HTTP Working Group J. Franks, Northwestern University INTERNET DRAFT P. Hallam-Baker, M.I.T. J. Hostetler, Spyglass, Inc. P. Leach, Microsoft Corporation A. Luotonen, Netscape Communications Corporation E. Sink, Spyglass, Inc. L. Stewart, Open Market, Inc. Expires: May 21, 1998 November 21, 1997 HTTP Authentication: Basic and Digest Access Authentication Status of this Memo This document is an Internet-Draft. 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 made obsolete 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". To learn the current status of any Internet-Draft, please check the "1id-abstracts.txt" listing contained in the Internet-Drafts Shadow Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe), munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), or ftp.isi.edu (US West Coast). Distribution of this document is unlimited. Please send comments to the HTTP working group at . Discussions of the working group are archived at . General discussions about HTTP and the applications which use HTTP should take place on the mailing list. Abstract "HTTP/1.0" includes the specification for a Basic Access Authentication scheme. This scheme is not considered to be a secure method of user authentication (unless used in conjunction with some external secure system such as SSL [5]), as the user name and password are passed over the network as clear text. This document also provides the specification for HTTP's authentication framework, the original Basic authentication scheme and a scheme based INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 on cryptographic hashes, referred to as "Digest Access Authentication". It is therefore intended to also serve as a replacement for RFC 2069.[6] Like Basic, Digest access authentication verifies that both parties to a communication know a shared secret (a password); unlike Basic, this verification can be done without sending the password in the clear, which is Basic's biggest weakness. As with most other authentication protocols, the greatest sources of risks are usually found not in the core protocol itself but in policies and procedures surrounding its use. Franks, et al. [Page 2] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Table of Contents HTTP AUTHENTICATION: BASIC AND DIGEST ACCESS AUTHENTICATION1 Status of this Memo........................................1 Abstract...................................................1 Table of Contents..........................................3 1 Access Authentication .................................5 1.1 Reliance on the HTTP/1.1 Specification ............5 1.2 Access Authentication Framework ...................5 2 Basic Authentication Scheme ...........................6 3 Digest Access Authentication Scheme ...................7 3.1 Introduction ......................................7 3.1.1 Purpose .........................................7 3.1.2 Overall Operation ...............................8 3.1.3 Representation of digest values .................8 3.1.4 Limitations .....................................8 3.2 Specification of Digest Headers ...................9 3.2.1 The WWW-Authenticate Response Header ............9 3.2.2 The Authorization Request Header ...............11 3.2.3 The Authentication-Info Header .................14 3.3 Digest Operation .................................15 3.4 Security Protocol Negotiation ....................16 3.5 Example ..........................................16 3.6 Proxy-Authentication and Proxy-Authorization .....17 4 Security Considerations ..............................18 4.1 Authentication of Clients using Basic Authentication 18 4.2 Authentication of Clients using Digest Authentication 19 4.3 Offering a Choice of Authentication Schemes ......19 4.4 Comparison of Digest with Basic Authentication ...20 4.5 Replay Attacks ...................................20 4.6 Man in the Middle ................................21 4.7 Spoofing by Counterfeit Servers ..................22 4.8 Storing passwords ................................22 4.9 Summary ..........................................23 5 Acknowledgments ......................................23 6 References ...........................................23 7 Authors' Addresses ...................................24 Index.....................................................26 Franks, et al. [Page 3] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Franks, et al. [Page 4] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 1 Access Authentication 1.1 Reliance on the HTTP/1.1 Specification This specification is a companion two the HTTP/1.1 specification [2]. It uses using the extended BNF section 2.1 of that document, and relies on both the BNF defined in that document, and other aspects of the HTTP/1.1 specification. 1.2 Access Authentication Framework HTTP provides a simple challenge-response authentication mechanism which MAY be used by a server to challenge a client request and by a client to provide authentication information. It uses an extensible, case-insensitive token to identify the authentication scheme, followed by a comma-separated list of attribute-value pairs which carry the parameters necessary for achieving authentication via that scheme. auth-scheme = token auth-param = token "=" ( token | quoted-string ) The 401 (Unauthorized) response message is used by an origin server to challenge the authorization of a user agent. This response MUST include a WWW-Authenticate header field containing at least one challenge applicable to the requested resource. The 407 (Proxy Authentication Required) response message is used by a proxy to challenge the authorization of a client and MUST include a Proxy-Authenticate header field containing a challenge applicable to the proxy for the requested resource. challenge = auth-scheme 1*SP 1#auth-param The authentication parameter realm is defined for all authentication schemes: realm = "realm" "=" realm-value realm-value = quoted-string The realm attribute (case-insensitive) is required for all authentication schemes which issue a challenge. The realm value (case- sensitive), in combination with the canonical root URL (see section 5.1.2 of [2]) of the server being accessed, defines the protection space. These realms allow the protected resources on a server to be partitioned into a set of protection spaces, each with its own authentication scheme and/or authorization database. The realm value is a string, generally