SIP WG R. Mahy Internet-Draft Plantronics Updates: 3261 (if approved) V. Gurbani, Ed. Expires: February 22, 2007 Lucent Technologies, Inc./Bell Laboratories B. Tate BroadSoft August 21, 2006 Connection Reuse in the Session Initiation Protocol (SIP) draft-ietf-sip-connect-reuse-06 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. This Internet-Draft will expire on February 22, 2007. Copyright Notice Copyright (C) The Internet Society (2006). Abstract When SIP entities use a connection oriented protocol to send a request, they typically originate their connections from an ephemeral port. The SIP protocol includes mechanisms which insure that responses to a request, and new requests sent in the original Mahy, et al. Expires February 22, 2007 [Page 1] Internet-Draft SIP Connection Reuse August 2006 direction reuse an existing connection. However, new requests sent in the opposite direction are unlikely to reuse the existing connection. This frequently causes a pair of SIP entities to use one connection for requests sent in each direction, and can result in potential scaling and performance problems. This document proposes requirements and a mechanism which address this deficiency in environments where the connection could be opened in either direction. Table of Contents 1. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Applicability Statement . . . . . . . . . . . . . . . . . . 3 3. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 4. Benefits of Connection Reuse . . . . . . . . . . . . . . . . 4 5. Overview of Operation . . . . . . . . . . . . . . . . . . . 5 6. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 9 7. Formal Syntax . . . . . . . . . . . . . . . . . . . . . . . 9 8. Normative Behavior . . . . . . . . . . . . . . . . . . . . . 9 8.1 Client Behavior . . . . . . . . . . . . . . . . . . . . . 10 8.2 Server Behavior . . . . . . . . . . . . . . . . . . . . . 11 9. Security Considerations . . . . . . . . . . . . . . . . . . 13 9.1 Authenticating TLS Connections: Client View . . . . . . . 13 9.2 Authenticating TLS Connections: Server View . . . . . . . 13 9.3 Security Considerations for the TCP Transport . . . . . . 13 10. Support for Virtual Servers . . . . . . . . . . . . . . . . 15 10.1 Virtual Servers and TLS Connections . . . . . . . . . . 15 10.2 Virtual Servers and TCP Connections . . . . . . . . . . 16 11. Connection Reuse and SRV Interaction . . . . . . . . . . . . 17 12. IANA Considerations . . . . . . . . . . . . . . . . . . . . 18 13. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . 18 14. References . . . . . . . . . . . . . . . . . . . . . . . . . 18 14.1 Normative References . . . . . . . . . . . . . . . . . . 18 14.2 Informational References . . . . . . . . . . . . . . . . 19 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . 20 Intellectual Property and Copyright Statements . . . . . . . 21 Mahy, et al. Expires February 22, 2007 [Page 2] Internet-Draft SIP Connection Reuse August 2006 1. Terminology 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 [2]. Additional terminology used in this document: Advertised address The address that occurs in the Via sent-by production rule, including the port number and transport. Alias A transport layer connection associated with a resolved address. Resolved address The network identifiers (IP address, port, transport) associated with a user agent as a result of executing RFC3263 [4] on a Uniform Resource Identifier (URI). 2. Applicability Statement The applicability of the mechanism described in this document is for two adjacent SIP entities to reuse connections when they are agnostic about the direction of the connection, i.e., either end can initiate the connection. SIP entities that can only open a connection in a specific direction -- perhaps because of Network Address Translation (NAT) and firewall reasons -- reuse their connections using the mechanism described in [8]. The connect reuse mechanism described in this document is defined only for Transport Layer Security (TLS) transports. Furthermore, it is assumed that both the peers using TLS possess X.509 certificates. Implementations MUST NOT use this mechanism for the TCP transport due to the possible attacks that can be launched with connection reuse over TCP. Such attacks and alternative methods for connection reuse over TCP are described in Section 9.3. 