\
[Search] [txt|pdfized|bibtex] [Tracker] [Email] [Nits]
Versions: 00 01 02                                                      
Network Working Group                                       Y. Pettersen
Internet-Draft                                        Opera Software ASA
Updates: RFC2109, RFC2965                                  March 1, 2010
(if approved)
Intended status: Standards Track
Expires: September 2, 2010


               Identifying origin server of HTTP Cookies
                    draft-pettersen-cookie-origin-00

Abstract

   HTTP Cookies, as originally defined by Netscape in [NETSC] andas
   later updated by [RFC2109] and [RFC2965] left unaddressed the issue
   of how to restrict which domains a server can set a cookie for, which
   is particularly a problem for servers hosted in top level domains
   have subdomains that are controlled by registries, not domain owners,
   e.g. co.uk and city.state.us domains.  In such situations, unless the
   client uses some kind of domain black-list it is possible for a
   malicious server to set cookies that the client will send to all
   servers in a domain the attacker does not control, and these cookies
   may adversly affect the function of servers receiving them.  The
   primary reason this is a problem is that the server receiving the
   cookie have no way of telling which server originally set it, and is
   therefore not able to reliably distinguish an invalid cookie from a
   valid cookie.  This document proposes a new attribute, "$Origin",
   that is associated with each cookie and sent in all the client's
   Cookie header in the request to the server.  Servers recognizing the
   attribute may then check to see if the cookie was set by a server
   allowed to set cookies for the server, and if necessary ignore the
   cookie.

Requirements Language

   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 [RFC2119].

Status of this Memo

   This Internet-Draft is submitted to IETF in full conformance with the
   provisions of BCP 78 and 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.



Pettersen               Expires September 2, 2010               [Page 1]


Internet-Draft             HTTP Cookie Origin                 March 2010


   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.

   This Internet-Draft will expire on September 2, 2010.

Copyright Notice

   Copyright (c) 2010 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the BSD License.
























Pettersen               Expires September 2, 2010               [Page 2]


Internet-Draft             HTTP Cookie Origin                 March 2010


Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . 4
   2.  The $Origin attribute . . . . . . . . . . . . . . . . . . . . . 4
     2.1.  General syntax  . . . . . . . . . . . . . . . . . . . . . . 4
     2.2.  Updated client processing of received cookies . . . . . . . 4
     2.3.  Updated Cookie header syntax  . . . . . . . . . . . . . . . 5
   3.  Examples  . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
   4.  IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 6
   5.  Security Considerations . . . . . . . . . . . . . . . . . . . . 6
   6.  Acknowledgements  . . . . . . . . . . . . . . . . . . . . . . . 7
   7.  Normative References  . . . . . . . . . . . . . . . . . . . . . 7
   Author's Address  . . . . . . . . . . . . . . . . . . . . . . . . . 7






































Pettersen               Expires September 2, 2010               [Page 3]


Internet-Draft             HTTP Cookie Origin                 March 2010


1.  Introduction

   When originally defined, Netscape's HTTP Cookie specification [NETSC]
   did not extensively specify how clients should check the domain of
   cookies, and although [RFC2109] and [RFC2965] did put restrictions on
   the domains a server setting cookies could set, these policies have
   not been widely implemented, and are also not able to protect against
   all possible abuses.

   Clients have attempted to limit this problem by using heuristics and
   domain blacklists to determine which domains they can set cookies
   for, but these workarounds have limits, both in terms of correctness,
   amount of data needed to implement them, and in timeliness of updates
   to the list.

   On the other side, servers have no way to determine whether or not a
   cookie it receives from a client is one of the cookies it sent to the
   client, or if it came from another server, and which server that
   originally set it.  The server may include information in the
   cookie's value to determine correctness, but this does not guard
   against a malicious server using a correctly generated cookie that
   was originally sent to a different client.

   A way to allow servers to learn if the received cookies are valid,
   and not set by an unauthorized server, is to include the name of the
   server setting the cookie in an attribute, "$Origin", associated with
   each cookie value in the Cookie header sent to the server.  This
   attribute would either identify the name of the server that set the
   cookie or, if the name of this server is not known, the domain for
   which the cookie has been set.  This allows the receiving server to
   remove or ignore cookies set by servers not allowed to set cookies
   for its domain, and also to log the information about the incorrectly
   set cookies.


2.  The $Origin attribute

2.1.  General syntax

   This specification uses the same syntax as is used by [RFC2109] ,
   [RFC2965] , and [RFC2616]
      attr        =     token
      value       =     token | quoted-string

2.2.  Updated client processing of received cookies

   When a client receives a Set-Cookie or Set-Cookie header, it will
   process the header as specified by the approiate specification,



Pettersen               Expires September 2, 2010               [Page 4]


Internet-Draft             HTTP Cookie Origin                 March 2010


   either [NETSC] , [RFC2109], or [RFC2965] , but when storing the
   cookie it MUST also register information about the host setting the
   cookie.  This information MUST include the hostname, and MAY include
   parts of, or the entire URI that set the cookie.