assigned by the origin server, which may have additional semantics specific to the authentication scheme. A user agent that wishes to authenticate itself with an origin server-- usually, but not necessarily, after receiving a 401 (Unauthorized)--MAY do so by including an Authorization header field with the request. A Franks, et al. [Page 5] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 client that wishes to authenticate itself with a proxy--usually, but not necessarily, after receiving a 407 (Proxy Authentication Required)--MAY do so by including a Proxy-Authorization header field with the request. Both the Authorization field value and the Proxy-Authorization field value consists of credentials containing the authentication information of the client for the realm of the resource being requested. credentials = basic-credentials | auth-scheme #auth-param The protection space determines the domain over which credentials can be automatically applied. If a prior request has been authorized, the same credentials MAY be reused for all other requests within that protection space for a period of time determined by the authentication scheme, parameters, and/or user preference. Unless otherwise defined by the authentication scheme, a single protection space cannot extend outside the scope of its server. If the origin server does not wish to accept the credentials sent with a request, it SHOULD return a 401 (Unauthorized) response. The response MUST include a WWW-Authenticate header field containing at least one (possibly new) challenge applicable to the requested resource. If a proxy does not accept the credentials sent with a request, it SHOULD return a 407 (Proxy Authentication Required). The response MUST include a Proxy-Authenticate header field containing a (possibly new) challenge applicable to the proxy for the requested resource. The HTTP protocol does not restrict applications to this simple challenge-response mechanism for access authentication. Additional mechanisms MAY be used, such as encryption at the transport level or via message encapsulation, and with additional header fields specifying authentication information. However, these additional mechanisms are not defined by this specification. Proxies MUST be completely transparent regarding user agent authentication by origin servers. That is, they MUST forward the WWW- Authenticate and Authorization headers untouched, and follow the rules found in section 14.8 of [2]. Both the Proxy-Authenticate and the Proxy- Authorization header fields are hop-by-hop headers (see section 13.5.1 of [2]). 2 Basic Authentication Scheme The "basic" authentication scheme is based on the model that the client must authenticate itself with a user-ID and a password for each realm. The realm value should be considered an opaque string which can only be compared for equality with other realms on that server. The server will service the request only if it can validate the user-ID and password for the protection space of the Request-URI. There are no optional authentication parameters. Franks, et al. [Page 6] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Upon receipt of an unauthorized request for a URI within the protection space, the origin server MAY respond with a challenge like the following: WWW-Authenticate: Basic realm="WallyWorld" where "WallyWorld" is the string assigned by the server to identify the protection space of the Request-URI. A proxy may respond with the same challenge using the Proxy-Authenticate header field. To receive authorization, the client sends the userid and password, separated by a single colon (":") character, within a base64 [7] encoded string in the credentials. basic-credentials = "Basic" SP base64-user-pass base64-user-pass = user-pass = userid ":" password userid = * password = *TEXT Userids might be case sensitive. If the user agent wishes to send the userid "Aladdin" and password "open sesame", it would use the following header field: Authorization: Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ== A client SHOULD assume that all paths at or deeper than the depth of the last symbolic element in the path field of the Request-URI also are within the protection space specified by the Basic realm value of the current challenge. A client MAY send the corresponding Authorization header with requests for resources in that space without receipt of another challenge from the server. If a client wishes to send the same userid and password to a proxy, it would use the Proxy-Authorization header field. See section 4 for security considerations associated with Basic authentication. 3 Digest Access Authentication Scheme 3.1 Introduction 3.1.1 Purpose The protocol referred to as "HTTP/1.0" includes specification for a Basic Access Authentication scheme[1]. This scheme is not considered to be a secure method of user authentication, as the user name and password are passed over the network in an unencrypted form. This document Franks, et al. [Page 7] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 provides specification for such a scheme, referred to as "Digest Access Authentication". The Digest Access Authentication scheme is not intended to be a complete answer to the need for security in the World Wide Web. This scheme provides no encryption of object content. The intent is simply to create a weak access authentication method, which avoids the most serious flaws of Basic authentication. 3.1.2 Overall Operation Like Basic Access Authentication, the Digest scheme is based on a simple challenge-response paradigm. The Digest scheme challenges using a nonce value. A valid response contains a checksum (by default the MD5 checksum) of the username, the password, the given nonce value, the HTTP method, and the requested URI. In this way, the password is never sent in the clear. Just as with the Basic scheme, the username and password must be prearranged in some fashion which is not addressed by this document. 3.1.3 Representation of digest values An optional header allows the server to specify the algorithm used to create the checksum or digest. By default the MD5 algorithm is used and that is the only algorithm described in this document. For the purposes of this document, an MD5 digest of 128 bits is represented as 32 ASCII printable characters. The bits in the 128 bit digest are converted from most significant to least significant bit, four bits at a time to their ASCII presentation as follows. Each four bits is represented by its familiar hexadecimal notation from the characters 0123456789abcdef. That is, binary 0000 gets represented by the character '0', 0001, by '1', and so on up to the representation of 1111 as 'f'. 