-opts" directive and the client responded
   with a "cipher" directive, then subsequent messages between the
   client and the server MUST be confidentiality protected. Using as a
   base session key the value of H(A1) as defined above the client and
   server calculate a pair of message integrity keys as follows.

   The key for confidentiality protecting messages from client to server
   is:

   Kcc = MD5({H(A1)[0..n-1],
   "Digest H(A1) to client-to-server sealing key magic constant"})

   The key for confidentiality protecting messages from server to client
   is:

   Kcs = MD5({H(A1)[0..n-1],
   "Digest H(A1) to server-to-client sealing key magic constant"})

   where MD5 is as specified in [RFC 1321]. For cipher "rc4-40" n is 5;
   for "rc4-56" n is 7; for the rest n is 16. The key for the "rc4-*"
   and "aes-cbc" ciphers is all 16 bytes of Kcc or Kcs.

   "aes-cbc" cipher works as described in section 2.4.1.

   rc4 cipher state MUST NOT be reset before sending/receiving a next
   buffer of protected data.


   If the blocksize of the chosen cipher is not 1 byte, the padding
   prefix is one or more octets each containing the number of padding
   bytes, such that the total length of the encrypted part of the
   message is a multiple of the blocksize.




Leach & Newman             Expires: April 2006                 [Page 22]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   The MAC block is 16 bytes formatted as follows: the first 10 bytes of
   the HMAC-MD5 [RFC 2104] of the message, a 2-byte message type number
   in network byte order with value 1, and the 4-byte sequence number in
   network byte order.

   The padding and first 10 bytes of the MAC block are encrypted with
   the chosen cipher along with the message and explicit IV (if
   present).

   SEAL(Ki, Kc, SeqNum, msg) = CIPHER(Kc, {exp_iv, msg, pad, MAC})

   MAC(Ki, SeqNum, exp_iv, msg) = {HMAC(Ki, {SeqNum, exp_iv, msg})[0..9],
                                   packet_type_data, SeqNum}

   packet_type_data = 0x0001

   where CIPHER is the chosen cipher, Ki and Kc are Kic and Kcc for
   messages sent by the client and Kis and Kcs for those sent by the
   server, exp_iv is empty string for rc4 ciphers and a randomly
   generated number R of the length 128 bit for the "aes-cbc" cipher.
   The sequence number (SeqNum) is an unsigned number initialized to
   zero after initial or subsequent authentication, and incremented by
   one for each message sent/successfully verified. (Note, that there
   are two independent counters for sending and receiving.) The sequence
   number wraps around to 0 after 2**32-1.

   Upon receipt, the message is decrypted, exp_iv is ignored (for the
   "aes-cbc" cipher only), HMAC(Ki, {SeqNum, msg}) is computed and
   compared with the received value; the padding and the packet type are
   verified.  The message is discarded if the received and the
   calculated HMACs differ and/or the padding is invalid. See also
   section 3.8 for important information about MAC and padding
   verification. The receiver's sequence counter is then compared with
   the received SeqNum value; the message is discarded if they differ
   and, as the result, the connection being used MUST be dropped. The
   receiver's sequence counter is incremented if they match.

2.4.1   AES cipher in CBC mode with explicit IV ("aes-cbc")

   Unlike previous versions of DIGEST-MD5, this document uses an
   explicit IV for ciphers in CBC mode. This is done in order to prevent
   the attacks described by [CBCATT].

   For each buffer of cleartext data to be encrypted the sender performs
   the following procedure:

   0) For the very first SASL packet sent the IV is calculated as
   follows:



Leach & Newman             Expires: April 2006                 [Page 23]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      The IV for the first SASL packet going from the client
      to the server (IVc) consists of 16 bytes calculated as follows:

       IVc = MD5({Kcc, "aes-128"})

      The IV for the first SASL packet going from the server
      to the client (IVs) consists of 16 bytes calculated as follows:

       IVs = MD5({Kcs, "aes-128"})

      For a subsequent packet: Em of the previous packet (see below)
      becomes the IV.

   1) Generate a cryptographically strong random number R of length 128
      bits (16 octets) and prepend it to the plaintext prior to
   encryption.

   2) padding and MAC block are constructed (see section 2.4) and
      appended to the end of the plaintext. After this step the data
      to be encrypted will look like:

       {R, msg, pad, MAC}

      As the total length of the data will be multiple of AES block size
      (i.e. 128 bit), this can also be represented as

       {P1, P2, P3, ..., Pm}

      where Pi is a chunk of data of the length 128 bit. Note, that
      P1 is R.