2.3.  Updated Cookie header syntax

   This specification updates the Cookie header as sent by the client by
   associating each cookie value with a $Origin attribute that specifies
   where the the cookie came from.

   This specification does not change the way cookies are selected for
   inclusion in the Cookie header.

   The syntax for the header field is:

   cookie          =  "Cookie:" cookie-value 0*(";" cookie-value)
   cookie-value    =  NAME "=" VALUE ";" cookie-origin

   NAME            =  attr
   VALUE           =  value
   cookie-origin   =  "$Origin" "=" <"> http_URL <">


   NAME and VALUE have the same meaning as in [RFC2109] and [RFC2965]

   The http_URL value of the $Origin attribute MUST be the URI of the
   resource setting the cookie, which SHOULD be restricted to the
   default path (remove the query part and the last path segment).  If
   the client does not know the URI that originally set the cookie,such
   as when the cookie was received by a version of the client that does
   not support $Origin, then it MUST instead send a generated default
   URL "http:// "+domainname+"/" where domainname is the name of the
   domain the cookie is set for, preceded by a single period (".") to
   differentiate the domain name from a hostname.

   The http_URL value MUST be encoded as described in [RFC3986] .

   When receiving a cookie header containing $Origin, servers
   recognizing it SHOULD check if the identified host or domain from the
   URI in the argument is acceptable to the server, and if the cookie is
   not from an acceptable host or domain the cookie can be ignored or
   reported to the server administrator.  The server SHOULD also ignore
   all cookies that are not followed by a $Origin attribute, if one
   cookie in the header have a $Origin attribute.

   [[Open issue: An option for cases with unknown origin is to send an
   empty $Origin attribute or, alternatively, no $Origin attribute for



Pettersen               Expires September 2, 2010               [Page 5]


Internet-Draft             HTTP Cookie Origin                 March 2010


   that cookie.  An argument against having a special dot-prefix is that
   these cookies will only exist for a limited time, after a client has
   been updated to set and send $Origin.  The author thinks it is better
   to provide some information to the server about the domain of the
   cookie, than to provide no information.  Either case would require
   special handling in the server, and there are a large number of
   cookies that are configured to live for years]]

   [[Open issue: An alternative requirement for the URI is to include
   all of the original URI, except the query portion]]


3.  Examples

   http://www.example.com/path1/resource?query sets the cookie:

      Set-Cookie: foo=value1; domain=.example.com; path=/

   http://www2.example.com/path2/resource2?query1 sets the cookie

      Set-Cookie: bar=value2; domain=.example.com; path=/

   An unkown server set the cookie

      Set-Cookie: xyz=value3;  domain=.example.com; path=/

   The resulting Cookie header is:

      Cookie:  foo=value1; $Origin="http://www.example.com/path1/";
               bar=value2; $Origin="http://www2.example.com/path2/";
               xyz=value3; $Origin="http://.example.com/"



4.  IANA Considerations

   This document makes no request of IANA.

   Note to RFC Editor: this section may be removed on publication as an
   RFC.


5.  Security Considerations

   This specification is intended to make sharing of cookies across
   domains detectable, whether the sharing is intentional,
   unintentional, or with malicious intent, and can therefore also be
   used to limit the potential for cookie spoofing, as discussed in the



Pettersen               Expires September 2, 2010               [Page 6]


Internet-Draft             HTTP Cookie Origin                 March 2010


   security considerations of [RFC2109] and [RFC2965] .  It is, however,
   still possible for servers within a permitted group of servers to set
   incorrect or malicioius cookies, which might adversly affect other
   servers in the domain.


6.  Acknowledgements


7.  Normative References

   [NETSC]    "Persistent Client State -- HTTP Cookies",
              <http://www.netscape.com/newsref/std/cookie_spec.html>.

              available at
              <http://www.netscape.com/newsref/std/cookie_spec.html>

   [RFC2109]  Kristol, D. and L. Montulli, "HTTP State Management
              Mechanism", RFC 2109, February 1997.

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

   [RFC2616]  Fielding, R., Gettys, J., Mogul, J., Frystyk, H.,
              Masinter, L., Leach, P., and T. Berners-Lee, "Hypertext
              Transfer Protocol -- HTTP/1.1", RFC 2616, June 1999.

   [RFC2965]  Kristol, D. and L. Montulli, "HTTP State Management
              Mechanism", RFC 2965, October 2000.

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


Author's Address

   Yngve N. Pettersen
   Opera Software ASA
   Waldemar Thranes gate 98
   N-0175 OSLO,
   Norway

   Email: yngve@opera.com







Pettersen               Expires September 2, 2010               [Page 7]