3.1.4 Limitations The digest authentication scheme described in this document suffers from many known limitations. It is intended as a replacement for basic authentication and nothing more. It is a password-based system and (on the server side) suffers from all the same problems of any password system. In particular, no provision is made in this protocol for the initial secure arrangement between user and server to establish the user's password. Users and implementors should be aware that this protocol is not as secure as kerberos, and not as secure as any client-side private-key Franks, et al. [Page 8] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 scheme. Nevertheless it is better than nothing, better than what is commonly used with telnet and ftp, and better than Basic authentication. 3.2 Specification of Digest Headers The Digest Access Authentication scheme is conceptually similar to the Basic scheme. The formats of the modified WWW-Authenticate header line and the Authorization header line are specified below. In addition, a new header, Authentication-Info, is specified. 3.2.1 The WWW-Authenticate Response Header If a server receives a request for an access-protected object, and an acceptable Authorization header is not sent, the server responds with a "401 Unauthorized" status code, and a WWW-Authenticate header, which is defined as follows: WWW-Authenticate = "WWW-Authenticate" ":" "Digest" digest-challenge digest-challenge = 1#( realm | [ domain ] | nonce | [ opaque ] |[ stale ] | [ algorithm ] | [ digest-required ]) domain = "domain" "=" <"> URI ( 1*SP URI ) <"> nonce = "nonce" "=" nonce-value nonce-value = quoted-string opaque = "opaque" "=" quoted-string stale = "stale" "=" ( "true" | "false" ) algorithm = "algorithm" "=" ( "MD5" | token ) digest-required = "digest-required" "=" ( "true" | "false" ) The meanings of the values of the parameters used above are as follows: realm A string to be displayed to users so they know which username and password to use. This string should contain at least the name of the host performing the authentication and might additionally indicate the collection of users who might have access. An example might be "registered_users@gotham.news.com". domain A space-separated list of URIs, as specified in RFC XURI [7]. The intent is that the client could use this information to know the set of URIs for which the same authentication information should be sent. The URIs in this list may exist on different servers. If this keyword Franks, et al. [Page 9] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 is omitted or empty, the client should assume that the domain consists of all URIs on the responding server. nonce A server-specified data string which may be uniquely generated each time a 401 response is made. It is recommended that this string be base64 or hexadecimal data. Specifically, since the string is passed in the header lines as a quoted string, the double-quote character is not allowed. The contents of the nonce are implementation dependent. The quality of the implementation depends on a good choice. A recommended nonce would include H(client-IP ":" time-stamp ":" private-key) Where client-IP is the dotted quad IP address of the client making the request, time-stamp is a server-generated time value, private-key is data known only to the server. With a nonce of this form a server would normally recalculate the nonce after receiving the client authentication header and reject the request if it did not match the nonce from that header. In this way the server can limit the reuse of a nonce to the IP address to which it was issued and limit the time of the nonce's validity. Further discussion of the rationale for nonce construction is in section 4.5 below. An implementation might choose not to accept a previously used nonce or a previously used digest to protect against a replay attack. Or, an implementation might choose to use one-time nonces or digests for POST or PUT requests and a time-stamp for GET requests. For more details on the issues involved see section 4 of this document. The nonce is opaque to the client. opaque A string of data, specified by the server, which should be returned by the client unchanged. It is recommended that this string be base64 or hexadecimal data. stale A flag, indicating that the previous request from the client was rejected because the nonce value was stale. If stale is TRUE (in upper or lower case), the client may wish to simply retry the request with a new encrypted response, without reprompting the user for a new username and password. The server should only set stale to true if it receives a request for which the nonce is invalid but with a valid digest for that nonce (indicating that the client knows the correct username/password). algorithm A string indicating a pair of algorithms used to produce the digest and a checksum. If this not present it is assumed to be "MD5". In Franks, et al. [Page 10] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 this document the string obtained by applying the digest algorithm to the data "data" with secret "secret" will be denoted by KD(secret, data), and the string obtained by applying the checksum algorithm to the data "data" will be denoted H(data). For the "MD5" algorithm H(data) = MD5(data) and KD(secret, data) = H(concat(secret, ":", data)) i.e., the digest is the MD5 of the secret concatenated with a colon concatenated with the data. digest-required If the value of the digest-required parameter is "true", then any request with an entity-body (such as a PUT or a POST) for the resource(s) to which this response applies MUST include the "digest" attribute in its Authorization header. If the request has no entity-body (such as a GET) then the digest- required value can be ignored. If the digest-required parameter is not specified, then its value is "false". If the value of the digest-required parameter is "false", then the "digest" attribute is OPTIONAL on requests for the resource(s) to which the response applies. 