   3) Data is encrypted as follows:

       E1 = CIPHER ( Kc, P1 XOR IV )
       E2 = CIPHER ( Kc, P2 XOR E1 )
       E3 = CIPHER ( Kc, P3 XOR E2 )
       ...
       Ei = CIPHER ( Kc, Pi XOR Ei-1)
       ...
       Em = CIPHER ( Kc, Pm XOR Em-1)

      This will generate ciphertext {E1, ..., Em} to be sent as a single
      SASL packet.


   The receiver performs the following steps:

   0) For the very first SASL packet sent the IV is calculated as



Leach & Newman             Expires: April 2006                 [Page 24]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      in step 0 for the sender.

      For a subsequent packet: Em of the previous packet becomes
      the IV of the immediately following packet.

   1) Data is decrypted as follows:

       P1 = CIPHER ( Kc, E1 ) XOR IV
       P2 = CIPHER ( Kc, E2 ) XOR E1
       P3 = CIPHER ( Kc, E3 ) XOR E2
       ...
       Pi = CIPHER ( Kc, Ei ) XOR Ei-1
       ...
       Pm = CIPHER ( Kc, Em ) XOR Em-1

      Em becomes the IV for the decryption of the subsequent SASL
      packet.

      This will generate plaintext {P1, ..., Pm}. P1 is discarded,
      {P2, ..., Pm} is {msg, pad, MAC}.

   2) pad and MAC block are verified as described in section 2.4.





























Leach & Newman             Expires: April 2006                 [Page 25]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


3  Security Considerations

   General SASL security considerations apply to this mechanism.
   "stringprep" and Unicode security considerations also apply.

   Detailed discussion of other DIGEST-MD5 specific security issues is
   below.

3.1   Authentication of Clients using Digest Authentication

   Digest Authentication does not provide a strong authentication
   mechanism, when compared to public key based mechanisms, for example.
   However, since it prevents chosen plaintext attacks, it is stronger
   than (e.g.) CRAM-MD5, which has been proposed for use with ACAP [RFC
   2244], POP and IMAP [RFC 2195]. It is intended to replace the much
   weaker and even more dangerous use of plaintext passwords; however,
   since it is still a password based mechanism it avoids some of the
   potential deployability issues with public-key, OTP or similar
   mechanisms.

   Digest Authentication offers no confidentiality protection beyond
   protecting the actual password. All of the rest of the challenge and
   response are available to an eavesdropper, including the user's name
   and authentication realm.

3.2   Comparison of Digest with Plaintext Passwords

   The greatest threat to the type of transactions for which these
   protocols are used is network snooping. This kind of transaction
   might involve, for example, online access to a mail service whose use
   is restricted to paying subscribers. With plaintext password
   authentication an eavesdropper can obtain the password of the user.
   This not only permits him to access anything in the database, but,
   often worse, will permit access to anything else the user protects
   with the same password.

3.3   Replay Attacks

   Replay attacks are defeated if the client or the server chooses a
   fresh nonce for each authentication, as this specification requires.

   As a security precaution, the server, when verifying a response from
   the client, must use the original server nonce ("nonce") it sent, not
   the one returned by the client in the response, as it might have been
   modified by an attacker.

   To prevent some redirection attacks it is recommended that the server
   verifies that the "serv-type" part of the "digest-uri" matches the



Leach & Newman             Expires: April 2006                 [Page 26]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   service name and that the hostname/IP address belongs to the server.

3.4  Online dictionary attacks

   If the attacker can eavesdrop, then it can test any overheard
   nonce/response pairs against a (potentially very large) list of
   common words. Such a list is usually much smaller than the total
   number of possible passwords. The cost of computing the response for
   each password on the list is paid once for each challenge.

   The server can mitigate this attack by not allowing users to select
   passwords that are in a dictionary.

3.5  Offline dictionary attacks

   If the attacker can choose the challenge, then it can precompute the
   possible responses to that challenge for a list of common words. Such
   a list is usually much smaller than the total number of possible
   passwords. The cost of computing the response for each password on
   the list is paid just once.

   Offline dictionary attacks are defeated if the client chooses a fresh
   nonce for each authentication, as this specification requires.

3.6  Man in the Middle

   Digest authentication is vulnerable to "man in the middle" (MITM)
   attacks. Clearly, a MITM would present all the problems of
   eavesdropping. But it also offers some additional opportunities to
   the attacker.

   A possible man-in-the-middle attack would be to substitute a weaker
   qop scheme for the one(s) sent by the server; the server will not be
   able to detect this attack. For this reason, the client should always
   use the strongest scheme that it understands from the choices
   offered, and should never choose a scheme that does not meet its
   minimum requirements.