3. Introduction SIP [1] entities can communicate using either unreliable/ connectionless (e.g., UDP) or reliable/connection-oriented (e.g., TCP, SCTP [14]) transport protocols. When SIP entities use a connection-oriented protocol (such as TCP or SCTP) to send a request, they typically originate their connections from an ephemeral port. In the following example, Entity A listens for SIP requests over TLS [3] on TCP port 5061 (the default port for SIP over TLS over TCP), but uses an ephemeral port (port 8293) for a new connection to Entity B. These entities could be SIP User Agents or SIP Proxy Servers. Mahy, et al. Expires February 22, 2007 [Page 3] Internet-Draft SIP Connection Reuse August 2006 +-----------+ 8293 (UAC) 5061 (UAS) +-----------+ | |--------------------------->| | | Entity | | Entity | | A | | B | | | 5061 (UAS) | | +-----------+ +-----------+ Figure 1: Uni-directional connection for requests from A to B. The SIP protocol includes mechanisms which insure that responses to a request reuse the existing connection which is typically still available, and also includes provisions for reusing existing connections for other requests sent by the originator of the connection. However, new requests sent in the opposite direction -- in the example above, requests from B destined to A -- are unlikely to reuse the existing connection. This frequently causes a pair of SIP entities to use one connection for requests sent in each direction, as shown below. +-----------+ 8293 5061 +-----------+ | |.......................>| | | Entity | | Entity | | A | 5061 9741 | B | | |<-----------------------| | +-----------+ +-----------+ Figure 2: Two connections for requests between A and B. While this is adequate for TCP, and indeed is the only way to securely do connection reuse over that transport (see Section 9.3), TLS connections can be reused since each end can be authenticated when the connection is initially set up. 4. Benefits of Connection Reuse Opening an extra connection where an existing one is sufficient can result in potential scaling and performance problems. Each new connection using TLS requires a TCP 3-way handshake, a handful of round-trips to establish TLS, typically expensive asymmetric authentication and key generation algorithms, and certificate verification. This may lead to a build up of considerable queues as the server CPU saturates by the TLS handshakes it is already performing (Section 6.19 of [9]). Consider the call flow shown below where Proxy A and Proxy B use the Record-Route mechanism to stay involved in a dialog. Proxy B will Mahy, et al. Expires February 22, 2007 [Page 4] Internet-Draft SIP Connection Reuse August 2006 establish a new TLS connection just to send a BYE request. Proxy A Proxy B | | Create connection 1 +---INV--->| | | |<---200---+ Response over connection 1 | | Re-use connection 1 +---ACK--->| | | = = | | |<---BYE---+ Create connection 2 | | Response over +---200--->| connection 2 Figure 3: Multiple connections for requests. Setting up a second connection (from B to A above) for subsequent requests, even requests in the context of an existing dialog (e.g., re-INVITE or BYE after an initial INVITE, or a NOTIFY after a SUBSCRIBE [13] or a REFER [12]), can also cause excessive delay (especially in networks with long round-trip times). Thus, it is advantageous to reuse connections whenever possible. From the user expectation point of view, it is advantageous if the re-INVITEs or UPDATE [10] requests are handled automatically and rapidly in order to avoid media and session state from being out of step. If a re-INVITE requires a new TLS connection, the reINVITE could be delayed by several extra round-trip times. Depending on the round-trip time, this combined delay could be perceptible or even annoying to a human user. This is especially problematic for some common SIP call flows (for example, the recommended example flow in figure number 4 in RFC3725 [11] use many reINVITEs). The mechanism described in this document can mitigate the delays associated with subsequent requests. 5. Overview of Operation This section is tutorial in nature, and does not specify any normative behavior. The act of reusing a connection is initiated by an user agent client (UAC) when it adds an "alias" parameter (defined later) to the Via header. When a user agent server (UAS) receives the request, it Mahy, et al. Expires February 22, 2007 [Page 5] Internet-Draft SIP Connection Reuse August 2006 examines the topmost Via header. If the header contained an "alias" parameter, the UAS establishes a binding such that subsequent requests going to the UAC will reuse the connection; i.e., requests are sent over the established connection. We now explain this working in more detail in the context of communication between two adjacent proxies. Without any loss of generality, it should be clear that the same technique can be used for connection reuse between a UAC and an edge proxy, or between an edge proxy and a UAS, or between an UAC and an UAS. P1 and P2 are proxies responsible for routing SIP requests