3.2.2 The Authorization Request Header The client is expected to retry the request, passing an Authorization header line, which is defined as follows. Authorization = "Authorization" ":" "Digest" digest-response Digest-response = 1#( username | realm | nonce | digest-uri | response | [ digest ] | [ algorithm ] | opaque ) username = "username" "=" username-value username-value = quoted-string digest-uri = "uri" "=" digest-uri-value digest-uri-value = request-uri ; As specified by HTTP/1.1 response = "response" "=" response-digest digest = "digest" "=" entity-digest response-digest = <"> *LHEX <"> entity-digest = <"> *LHEX <"> Franks, et al. [Page 11] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 LHEX = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" |"8" | "9" | "a" | "b" | "c" | "d" | "e" | "f" The values of the opaque and algorithm fields must be those supplied in the WWW-Authenticate response header for the entity being requested. If the value of the digest-required parameter is "true", the response to this request MUST either include the "digest" field in its Authentication-Info header or the response should be an error message indicating the server is unable or unwilling to supply this field. In the latter case the requested entity MUST not be returned as part of the response. If the digest-required parameter is not specified in the request, then its value is "false". If the value of the digest-required parameter is "false", then the "digest" attribute is OPTIONAL for the response to this request. The definitions of response-digest and entity-digest above indicate the encoding for their values. The following definitions show how the value is computed: response-digest = <"> < KD ( H(A1), unquoted nonce-value ":" H(A2) ) > <"> A1 = unquoted username-value ":" unquoted realm-value ":" password password = < user's password > A2 = Method ":" digest-uri-value The "username-value" field is a "quoted-string". However, the surrounding quotation marks are removed in forming the string A1. Thus if the Authorization header includes the fields username="Mufasa", realm="myhost@testrealm.com" and the user Mufasa has password "CircleOfLife" then H(A1) would be H(Mufasa:myhost@testrealm.com:CircleOfLife) with no quotation marks in the digested string. No white space is allowed in any of the strings to which the digest function H() is applied unless that white space exists in the quoted strings or entity body whose contents make up the string to be digested. For example, the string A1 illustrated above must be Mufasa:myhost@testrealm.com:CircleOfLife with no white space on either side of the colons. Likewise, the other strings digested by H() must not have white space on either side Franks, et al. [Page 12] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 of the colons which delimit their fields unless that white space was in the quoted strings or entity body being digested. "Method" is the HTTP request method as specified in section 5.1 of [2]. The "request-uri" value is the Request-URI from the request line as specified in section 5.1 of [2]. This may be "*", an "absoluteURL" or an "abs_path" as specified in section 5.1.2 of [2], but it MUST agree with the Request-URI. In particular, it MUST be an "absoluteURL" if the Request-URI is an "absoluteURL". The authenticating server must assure that the document designated by the "uri" parameter is the same as the document served. The purpose of duplicating information from the request URL in this field is to deal with the possibility that an intermediate proxy may alter the client's request. This altered (but presumably semantically equivalent) request would not result in the same digest as that calculated by the client. The optional "digest" field contains a digest of the entity body and some of the associated entity headers. This digest can be useful in both request and response transactions. In a request it can insure the integrity of POST data or data being PUT to the server. In a response it insures the integrity of the served document. The value of the "digest" field is an , which is defined as follows. entity-digest<"> KD (H(A1), unquoted nonce-value ":" Method ":" date ":" entity-info ":" H(entity-body)) <"> ; format is <"> *LHEX <"> date = rfc1123-date ; see section 3.3.1 of[2] entity-info = H( digest-uri-value ":" media-type ":" ; Content-Type, see section 3.7 of [2] *DIGIT ":" ; Content-Length, see 10.12 of [2] content-coding ":" ; Content-Encoding, see 3.5 of [2] last-modified ":" ; last modified date, see 10.25 of [2] expires ; expiration date; see 10.19 of [2] ) last-modified = rfc1123-date ; see section 3.3.1 of [2] expires = rfc1123-date The entity-info elements incorporate the values of the URI used to request the entity as well as the associated entity headers Content-Type, Content-Length, Content-Encoding, Last-Modified, and Expires. These headers are all end-to-end headers (see section 13.5.1 of [2]) which must not be modified by proxy Franks, et al. [Page 13] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 caches. The "entity-body" is as specified by section 10.13 of [2] or RFC 1864. The content length MUST always be included. The HTTP/1.1 spec requires that content length is well defined in all messages, whether or not there is a Content-Length header. Note that not all entities will have an associated URI or all of these headers. For example, an entity which is the data of a POST request will typically not have a digest-uri-value or Last- modified or Expires headers. If an entity does not have a digest- uri-value or a header corresponding to one of the entity-info fields, then that field is left empty in the computation of entity-info. All the colons specified above are present, however. For example the value of the entity-info associated with POST data which has content-type "text/plain", no content-encoding and a length of 255 bytes would be H(:text/plain:255:::). Similarly a request may not have a "Date" header. In this case the date field of the entity-digest should be empty. In the entity-info and entity-digest computations, except for the blank after the comma in "rfc1123-date", there must be no white space between "words" and "separators", and exactly one blank between "words" (see section 2.2 of [2]). Implementers should be aware of how authenticated transactions interact with proxy caches. The HTTP/1.1 protocol specifies that when a shared cache (see section 13.10 of [2]) has received a request containing an Authorization header and a response from relaying that request, it MUST NOT return that response as a reply to any other request, unless one of two Cache-Control (see section 14.9 of [2]) directives was present in the response. If the original response included the "must-revalidate" Cache- Control directive, the cache MAY use the entity of that response in replying to a subsequent request, but MUST first revalidate it with the origin server, using the request headers from the new request to allow the origin server to authenticate the new request. Alternatively, if the original response included the "public" Cache-Control directive, the response entity MAY be returned in reply to any subsequent request. 