   A man-in-the-middle attack may also make the client and the server
   that agreed to use confidentiality protection to use different (and
   possibly weaker) cipher's. This is because the chosen cipher is not
   used in the shared secret calculation.

3.7  Chosen plaintext attacks

   A chosen plaintext attack is where a MITM or a malicious server can
   arbitrarily choose the challenge that the client will use to compute
   the response. The ability to choose the challenge is known to make



Leach & Newman             Expires: April 2006                 [Page 27]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   cryptanalysis much easier [MD5].

   However, Digest does not permit the attack to choose the challenge as
   long as the client chooses a fresh nonce for each authentication, as
   this specification requires.

3.8  CBC Mode attacks

   The following attack can be launched when the connection uses
   Confidentiality protection with ciphers in CBC mode. If bad padding
   is treated differently from bad MACs when decrypting a DIGEST-MD5
   buffer of protected data, the attacker may be able to launch
   Vaudenay's [VAUDENAY] attack on padding.

   An error logfile will suffice to launch the attack if it reveals the
   type of error -- even if file permissions prevent the attacker from
   actually reading the file (the file length increase cause by the
   attack is likely to reveal which of the two errors occured).

   A different approach to distinguish these two error cases and launch
   the attack is to examine the timing of error responses: if the MAC
   verification is skipped when bad padding has been found, the error
   will appear quicker in the case of incorrect block cipher padding
   than in the case of an incorrect MAC.

   A countermeasure is to compute a MAC of the plaintext anyway, even if
   the usual padding removal step fails because of incorrect padding, to
   obtain (nearly) uniform timing.

3.9  Spoofing by Counterfeit Servers

   If a user can be led to believe that she is connecting to a host
   containing information protected by a password she knows, when in
   fact she is connecting to a hostile server, then the hostile server
   can obtain challenge/response pairs where it was able to partly
   choose the challenge. There is no known way that this can be
   exploited.

3.10  Storing passwords

   Digest authentication requires that the authenticating agent (usually
   the server) store some data derived from the user's name and password
   in a "password file" associated with a given realm. Normally this
   might contain pairs consisting of username and H({ username-value,
   ":", realm-value, ":", password }), which is adequate to compute
   H(A1) as described above without directly exposing the user's
   password.




Leach & Newman             Expires: April 2006                 [Page 28]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   The security implications of this are that if this password file is
   compromised, then an attacker gains immediate access to documents on
   the server using this realm. Unlike, say a standard UNIX password
   file, this information need not be decrypted in order to access
   documents in the server realm associated with this file. On the other
   hand, decryption, or more likely a brute force attack, would be
   necessary to obtain the user's password. This is the reason that the
   realm is part of the digested data stored in the password file. It
   means that if one Digest authentication password file is compromised,
   it does not automatically compromise others with the same username
   and password (though it does expose them to brute force attack).

   There are two important security consequences of this. First the
   password file must be protected as if it contained plaintext
   passwords, because for the purpose of accessing documents in its
   realm, it effectively does.

   A second consequence of this is that the realm string should be
   unique among all realms that any single user is likely to use. In
   particular a realm string should include the name of the host doing
   the authentication.






























Leach & Newman             Expires: April 2006                 [Page 29]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


3.11  Multiple realms

   Use of multiple realms may mean both that compromise of a the
   security database for a single realm does not compromise all
   security, and that there are more things to protect in order to keep
   the whole system secure.

3.11  Summary

   By modern cryptographic standards Digest Authentication is weak,
   compared to (say) public key based mechanisms. But for a large range
   of purposes it is valuable as a replacement for plaintext passwords.
   Its strength may vary depending on the implementation.


4  Example

   This example shows the use of the Digest SASL mechanism with the
   IMAP4 AUTHENTICATE command [RFC 3501].

   In this example, "C:" and "S:" represent a line sent by the client or
   server respectively including a CRLF at the end.  Linebreaks and
   indentation within a "C:" or "S:" are editorial and not part of the
   protocol. The password in this example was "secret".  Note that the
   base64 encoding of the challenges and responses is part of the IMAP4
   AUTHENTICATE command, not part of the Digest specification itself.

