Network Working Group
Request for Comments: 4453
Category: Informational
J. Rosenberg
Cisco Systems
G. Camarillo, Ed.
D. Willis
Cisco Systems
April 2006

Requirements for Consent-Based Communications

in the Session Initiation Protocol (SIP)

Status of This Memo

This memo provides information for the Internet community. It does not specify an Internet standard of any kind. Distribution of this memo is unlimited.

Copyright Notice

Copyright © The Internet Society (2006).


The Session Initiation Protocol (SIP) supports communications across many media types, including real-time audio, video, text, instant messaging, and presence. In its current form, it allows session invitations, instant messages, and other requests to be delivered from one party to another without requiring explicit consent of the recipient. Without such consent, it is possible for SIP to be used for malicious purposes, including spam and denial-of-service attacks. This document identifies a set of requirements for extensions to SIP that add consent-based communications.

Table of Contents

   1. Introduction ....................................................2
   2. Problem Statement ...............................................2
   3. Requirements ....................................................4
   4. Security Considerations .........................................5
   5. References ......................................................6
      5.1. Normative References .......................................6
      5.2. Informational References ...................................6

1. Introduction

The Session Initiation Protocol (SIP) [1] supports communications across many media types, including real-time audio, video, text, instant messaging, and presence. This communication is established by the transmission of various SIP requests (such as INVITE and MESSAGE [3]) from an initiator to the recipient, with whom communication is desired. Although a recipient of such a SIP request can reject the request, and therefore decline the session, a SIP network will deliver a SIP request to the recipient without their explicit consent.

Receipt of these requests without explicit consent can cause a number of problems in SIP networks. These include amplification attacks. These problems have plagued email. At the time of this writing, most SIP services are not interconnected, so the incidence of amplification attacks directed at SIP services is low compared to the same attacks on email services. The SIPPING working group believes it is necessary to address these attacks proactively so the attacks do not become as burdensome as attacks on email have become.

This document elaborates on the problems posed by the current open model in which SIP was designed, and then goes on to define a set of requirements for adding a consent framework to SIP.

2. Problem Statement

In SIP networks designed according to the principles of RFC 3261 [1] and RFC 3263 [2], anyone on the Internet can create and send a SIP request to any other SIP user, by identifying that user with a SIP Uniform Resource Identifier (URI). The SIP network will usually deliver this request to the user identified by that URI. It is possible, of course, for network services, such as call screening, to block such messaging from occurring, but this is not widespread and certainly not a systematic solution to the problem under consideration here.

Once the SIP request is received by the recipient, the user agent typically takes some kind of automated action to alert the user about receipt of the message. For INVITE requests, this usually involves delivering an audible alert (e.g., "ringing the phone"), or a visual alert (e.g., creating a screen pop-up window). These indicators frequently convey the subject of the call and the identity of the caller. Due to the real-time nature of the session, these alerts are typically disruptive in nature, so as to get the attention of the user.

For MESSAGE requests, the content of the message is usually rendered to the user.

SUBSCRIBE [4] requests do not normally get delivered to the user agents residing on a user's devices. Rather, they are normally processed by network-based state agents. The watcher information event package allows a user to find out that such requests were generated for them, affording the user the opportunity to approve or deny the request. As a result, SUBSCRIBE processing, and most notably presence, already has a consent-based operation. Nevertheless, this already-existing consent mechanism for SIP subscriptions does not protect network agents against denial-of- service (DoS) attacks.

A problem that arises when requests can be delivered to user agents directly, without their consent, is amplification attacks. SIP proxies provide a convenient relay point for targeting a message to a particular user or IP address and, in particular, forwarding to a recipient that is often not directly reachable without usage of the proxy. Some SIP proxy servers forward a single request to several instances or contacts for the same user or resource. This process is called "forking". Another type of SIP server provides the SIP URI- list service [5], which sends a new copy of the same request to each recipient in the URI-list. Examples of URI-list services are subscriptions to resource lists [6], dial-out conference servers [8], and MESSAGE URI-list services [7]. A SIP URI-list service could be used as an amplifier, allowing a single SIP request to flood a single target host or network. For example, a user can create a resource list with 100 entries, each of which is a URI of the form "sip:identifier@target-IP", where target-IP is the IP address to which the attack is to be directed. Sending a single SIP SUBSCRIBE request to such a list will cause the resource list server to generate 100 SUBSCRIBE requests, each to the IP address of the target, which does not even need to be a SIP node.

Note that the target-IP does not need to be the same in all the URIs in order to attack a single machine. For example, the target-IP addresses may all belong to the same subnetwork, in which case the target of the attack would be the access router of the subnetwork.

In addition to launching DoS attacks, attackers could also use SIP URI-list servers as amplifiers to deliver spam. For INVITE requests, this takes the form of typical "telemarketer" calls. A user might receive a stream of never-ending requests for communications, each of them disrupting the user and demanding their attention. For MESSAGE requests, the problem is even more severe. The user might receive a never-ending stream of visual alerts (e.g., screen pop-up windows) that deliver unwanted, malicious, or otherwise undesired content.

Both amplification attacks related to spam and DoS can be alleviated by adding a consent-based communications framework to SIP. Such a framework keeps servers from relaying messages to users without their consent.