3.2.3 The Authentication-Info Header When authentication succeeds, the server may optionally provide a Authentication-Info header indicating that the server wants to communicate some information regarding the successful authentication (such as an entity digest or a new nonce to be used for the next transaction). It has two fields, digest and nextnonce. Both are optional. AuthenticationInfo = "Authentication-Info" ":" 1#( digest | nextnonce ) Franks, et al. [Page 14] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 nextnonce = "nextnonce" "=" nonce-value digest = "digest" "=" entity-digest The optional digest allows the client to verify that the body of the response has not been changed en-route. The server would probably only send this when it has the document and can compute it. The server would probably not bother generating this header for CGI output. The value of the "digest" is an which is computed as described above. The value of the nextnonce parameter is the nonce the server wishes the client to use for the next authentication response. Note that either field is optional. In particular the server may send the Authentication-Info header with only the nextnonce field as a means of implementing one-time nonces. If the nextnonce field is present the client is strongly encouraged to use it for the next WWW- Authenticate header. Failure of the client to do so may result in a request to re-authenticate from the server with the "stale=TRUE ". The Authentication-Info header is allowed in the trailer of an HTTP message transferred via chunked transfer-coding. 3.3 Digest Operation Upon receiving the Authorization header, the server may check its validity by looking up its known password which corresponds to the submitted username. Then, the server must perform the same MD5 operation performed by the client, and compare the result to the given response-digest. Note that the HTTP server does not actually need to know the user's clear text password. As long as H(A1) is available to the server, the validity of an Authorization header may be verified. A client may remember the username, password and nonce values, so that future requests within the specified may include the Authorization header preemptively. The server may choose to accept the old Authorization header information, even though the nonce value included might not be fresh. Alternatively, the server could return a 401 response with a new nonce value, causing the client to retry the request. By specifying stale=TRUE with this response, the server hints to the client that the request should be retried with the new nonce, without reprompting the user for a new username and password. The opaque data is useful for transporting state information around. For example, a server could be responsible for authenticating content which actually sits on another server. The Franks, et al. [Page 15] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 first 401 response would include a domain field which includes the URI on the second server, and the opaque field for specifying state information. The client will retry the request, at which time the server may respond with a 301/302 redirection, pointing to the URI on the second server. The client will follow the redirection, and pass the same Authorization header, including the data which the second server may require. As with the basic scheme, proxies must be completely transparent in the Digest access authentication scheme. That is, they must forward the WWW-Authenticate, Authentication-Info and Authorization headers untouched. If a proxy wants to authenticate a client before a request is forwarded to the server, it can be done using the Proxy-Authenticate and Proxy-Authorization headers described in section 3.6 below. 3.4 Security Protocol Negotiation It is useful for a server to be able to know which security schemes a client is capable of handling. It is possible that a server may want to require Digest as its authentication method, even if the server does not know that the client supports it. A client is encouraged to fail gracefully if the server specifies any authentication scheme it cannot handle. 3.5 Example The following example assumes that an access-protected document is being requested from the server. The URI of the document is "http://www.nowhere.org/dir/index.html". Both client and server know that the username for this document is "Mufasa", and the password is "CircleOfLife". The first time the client requests the document, no Authorization header is sent, so the server responds with: HTTP/1.1 401 Unauthorized WWW-Authenticate: Digest realm="testrealm@host.com", nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", opaque="5ccc069c403ebaf9f0171e9517f40e41" The client may prompt the user for the username and password, after which it will respond with a new request, including the following Authorization header: Authorization: Digest username="Mufasa", Franks, et al. [Page 16] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 realm="testrealm@host.com", nonce="dcd98b7102dd2f0e8b11d0f600bfb0c093", uri="/dir/index.html", response="1949323746fe6a43ef61f9606e7febea", opaque="5ccc069c403ebaf9f0171e9517f40e41" 3.6 Proxy-Authentication and Proxy-Authorization The digest authentication scheme may also be used for authenticating users to proxies, proxies to proxies, or proxies to end servers by use of the Proxy-Authenticate and Proxy- Authorization headers. These headers are instances of the general Proxy-Authenticate and Proxy-Authorization headers specified in sections 10.30 and 10.31 of the HTTP/1.1 specification [2] and their behavior is subject to restrictions described there. The transactions for proxy authentication are very similar to those already described. Upon receiving a request which requires