    S: * OK elwood.innosoft.com PMDF IMAP4rev1 V6.0-9
    C: c CAPABILITY
    S: * CAPABILITY IMAP4 IMAP4rev1 ACL LITERAL+ NAMESPACE QUOTA
                UIDPLUS AUTH=CRAM-MD5 AUTH=DIGEST-MD5 AUTH=PLAIN
    S: c OK Completed
    C: a AUTHENTICATE DIGEST-MD5
    S: + cmVhbG09ImVsd29vZC5pbm5vc29mdC5jb20iLG5vbmNlPSJPQTZNRzl0
         RVFHbTJoaCIscW9wPSJhdXRoIixhbGdvcml0aG09bWQ1LXNlc3MsY2hh
         cnNldD11dGYtOA==
    C: Y2hhcnNldD11dGYtOCx1c2VybmFtZT0iY2hyaXMiLHJlYWxtPSJlbHdvb2
       QuaW5ub3NvZnQuY29tIixub25jZT0iT0E2TUc5dEVRR20yaGgiLG5jPTAw
       MDAwMDAxLGNub25jZT0iT0E2TUhYaDZWcVRyUmsiLGRpZ2VzdC11cmk9Im
       ltYXAvZWx3b29kLmlubm9zb2Z0LmNvbSIscmVzcG9uc2U9ZDM4OGRhZDkw
       ZDRiYmQ3NjBhMTUyMzIxZjIxNDNhZjcscW9wPWF1dGg=
    S: + cnNwYXV0aD1lYTQwZjYwMzM1YzQyN2I1NTI3Yjg0ZGJhYmNkZmZmZA==
    C:
    S: a OK User logged in
    ---

    The base64-decoded version of the SASL exchange is:




Leach & Newman             Expires: April 2006                 [Page 30]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


    S: realm="elwood.innosoft.com",nonce="OA6MG9tEQGm2hh",qop="auth",
       algorithm=md5-sess,charset=utf-8
    C: charset=utf-8,username="chris",realm="elwood.innosoft.com",
       nonce="OA6MG9tEQGm2hh",nc=00000001,cnonce="OA6MHXh6VqTrRk",
       digest-uri="imap/elwood.innosoft.com",
       response=d388dad90d4bbd760a152321f2143af7,qop=auth
    S: rspauth=ea40f60335c427b5527b84dbabcdfffd

    The password in this example was "secret".

   This example shows the use of the Digest SASL mechanism with the
   ACAP, using the same notational conventions and password as in the
   previous example. Note that ACAP does not base64 encode and uses
   fewer round trips that IMAP4.

    S: * ACAP (IMPLEMENTATION "Test ACAP server") (SASL "CRAM-MD5"
               "DIGEST-MD5" "PLAIN")
    C: a AUTHENTICATE "DIGEST-MD5"
    S: + {94}
    S: realm="elwood.innosoft.com",nonce="OA9BSXrbuRhWay",qop="auth",
       algorithm=md5-sess,charset=utf-8
    C: {206}
    C: charset=utf-8,username="chris",realm="elwood.innosoft.com",
       nonce="OA9BSXrbuRhWay",nc=00000001,cnonce="OA9BSuZWMSpW8m",
       digest-uri="acap/elwood.innosoft.com",
       response=6084c6db3fede7352c551284490fd0fc,qop=auth
    S: a OK (SASL {40}
    S: rspauth=2f0b3d7c3c2e486600ef710726aa2eae) "AUTHENTICATE
    Completed"
    ---

   The server uses the values of all the directives, plus knowledge of
   the users password (or the hash of the user's name, server's realm
   and the user's password) to verify the computations above. If they
   check, then the user has authenticated.
















Leach & Newman             Expires: April 2006                 [Page 31]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


5   References

5.1   Normative references

   [Digest]   Franks, J., et al., "HTTP Authentication: Basic and Digest
              Access Authentication", RFC 2617, June 1999.

   [ISO-8859] ISO-8859. International Standard--Information Processing--
              8-bit Single-Byte Coded Graphic Character Sets --
              Part 1: Latin alphabet No. 1, ISO-8859-1:1987.
              Part 2: Latin alphabet No. 2, ISO-8859-2, 1987.
              Part 3: Latin alphabet No. 3, ISO-8859-3, 1988.
              Part 4: Latin alphabet No. 4, ISO-8859-4, 1988.
              Part 5: Latin/Cyrillic alphabet, ISO-8859-5, 1988.
              Part 6: Latin/Arabic alphabet, ISO-8859-6, 1987.
              Part 7: Latin/Greek alphabet, ISO-8859-7, 1987.
              Part 8: Latin/Hebrew alphabet, ISO-8859-8, 1988.
              Part 9: Latin alphabet No. 5, ISO-8859-9, 1990.

   [RFC 1321] Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321,
              April 1992.

   [RFC 2052] Gulbrandsen, A. and P. Vixie, "A DNS RR for specifying the
              location of services (DNS SRV)", RFC 2052, October 1996.