through user agents that use them as edge proxies (see Figure 4). P1 <===================> P2 p1.example.com p2.example.com (192.0.2.1) (192.0.2.128) +---+ +---+ | | 0---0 0---0 | | |___| /-\ /-\ |___| / / +---+ +---+ / / +----+ +----+ User Agents User Agents Figure 4: Proxy setup. This document is concerned with specifying an extension to SIP for connection reuse at the receiving end; i.e., reusing the connection when P2 wants to send a request to P1. However, it should be sufficiently clear that the sending end (P1) may also reuse a connection previously established downstream (to P2, for instance). With reference to Figure 4, in order for P2 to reuse a connection for requests in the opposite direction, it is important to note that the validation model for requests sent in this direction (i.e., P2 to P1) should be equivalent to the normal "connection in each direction" model, wherein P2 acting as client would open up a new connection in the backwards direction and validate the connection by examining the X.509 certificate presented. The act of reusing a connection must have the desired property that requests get delivered in the reverse direction only if they would have been delivered to the same destination had connection reuse not been employed. To guarantee this property, the X.509 certificate presented by P1 to P2 when a TLS connection is first authenticated must be cached for later use. To aid in connection reuse, a new data structure, the connection alias table (or simply, alias table) is defined in this document and Mahy, et al. Expires February 22, 2007 [Page 6] Internet-Draft SIP Connection Reuse August 2006 it is used to store the resolved addresses. User agents can consult the alias table for an existing connection before opening up a new one. P1 gets a request from one of its upstream user agents, and after performing RFC3263 server selection, arrives at a resolved address of P2. P1 maintains an alias table, and it populates the alias table with the IP address, port number, and transport of P2 as determined through RFC3263 server selection. P1 adds an "alias" parameter to the topmost Via header (inserted by it) before sending the request to P2. The value in the sent-by production rule of the Via header (including the port number), and the transport over which the request was sent becomes the advertised address of P1: Via: SIP/2.0/TLS p1.example.com;branch=z9hG4bKa7c8dze;alias Assuming that P1 does not already have an existing aliased connection with P2, P1 now opens a connection with P2. P2 presents its X.509 certificate to P1 for validation (see Section 9.1). Upon connection authentication and acceptance, P1 adds P2s to its alias table. P1's alias table now looks like: Destination Destination Destination Destination Alias IP Address Port Transport Identity Descriptor ... 192.0.2.128 5061 TLS DNS:p2.example.com, 25 sips:example.com Figure 5: Alias table at the client. Subsequent requests that traverse from P1 to P2 will reuse this connection; i.e., the requests will be sent over the descriptor 25. There are three items of interest in the alias table created at the client: 1. The IP address, port and transport are a result of executing RFC3263 server resolution process on a next hop URI. 2. The entries in the fourth column consists of the identities of the server as asserted in the X.509 certificate presented by the server. These identities are cached by the client after the server has been duly authenticated (see Section 9.1). 3. The entry in the last column is the socket descriptor over which P1, acting as a client, actively opened a TLS connection. At some later time, when P1 gets a request from one of the user agents in its domain, it will reuse the aliased connection accessible through socket descriptor 25 if and only if all of the following conditions hold: Mahy, et al. Expires February 22, 2007 [Page 7] Internet-Draft SIP Connection Reuse August 2006 A. P1 determines through RFC3263 server resolution process that the request should be sent to P2 on port 5061 using TLS, and B. The URI used for RFC3263 server resolution matches one of the identities stored in the cached certificate (fourth column). When the server, P2, receives the request, it may add a "received" parameter to the topmost Via and examines the topmost Via to determine whether P1 supports aliased connections. The Via at P2 now looks like: Via: SIP/2.0/TLS p1.example.com;branch=z9hG4bKa7c8dze;alias; received=192.0.2.1 The presence of the "alias" parameter indicates that P1 does support aliasing. P2 now authenticates the connection (see Section 9.2) and if the authentication was successful, P2 creates an alias to P1 using the advertised address in the topmost Via. P2's alias table looks like: Destination Destination Destination Destination Alias IP Address Port Transport Identity Descriptor ... 