authentication, the proxy/server must issue the "HTTP/1.1 401 Unauthorized" header followed by a "Proxy-Authenticate" header of the form Proxy-Authentication = "Proxy-Authentication" ":" "Digest" digest-challenge where digest-challenge is as defined above in section 2.1. The client/proxy must then re-issue the request with a Proxy- Authenticate header of the form Proxy-Authorization = "Proxy-Authorization" ":" digest-response where digest-response is as defined above in section 2.1. When authentication succeeds, the server may optionally provide a Proxy-Authentication-info header of the form Proxy-Authentication-Info = "Proxy-Authentication-Info" ":" nextnonce where nextnonce has the same semantics as the nextnonce field in the Authentication-Info header described above in section 3.2.3. Note that in principle a client could be asked to authenticate itself to both a proxy and an end-server. It might receive an "HTTP/1.1 401 Unauthorized" header followed by both a WWW- Authenticate and a Proxy-Authenticate header. However, it can never receive more than one Proxy-Authenticate header since such headers are only for immediate connections and must not be passed Franks, et al. [Page 17] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 on by proxies. If the client receives both headers, it must respond with both the Authorization and Proxy-Authorization headers as described above, which will likely involve different combinations of username, password, nonce, etc. 4 Security Considerations 4.1 Authentication of Clients using Basic Authentication The Basic authentication scheme is not a secure method of user authentication, nor does it in any way protect the entity, which is transmitted in clear text across the physical network used as the carrier. HTTP does not prevent additional authentication schemes and encryption mechanisms from being employed to increase security or the addition of enhancements (such as schemes to use one-time passwords) to Basic authentication. The most serious flaw in Basic authentication is that it results in the essentially clear text transmission of the user's password over the physical network. It is this problem which Digest Authentication attempts to address. Because Basic authentication involves the clear text transmission of passwords it SHOULD never be used (without enhancements) to protect sensitive or valuable information. A common use of Basic authentication is for identification purposes -- requiring the user to provide a user name and password as a means of identification, for example, for purposes of gathering accurate usage statistics on a server. When used in this way it is tempting to think that there is no danger in its use if illicit access to the protected documents is not a major concern. This is only correct if the server issues both user name and password to the users and in particular does not allow the user to choose his or her own password. The danger arises because naive users frequently reuse a single password to avoid the task of maintaining multiple passwords. If a server permits users to select their own passwords, then the threat is not only illicit access to documents on the server but also illicit access to the accounts of all users who have chosen to use their account password. If users are allowed to choose their own password that also means the server must maintain files containing the (presumably encrypted) passwords. Many of these may be the account passwords of users perhaps at distant sites. The owner or administrator of such a system could conceivably incur liability if this information is not maintained in a secure fashion. Basic Authentication is also vulnerable to spoofing by counterfeit servers. If a user can be led to believe that he is connecting to a host Franks, et al. [Page 18] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 containing information protected by basic authentication when in fact he is connecting to a hostile server or gateway then the attacker can request a password, store it for later use, and feign an error. This type of attack is not possible with Digest Authentication. Server implementers SHOULD guard against the possibility of this sort of counterfeiting by gateways or CGI scripts. In particular it is very dangerous for a server to simply turn over a connection to a gateway. That gateway can then use the persistent connection mechanism to engage in multiple transactions with the client while impersonating the original server in a way that is not detectable by the client. 4.2 Authentication of Clients using Digest Authentication Digest Authentication does not provide a strong authentication mechanism. That is not its intent. It is intended solely to replace a much weaker and even more dangerous authentication mechanism: Basic Authentication. An important design constraint is that the new authentication scheme be free of patent and export restrictions. Most needs for secure HTTP transactions cannot be met by Digest Authentication. For those needs SSL or SHTTP are more appropriate protocols. In particular digest authentication cannot be used for any transaction requiring encrypted content. Nevertheless many functions remain for which digest authentication is both useful and appropriate. 4.3 Offering a Choice of Authentication Schemes An HTTP/1.1 server may return multiple challenges with a 401 (Authenticate) response, and each challenge may use a different scheme. The order of the challenges returned to the user agent is in the order that the server would prefer they be chosen. The server should order its challenges with the "most secure" authentication scheme first. A user agent should choose as the challenge to be made to the user the first one that the user agent understands. When the server offers choices of authentication schemes using the WWW- Authenticate header, the "security" of the authentication is only as good as the security of the weakest of the authentication schemes. A malicious user could capture the set of challenges and try to authenticate him/herself using the weakest of the authentication schemes. Thus, the ordering serves more to protect the user's credentials than the server's information. A possible man-in-the-middle (MITM) attack would be to add a weak authentication scheme to the set of choices, hoping that the client will use one that exposes the user's credentials (e.g. password). For this Franks, et al. [Page 19] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 reason, the client should always use the strongest scheme that it understands from the choices accepted. An even better MITM attack would be to remove all offered choices, and to insert a challenge that requests Basic authentication. For this reason, user agents that are concerned about this kind of attack could remember the strongest authentication scheme ever requested by a server and produce a warning message that requires user confirmation before using a weaker one. A particularly insidious way to mount such a MITM attack would be to offer a "free" proxy caching service to gullible users. 