   [RFC 2104] Krawczyk, H., Bellare, M. and R. Canetti, "HMAC: Keyed-
              Hashing for  Message Authentication", RFC 2104, February
              1997.

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

   [RFC 2222] Melnikov, A. (editor), "Simple Authentication and Security
              Layer (SASL)", draft-ietf-sasl-rfc2222bis-xx.txt, a work
              in progress.

   [RFC 3454] Hoffman, P., Blanchet, M., "Preparation of
              Internationalized Strings ("stringprep")", RFC 3454,
              December 2002.

   [Unicode]  The Unicode Consortium, "The Unicode Standard, Version
              3.2.0", defined by: The Unicode Standard, Version 3.0
              (Reading, MA, Addison-Wesley, 2000.  ISBN 0-201-61633-5),
              as amended by the Unicode Standard Annex #28: Unicode 3.2
              (http://www.unicode.org/reports/tr28/tr28-3.html).

   [UTF-8]    Yergeau, "UTF-8, a transformation format of ISO 10646",
              RFC 2279, Janyary 1998.



Leach & Newman             Expires: April 2006                 [Page 32]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   [USASCII]  US-ASCII. Coded Character Set - 7-Bit American Standard
              Code for Information Interchange. Standard ANSI X3.4-1986,
              ANSI, 1986.

   [SASLPrep] Zeilenga, K., "SASLprep: Stringprep profile for user names
              and passwords", RFC 4013, February 2005.

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

   [AES]      Daemen, J., Rijmen, V., "The Rijndael Block Cipher",
              http://csrc.nist.gov/encryption/aes/rijndael/Rijndael.pdf,
              3rd September 1999.

   [RFC2743] <<TBD>>

5.2   Informative references

   [RFC 2195] Klensin, J., Catoe, R. and P. Krumviede, "IMAP/POP
              AUTHorize Extension for Simple Challenge/Response", RFC
              2195, September 1997.

   [MD5]      Kaliski, B.,Robshaw, M., "Message Authentication with
              MD5", CryptoBytes, Sping 1995, RSA Inc,
              (http://www.rsa.com/rsalabs/pubs/cryptobytes/spring95/md5.htm)

   [ABNF] Crocker, D. (Ed.) and P. Overell , "Augmented BNF for Syntax
              Specifications: ABNF", RFC 4234, October 2005.

   [RFC 2078] Linn, J., "Generic Security Service Application Program
              Interface, Version 2", RFC 2078, January 1997.

   [RFC 3501] Crispin, M., "Internet Message Access Protocol - Version
              4rev1", RFC 3501, March 2003.

   [RFC 2244] Newman, C., Myers, J., "ACAP -- Application Configuration
              Access Protocol", RFC 2244, November 1997.

   [RFC 2616] Fielding, R., Gettys, J., Mogul, J., Frystyk, H.,
              Masinter, L., Leach, P., Berners-Lee, T., "Hypertext
              Transfer Protocol -- HTTP/1.1", RFC 2616, June 1999.

   [TLS-CBC]  Moeller, B., "Security of CBC Ciphersuites in SSL/TLS:
              Problems and Countermeasures",
              http://www.openssl.org/~bodo/tls-cbc.txt.

   [CBCATT]   Canvel, B., "Password Interception in a SSL/TLS Channel",



Leach & Newman             Expires: April 2006                 [Page 33]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


              published 2003-02-20:
              http://lasecwww.epfl.ch/memo_ssl.shtml

   [VAUDENAY] Serge Vaudenay, "Security Flaws Induced by CBC Padding -
              Applications to SSL, IPSEC, WTLS ...". L.R. Knudsen (Ed.):
              EUROCRYPT 2002, LNCS 2332, pp. 534-545, 2002.

   [RFC 3548] Josefsson, S., "The Base16, Base32, and Base64 Data
              Encodings", RFC 3548, July 2003.

   [IANA-SASL] IANA, "SIMPLE AUTHENTICATION AND SECURITY LAYER (SASL)
              MECHANISMS", <http://www.iana.org/assignments/sasl-
              mechanisms>.






































Leach & Newman             Expires: April 2006                 [Page 34]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


6  IANA Considerations

   It is requested that the SASL Mechanism registry [IANA-SASL] entry for
   the DIGEST-MD5 mechanism be updated to reflect that this document now
   provides its technical specification.