The framework for SIP URI-list services [5] identifies amplification attacks as a problem in the context of URI-list services. That framework mandates the use of opt-in lists, which are a form of consent-based communications. The reader can find an analysis on how a consent-based framework helps alleviate spam-related problems in [9].

3. Requirements

The following identify requirements for a solution that provides consent-based communications in SIP. A relay is defined as any SIP server, be it a proxy, Back-to-Back User Agent (B2BUA), or some hybrid, that receives a request and translates the request URI into one or more next-hop URIs to which it then delivers a request.

   REQ 1:  The solution must keep relays from delivering a SIP request
      to a recipient unless the recipient has explicitly granted
      permission to the relay using appropriately authenticated
   REQ 2:  The solution shall prevent relays from generating more than
      one outbound request in response to an inbound request, unless
      permission to do so has been granted by the resource to whom the
      outbound request was to be targeted.  This requirement avoids the
      consent mechanism itself becoming the focus of DoS attacks.
   REQ 3:  The permissions shall be capable of specifying that messages
      from a specific user, identified by a SIP URI that is an Address-
      of-Record (AOR), are permitted.
   REQ 4:  Each recipient AOR must be able to specify permissions
      separately for each SIP service that forwards messages to the
      recipient.  For example, Alice may authorize forwarding to her
      from domain A, but not from domain B.
   REQ 5:  It shall be possible for a user to revoke permissions at any
   REQ 6:  It shall not be required for a user or user agent to store
      information in order to be able to revoke permissions that were
      previously granted for a relay resource.
   REQ 7:  The solution shall work in an inter-domain context, without
      requiring preestablished relationships between domains.
   REQ 8:  The solution shall work for all current and future SIP
   REQ 9:  The solution shall be applicable to forking proxies.
   REQ 10:  The solution shall be applicable to URI-list services, such
      as resource list servers [5], MESSAGE URI-list services [7], and
      conference servers performing dial-out functions [8].
   REQ 11:  In SIP, URI-lists can be stored on the URI-list server or
      provided in a SIP request.  The consent framework must work in
      both cases.
   REQ 12:  The solution shall allow anonymous communications, as long
      as the recipient is willing to accept anonymous communications.
   REQ 13:  If the recipient of a request wishes to be anonymous with
      respect to the original sender, it must be possible for the
      recipient to grant permission for the sender without the original
      sender learning the recipient's identity.
   REQ 14:  The solution shall prevent attacks that seek to undermine
      the underlying goal of consent.  That is, it should not be
      possible to "fool" the system into delivering a request for which
      permission was not, in fact, granted.
   REQ 15:  The solution shall not require the recipient of the
      communications to be connected to the network at the time
      communications are attempted.
   REQ 16:  The solution shall not require the sender of a SIP request
      to be connected at the time that a recipient provides permission.
   REQ 17:  The solution should scale to Internet-wide deployment.

4. Security Considerations

Security has been discussed throughout this document.

5. References

5.1. Normative References

   [1]  Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A.,
        Peterson, J., Sparks, R., Handley, M., and E. Schooler, "SIP:
        Session Initiation Protocol", RFC 3261, June 2002.
   [2]  Rosenberg, J. and H. Schulzrinne, "Session Initiation Protocol
        (SIP): Locating SIP Servers", RFC 3263, June 2002.
   [3]  Campbell, B., Rosenberg, J., Schulzrinne, H., Huitema, C., and
        D. Gurle, "Session Initiation Protocol (SIP) Extension for
        Instant Messaging", RFC 3428, December 2002.

5.2. Informational References

   [4]  Roach, A.B., "Session Initiation Protocol (SIP)-Specific Event
        Notification", RFC 3265, June 2002.
   [5]  Camarillo, G. and A.B. Roach, "Framework and Security
        Considerations for Session Initiation Protocol (SIP) Uniform
        Resource Identifier (URI)-List Services", Work in Progress,
        January 2006.
   [6]  Roach, A.B., Rosenberg, J., and B. Campbell, "A Session
        Initiation Protocol (SIP) Event Notification Extension for
        Resource Lists", Work in Progress, January 2005.
   [7]  Garcia-Martin, M. and G. Camarillo, "Multiple-Recipient MESSAGE
        Requests in the Session Initiation Protocol (SIP)", Work in
        Progress, February 2006.
   [8]  Camarillo, G. and A. Johnston, "Conference Establishment Using
        Request-Contained Lists in the Session Initiation Protocol
        (SIP)", Work in Progress, February 2006.
   [9]  Rosenberg, J., "The Session Initiation Protocol (SIP) and Spam",
        Work in Progress, July 2005.

Authors' Addresses

   Jonathan Rosenberg
   Cisco Systems
   600 Lanidex Plaza
   Parsippany, NJ  07054
   Phone: +1 973 952-5000
   Gonzalo Camarillo (Editor)
   Hirsalantie 11
   Jorvas  02420
   Dean Willis
   Cisco Systems
   2200 E. Pres. George Bush Turnpike
   Richardson, TX  75082

Full Copyright Statement

Copyright © The Internet Society (2006).

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.


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

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


Funding for the RFC Editor function is provided by the IETF Administrative Support Activity (IASA).