4.4 Comparison of Digest with Basic Authentication Both Digest and Basic Authentication are very much on the weak end of the security strength spectrum. But a comparison between the two points out the utility, even necessity, of replacing Basic by Digest. 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 database whose use is restricted to paying subscribers. With Basic 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. By contrast, with Digest Authentication the eavesdropper only gets access to the transaction in question and not to the user's password. The information gained by the eavesdropper would permit a replay attack, but only with a request for the same document, and even that might be difficult. 4.5 Replay Attacks A replay attack against digest authentication would usually be pointless for a simple GET request since an eavesdropper would already have seen the only document he could obtain with a replay. This is because the URI of the requested document is digested in the client response and the server will only deliver that document. By contrast under Basic Authentication once the eavesdropper has the user's password, any document protected by that password is open to him. A GET request containing form data could only be "replayed" with the identical data. However, this could be problematic if it caused a CGI script to take some action on the server. Franks, et al. [Page 20] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Thus, for some purposes, it is necessary to protect against replay attacks. A good digest implementation can do this in various ways. The server created "nonce" value is implementation dependent, but if it contains a digest of the client IP, a time- stamp, and a private server key (as recommended above) then a replay attack is not simple. An attacker must convince the server that the request is coming from a false IP address and must cause the server to deliver the document to an IP address different from the address to which it believes it is sending the document. An attack can only succeed in the period before the time-stamp expires. Digesting the client IP and time-stamp in the nonce permits an implementation which does not maintain state between transactions. For applications where no possibility of replay attack can be tolerated the server can use one-time response digests which will not be honored for a second use. This requires the overhead of the server remembering which digests have been used until the nonce time-stamp (and hence the digest built with it) has expired, but it effectively protects against replay attacks. Instead of maintaining a list of the values of used digests, a server would hash these values and require re-authentication whenever a hash collision occurs. An implementation must give special attention to the possibility of replay attacks with POST and PUT requests. A successful replay attack could result in counterfeit form data or a counterfeit version of a PUT file. The use of one-time digests or one-time nonces is recommended. It is also recommended that the optional be implemented for use with POST or PUT requests to assure the integrity of the posted data. Alternatively, a server may choose to allow digest authentication only with GET requests. Responsible server implementors will document the risks described here as they pertain to a given implementation. 4.6 Man in the Middle Both Basic and Digest authentication are vulnerable to "man in the middle" attacks, for example, from a hostile or compromised proxy. Clearly, this would present all the problems of eavesdropping. But it could also offer some additional threats. A simple but effective attack would be to replace the Digest challenge with a Basic challenge, to spoof the client into revealing their password. To protect against this attack, clients should remember if a site has used Digest authentication in the past, and warn the user if the site stops using it. It might also be a good idea for the browser to be configured to demand Digest authentication in general, or from specific sites. Franks, et al. [Page 21] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Or, a hostile proxy might spoof the client into making a request the attacker wanted rather than one the client wanted. Of course, this is still much harder than a comparable attack against Basic Authentication. There are several attacks on the "digest" field in the Authentication- Info header. A simple but effective attack is just to remove the field, so that the client will not be able to use it to detect modifications to the response entity. Sensitive applications may wish to allow configuration to require that the digest field be present when appropriate. More subtly, the attacker can alter any of the entity- headers not incorporated in the computation of the digest. The attacker can alter most of the request headers in the client's request, and can alter any response header in the origin-server's reply, except those headers whose values are incorporated into the "digest" field. Alteration of Accept* or User-Agent request headers can only result in a denial of service attack that returns content in an unacceptable media type or language. Alteration of cache control headers also can only result in denial of service. Alteration of Host will be detected, if the full URL is in the response-digest. Alteration of Referer or From is not important, as these are only hints. 4.7 Spoofing by Counterfeit Servers Basic Authentication is vulnerable to 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 request a password, store it away for later use, and feign an error. This type of attack is more difficult with Digest Authentication -- but the client must know to demand that Digest authentication be used, perhaps using some of the techniques described above to counter "man-in- the-middle" attacks. 