     To: iana@iana.org
     Subject: Updated Registration of SASL mechanism DIGEST-MD5

     Family of SASL mechanisms: NO
     SASL mechanism name: DIGEST-MD5
     Security considerations: See RFC XXXX.
     Published specification (optional, recommended): RFC XXXX
     Person & email address to contact for further information:
              Alexey Melnikov <alexey.melnikov@isode.com>
              IETF SASL WG <ietf-sasl@imc.org>
     Intended usage: COMMON
     Author/Change controller: IESG <iesg@ietf.org>
     Note: Updates existing entry for DIGEST-MD5
































Leach & Newman             Expires: April 2006                 [Page 35]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


7  ABNF

   What follows is the definition of the notation as is used in the
   HTTP/1.1 specification [RFC 2616] and the HTTP authentication
   specification [Digest]; it is reproduced here for ease of reference.
   Since it is intended that a single Digest implementation can support
   both HTTP and SASL-based protocols, the same notation is used in both
   to facilitate comparison and prevention of unwanted differences.
   Since it is cut-and-paste from the HTTP specifications, not all
   productions may be used in this specification.

7.1   Augmented BNF

   All of the mechanisms specified in this document are described in
   both prose and an Augmented Backus-Naur Form (BNF) similar to that
   used by [ABNF]. Implementers will need to be familiar with the
   notation in order to understand this specification.

   The augmented BNF includes the following constructs: <<Can this be
   dropped in favor of ABNF?>>

   name = definition
      The name of a rule is simply the name itself (without any
      enclosing "<" and ">") and is separated from its definition by the
      equal "=" character. White space is only significant in that
      indentation of continuation lines is used to indicate a rule
      definition that spans more than one line. Certain basic rules are
      in uppercase, such as SP, LWS, HT, CRLF, DIGIT, ALPHA, etc. Angle
      brackets are used within definitions whenever their presence will
      facilitate discerning the use of rule names.

   "literal"
      Quotation marks surround literal text. Unless stated otherwise,
      the text is case-insensitive.

   rule1 | rule2
      Elements separated by a bar ("|") are alternatives, e.g., "yes |
      no" will accept yes or no.

   (rule1 rule2)
      Elements enclosed in parentheses are treated as a single element.
      Thus, "(elem (foo | bar) elem)" allows the token sequences
      "elem foo elem" and "elem bar elem".

   *rule
      The character "*" preceding an element indicates repetition. The
      full form is "<n>*<m>element" indicating at least <n> and at most
      <m> occurrences of element. Default values are 0 and infinity so



Leach & Newman             Expires: April 2006                 [Page 36]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


      that "*(element)" allows any number, including zero; "1*element"
      requires at least one; and "1*2element" allows one or two.

   [rule]
      Square brackets enclose optional elements; "[foo bar]" is
      equivalent to "*1(foo bar)".

   N rule
      Specific repetition: "<n>(element)" is equivalent to
      "<n>*<n>(element)"; that is, exactly <n> occurrences of (element).
      Thus 2DIGIT is a 2-digit number, and 3ALPHA is a string of three
      alphabetic characters.

   #rule
      A construct "#" is defined, similar to "*", for defining lists of
      elements. The full form is "<n>#<m>element" indicating at least
      <n> and at most <m> elements, each separated by one or more commas
      (",") and OPTIONAL linear white space (LWS). This makes the usual
      form of lists very easy; a rule such as
        ( *LWS element *( *LWS "," *LWS element ) *LWS )
      can be shown as
        1#element
      Wherever this construct is used, null elements are allowed, but do
      not contribute to the count of elements present. That is,
      "(element), , (element) " is permitted, but counts as only two
      elements.  Therefore, where at least one element is required, at
      least one non-null element MUST be present. Default values are 0
      and infinity so that "#element" allows any number, including zero;
      "1#element" requires at least one; and "1#2element" allows one or
      two.

   ; comment
      A semi-colon, set off some distance to the right of rule text,
      starts a comment that continues to the end of line. This is a
      simple way of including useful notes in parallel with the
      specifications.

   implied *LWS
      The grammar described by this specification is word-based. Except
      where noted otherwise, linear white space (LWS) can be included
      between any two adjacent words (token or quoted-string), and
      between adjacent words and separators, without changing the
      interpretation of a field. At least one delimiter (LWS and/or
      separators) MUST exist between any two tokens (for the definition
      of "token" below), since they would otherwise be interpreted as a
      single token.

7.2   Basic Rules



Leach & Newman             Expires: April 2006                 [Page 37]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   The following rules are used throughout this specification to
   describe basic parsing constructs. The US-ASCII coded character set
   is defined by ANSI X3.4-1986 [USASCII].