4.8 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(A1), where H(A1) is the digested value of the username, realm, and password as described above. 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 Franks, et al. [Page 22] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 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 unencrypted 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 which any single user is likely to use. In particular a realm string should include the name of the host doing the authentication. The inability of the client to authenticate the server is a weakness of Digest Authentication. 4.9 Summary By modern cryptographic standards Digest Authentication is weak. But for a large range of purposes it is valuable as a replacement for Basic Authentication. It remedies many, but not all, weaknesses of Basic Authentication. Its strength may vary depending on the implementation. In particular the structure of the nonce (which is dependent on the server implementation) may affect the ease of mounting a replay attack. A range of server options is appropriate since, for example, some implementations may be willing to accept the server overhead of one-time nonces or digests to eliminate the possibility of replay. Others may satisfied with a nonce like the one recommended above restricted to a single IP address and with a limited lifetime. The bottom line is that *any* compliant implementation will be relatively weak by cryptographic standards, but *any* compliant implementation will be far superior to Basic Authentication. 5 Acknowledgments In addition to the authors, valuable discussion instrumental in creating this document has come from Peter J. Churchyard, Ned Freed, and David M. Kristol. Jim Gettys edited this document for its update. 6 References [1] Berners-Lee, T., Fielding, R., and H. Frystyk, "Hypertext Transfer Protocol -- HTTP/1.0", RFC 1945, May 1996. Franks, et al. [Page 23] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 [2] Fielding, R., Gettys, J., Mogul, J. C., Frysyk, H, Berners-Lee, T., " Hypertext Transfer Protocol -- HTTP/1.1", Work In Progress of the HTTP working group, November 1997. [3] Rivest, R., "The MD5 Message-Digest Algorithm", RFC 1321, April 1992. [4] Freed, N., and N. Borenstein. "Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies." RFC 2045, Innosoft, First Virtual, November 1996. [5] Dierks, T. and C. Allen "The TLS Protocol, Version 1.0," Work In Progress of the TLS working group, November, 1997. [6] Franks, J., Hallam-Baker, P., Hostetler, J., Leach, P., Luotonen, A., Sink, E., Stewart, L.," An Extension to HTTP : Digest Access Authentication." RFC 2069, January, 1997. [7] Berners Lee, T, Fielding, R., Masinter, L., "Uniform Resource Identifiers (URI): Generic Syntax and Semantics ," Work in Progress, November, 1997. 7 Authors' Addresses John Franks Professor of Mathematics Department of Mathematics Northwestern University Evanston, IL 60208-2730, USA EMail: john@math.nwu.edu Phillip M. Hallam-Baker Principal Consultant Verisign Inc. One Alewife Center Cambridge, MA 02138, USA EMail: pbaker@verisign.com Jeffery L. Hostetler Senior Software Engineer Spyglass, Inc. 3200 Farber Drive Champaign, IL 61821, USA EMail: jeff@spyglass.com Franks, et al. [Page 24] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Paul J. Leach Microsoft Corporation 1 Microsoft Way Redmond, WA 98052, USA EMail: paulle@microsoft.com Ari Luotonen Member of Technical Staff Netscape Communications Corporation 501 East Middlefield Road Mountain View, CA 94043, USA EMail: luotonen@netscape.com Eric W. Sink Senior Software Engineer Spyglass, Inc. 3200 Farber Drive Champaign, IL 61821, USA EMail: eric@spyglass.com Lawrence C. Stewart Open Market, Inc. 215 First Street Cambridge, MA 02142, USA EMail: stewart@OpenMarket.com Franks, et al. [Page 25] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Index While some care was taken producing this index, there is no guarantee that all occurrences of an index term have been entered into the index. Italics indicate the definition of a term; bold face is used for the definition of a header. credentials, 6 301, 16 13 digest, 11, 12, 13, 14, 15, 21, 22 Digest Access Authentication, 2, 401, 5, 6, 9, 10, 15, 16, 17, 19 8, 9 407, 5, 6 Digest Authentication, 18, 19 411, 6 digest-challenge, 9, 17 digest-required, 9, 11, 12 digest-response, 11, 17 digest-uri, 11 absoluteURL, 13 digest-uri-value, 11, 12, 13, 14 Accept*, 22 domain, 9, 10, 15, 16 Access Authentication, 5 algorithm, 8, 9, 10, 11, 12 AuthenticationInfo, 302, 16 date, 14 Authentication-Info, 9, 12, 14, entity-body, 13, 14 15, 16, 17, 22 entity-digest, 11, 12, 13, 14, 15 Authorization, 5, 6, 7, 9, 11, entity-info, 13, 14 12, 14, 15, 16, 17, 18 expires, 13 auth-param, 5 Expires, 13, 14 auth-scheme, 5 From, 22 base64-user-pass, 7 Basic Access Authentication, 1, 7, 8 Basic authentication, 7, 18, 20 GET, 10, 11, 20, 21 Basic Authentication Scheme, 6 basic-credentials, 7 last-modified, 13 Last-Modified, 13 Cache-Control, 14 LHEX, 11, 12, 13 challenge, 5 content-coding, 13 Content-Encoding, 13 Content-Length, 13 MD5, 8, 9, 10, 11, 15, 24 Content-Type, 13 media-type, 13 Franks, et al. [Page 26] INTERNET-DRAFT HTTP Authentication Friday 21 November 1997 Method, 12, 13 response, 11, 17 MIME, 24 response-digest, 11, 12, 15, 22 must-revalidate, 14 rfc1123-date, 13, 14 , 17 Security Considerations nonce, 8, 9, 10, 11, 14, 15, 16, basic scheme is insecure, 18 17, 18, 21, 23 comparison of digest with basic, nonce-value, 9, 12, 13, 15 20 man in the middle attacks, 21 offering multiple authentication schemes, 19 opaque, 9, 10, 11, 12, 15, 16, 17 replay attacks against digest nextnonce, 14, 15 authentication, 20 spoofing by counterfeit servers, 22 password, 1, 7, 8, 9, 10, 12, 15, digest weak, 23 16, 18, 20, 21, 22, 23 separators, 14 POST, 10, 11, 13, 14, 21 stale, 9, 10, 15 Proxy-Authenticate, 5, 6, 7, 16, 17 Proxy-Authentication, 17 Proxy-Authentication-Info, 17 token, 5 Proxy-Authorization, 6 true, 12 public, 14 PUT, 10, 11, 13, 21 User-Agent, 22 userid, 7 quoted-string, 5, 9, 11, 12 username, 8, 9, 10, 11, 12, 15, 16, 18, 22, 23 username-value, 11, 12 user-pass, 7 realm, 5, 9, 11, 12, 16, 17, 22, 23 realm-value, 5, 12 Referer, 22 words, 14 request-uri, 11, 13 WWW-Authenticate, 5, 6, 7, 9, 12, Request-URI, 6, 7, 13 16, 17, 19 Franks, et al. [Page 27]