       OCTET          = <any 8-bit character>
       CHAR           = <any US-ASCII character (octets 0 - 127)>
       UPALPHA        = <any US-ASCII uppercase letter "A".."Z">
       LOALPHA        = <any US-ASCII lowercase letter "a".."z">
       ALPHA          = UPALPHA | LOALPHA
       DIGIT          = <any US-ASCII digit "0".."9">
       CTL            = <any US-ASCII control character
                        (octets 0 - 31) and DEL (127)>
       CR             = <US-ASCII CR, carriage return (13)>
       LF             = <US-ASCII LF, linefeed (10)>
       SP             = <US-ASCII SP, space (32)>
       HT             = <US-ASCII HT, horizontal-tab (9)>
       <">            = <US-ASCII double-quote mark (34)>
       TEXTCHAR       = <any OCTET except CTLs, but including HT>
       CRLF           = CR LF

   All linear white space, including folding, has the same semantics as
   SP.  A recipient MAY replace any linear white space with a single SP
   before interpreting the field value or forwarding the message
   downstream.

       LWS            = [CRLF] 1*( SP | HT )

   The TEXT rule is only used for descriptive field contents and values
   that are not intended to be interpreted by the message parser. Words
   of TEXT contains characters either from ISO-8859-1 [ISO-8859]
   character set or UTF-8 [UTF-8].

       TEXT           = <any *OCTET except CTLs,
                        but including LWS>

   A CRLF is allowed in the definition of TEXT only as part of a header
   field continuation. It is expected that the folding LWS will be
   replaced with a single SP before interpretation of the TEXT value.

   Hexadecimal numeric characters are used in several protocol elements.

       HEX            = "A" | "B" | "C" | "D" | "E" | "F"
                      | "a" | "b" | "c" | "d" | "e" | "f" | DIGIT

   Many HTTP/1.1 header field values consist of words separated by LWS
   or special characters. These special characters MUST be in a quoted
   string to be used within a parameter value.




Leach & Newman             Expires: April 2006                 [Page 38]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


       token          = 1*TOKENCHAR
       separators     = "(" | ")" | "<" | ">" | "@"
                      | "," | ";" | ":" | "\" | <">
                      | "/" | "[" | "]" | "?" | "="
                      | "{" | "}" | SP | HT
       TOKENCHAR      = <any CHAR except CTLs or separators>

   A string of text is parsed as a single word if it is quoted using
   double-quote marks.

       quoted-string  = ( <"> qdstr-val <"> )
       qdstr-val      = *( qdtext | quoted-pair )
       qdtext         = <any TEXTCHAR except <"> and "\">

   Note that LWS is NOT implicit between the double-quote marks (<">)
   surrounding a qdstr-val and the qdstr-val; any LWS will be considered
   part of the qdstr-val.  This is also the case for quotation marks
   surrounding any other construct.

   The backslash character ("\") MAY be used as a single-character
   quoting mechanism only within qdstr-val and comment constructs.

       quoted-pair    = "\" CHAR

   The value of this construct is CHAR. Note that an effect of this rule
   is that backslash itself MUST be quoted.

























Leach & Newman             Expires: April 2006                 [Page 39]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


8  Sample Code

   The sample implementation in [Digest] also applies to DIGEST-MD5.

   The following code implements the conversion from UTF-8 to 8859-1 if
   necessary.

    /* if the string is entirely in the 8859-1 subset of UTF-8, then
     * translate to 8859-1 prior to MD5
     */
    void MD5_UTF8_8859_1(MD5_CTX *ctx, const unsigned char *base,
        int len)
    {
        const unsigned char *scan, *end;
        unsigned char cbuf;

        end = base + len;
        for (scan = base; scan < end; ++scan) {
            if (*scan > 0xC3) break; /* abort if outside 8859-1 */
            if (*scan >= 0xC0 && *scan <= 0xC3) {
                if (++scan == end || *scan < 0x80 || *scan > 0xBF)
                    break;
            }
        }
        /* if we found a character outside 8859-1, don't alter string
         */
        if (scan < end) {
            MD5Update(ctx, base, len);
            return;
        }

        /* convert to 8859-1 prior to applying hash
         */
        do {
            for (scan = base; scan < end && *scan < 0xC0; ++scan)
                ;
            if (scan != base) MD5Update(ctx, base, scan - base);
            if (scan + 1 >= end) break;
            cbuf = ((scan[0] & 0x3) << 6) | (scan[1] & 0x3f);
            MD5Update(ctx, &cbuf, 1);
            base = scan + 2;
        } while (base < end);
    }








Leach & Newman             Expires: April 2006                 [Page 40]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


9  Authors' Addresses

   Paul Leach
   Microsoft
   1 Microsoft Way
   Redmond, WA 98052, USA

   EMail: paulle@microsoft.com


   Chris Newman
   Sun Microsystems
   1050 Lakes Drive
   West Covina, CA 91790, USA

   EMail: Chris.Newman@Sun.COM


   Alexey Melnikov
   Isode Ltd.
   5 Castle Business Village,
   36 Station Road,
   Hampton,
   Middlesex,
   TW12 2BX,
   United Kingdom

   Email: Alexey.Melnikov@isode.com


10  Acknowledgements

   The following people had substantial contributions to the development
   and/or refinement of this document:

   Lawrence Greenfield
   John Gardiner Myers
   Simon Josefsson
   RL Bob Morgan
   Jeff Hodges
   Claus Assmann
   Tony Hansen
   Ken Murchison
   Sam Hartman
   Kurt D. Zeilenga
   Hallvard B. Furuseth
   Abhijit Menon-Sen
   Tom Yu



Leach & Newman             Expires: April 2006                 [Page 41]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


   as well as other members of the SASL mailing list.

   The text used is section 3.8 was taken from [TLS-CBC] by Bodo
   Moeller.















































Leach & Newman             Expires: April 2006                 [Page 42]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


11  Full Copyright Statement

   Copyright (C) The Internet Society (2005).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
   ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED,
   INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
   INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Acknowledgement

   Funding for the RFC Editor function is currently provided by the
   Internet Society.

12  Intellectual Property

   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at ietf-
   ipr@ietf.org.







Leach & Newman             Expires: April 2006                 [Page 43]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


Appendix A: Changes from 2831

   1). Fixed various typos in formulas.

   2). Dropped DES as mandatory to implement cipher (rc4 is mandatory to
       implement). Removed "des" and "3des" ciphers because of known
       interoperability problems and vulnerability to CBC mode attack.

   3). Tighten ABNF. Fixed some bugs.

   4). Clarified nc-value verification and which side is aborting
       exchange.

   5). Added text saying that for interoperability
       username/password/realm SHOULD be prepared using the "SASLPrep"
       profile [SASLPrep] of the "stringprep" algorithm [RFC 3454].

   6). Clarified that unquoted version of the username, etc. used in A1
       calculation.

   7). Various cleanup to References section. Split all references to
       Normative and Informative.

   8). Added minimal and maximal limits on maxbuf. Clarified how to
       calculate "maximal sender size".

   9). Change ABNF for host to allow for IPv6 addresses. ABNF now
       references RFC 3986.

   10). Added man-in-the-middle considerations for ciphers.

   11). Clarified how sequence counters are updated.

   12). Addition warnings about preventing reply/redirection attacks.

   13). Specified that "charset" directive affects "realm" and doesn't
        affect "authzid".

   14). Removed text that described that "authzid" is in Unicode in
        Normalization Form KC, encoded as UTF-8.

   15). Clarified that rc4 state is not reset between two consecutive
        sent/received buffers of protected data.

   16). Allow for extensibility in step 3. Use "auth-info" as in RFC
        2617.

   17). Prohibit an empty authzid, as this caused interoperability



Leach & Newman             Expires: April 2006                 [Page 44]


INTERNET DRAFT          DIGEST-MD5 SASL Mechanism           October 2005


        problems.

   18). Added AES cipher defined in "AES Ciphersuite for DIGEST-MD5 SASL
        mechanism" document (expired draft-ietf-sasl-digest-aes-00.txt).
        Use explicit IV with aes cipher in CBC mode.

   18). Cleaned up Confidentiality protection section. Added step by
        step exlanation how CBC mode is used.

   19). Clarified client behavior, if it recognizes no ciphers.

   20). Clarified that the server is not required to advertise all
        realms it supports.

   21). Clarified how UIs should present realms.

   22). Changed some informative text to normative MUST/SHOULDs.

   23). Changed nonce/cnonce to allow for channel bindings.

   24). Replace RFC 822 ABNF with [ABNF].
        (in progress)

   And other minor text clarifications.


Appendix B: Open Issues/ToDo List

   1). Normative vs. Informative references must be carefully rechecked.

   2). Replace ABNF with the reference to RFC 2234bis?

   3). Resolve ISO-8859-1 and SASLPrep interaction issue as reported by
       Simon Josefsson.

   4). Can we drop ISO-8859-1 stuff in favor of UTF-8? This is what
       people have suggested for HTTP Digest.

   5). "SHOULD SASLprep" / "MUST fail authentication exchange" language
       should be checked for consistency. Also, should SASLPrep be
       applied BEFORE or AFTER checking for ISO-8859-1 subset? (I think
       it should be applied before)

   6). Pick a way to fix CBC mode attack.







Leach & Newman             Expires: April 2006                 [Page 45]