192.0.2.1 5061 TLS DNS:p1.example.com, 18 sips:example.com Figure 6: Alias table at the server. There are a few items of interest here: 1. The IP address field is populated with the source address of the client. 2. The port field is populated from the advertised address (topmost Via header), if a port is present in it, or 5061 if it is not. 3. The transport field is populated from the advertised address (topmost Via header). 4. The entries in the fourth column consist of the identities of the client as asserted in the X.509 certificate presented by the client. These identities are cached by the server after the client has been duly authenticated (see Section 9.2). 5. The entry in the last column is the socket descriptor over which the connection was passively accepted. At some later time, when P2 gets a request from one of the user agents in its domain, it will reuse the aliased connection accessible through socket descriptor 18 if and only if all of the following conditions hold: A. P2 determines through RFC3263 server resolution process that the request should be sent to P1 on port 5061 using TLS, and B. The URI used for RFC3263 server resolution matches one of the identities stored in the cached certificate (fourth column). Mahy, et al. Expires February 22, 2007 [Page 8] Internet-Draft SIP Connection Reuse August 2006 6. The network address inserted in the "Destination IP Address" column should be the source address as seen by P2 (i.e., the "received" parameter). It could be the case that the host name of P1 resolves to different IP addresses due to round-robin DNS. However, the aliased connection is to be established with the original sender of the request. 6. Requirements The following are the requirements that motivated this specification: 1. A connection sharing mechanism SHOULD allow SIP entities to reuse existing connections for requests and responses originated from either peer in the connection. 2. A connection sharing mechanism MUST NOT require clients to send all traffic from well-know SIP ports. 3. A connection sharing mechanism MUST NOT require configuring ephemeral port numbers in DNS. 4. A connection sharing mechanism MUST prevent unauthorized hijacking of other connections. 5. Connection sharing SHOULD persist across SIP transactions and dialogs. 6. Connection sharing MUST work across name-based virtual SIP servers. 7. There is no requirement to share a complete path for ordinary connection reuse. Hop-by-hop connection sharing is more appropriate. 7. Formal Syntax The following syntax specification uses the augmented Backus-Naur Form (BNF) as described in RFC 4234 [5]. This document extends the via-params to include a new via-alias defined below. via-params =/ via-alias via-alias = "alias" 8. Normative Behavior This document specifies how to reuse connections. It is RECOMMENDED that servers keep connections up unless they need to reclaim resources, and that clients keep connections up as long as they are needed. Connection reuse works best when the client and the server maintain their connections for long periods of time. SIP entities therefore SHOULD NOT automatically drop connections on completion of a transaction or termination of a dialog. Mahy, et al. Expires February 22, 2007 [Page 9] Internet-Draft SIP Connection Reuse August 2006 When this document is followed to specify connection reuse, an alias is formed at the receiver of a request when it gets a request with the "alias" parameter in the topmost Via header. If the receiver decides to accept the alias, then the alias corresponds to the source IP address, transport, port (if one exists in the Via sent-by, or the default port if it does not), and the identities of the sender as asserted in the certificate of the sender of the request. Whenever the RFC3263 server selection mechanism executed at the receiver results in the choice of this IP address, port, transport, and identity tuple, the alias SHOULD be used instead. Note that at the receiver, the responses are sent over the same connection as specified by RFC3261. The aliasing mechanism at the receiver allows subsequent requests going from the receiver to the original sender of the request to reuse the same connection. An alias is formed at the sender of the request when it executes the RFC3263 server selection mechanism to arrive at an IP address, port, and transport tuple to send a request to. Subsequent requests going to the same resolved address SHOULD use the alias instead. Because an alias connection might be reclaimed during a transaction, clients SHOULD NOT enforce the RFC 3261 requirement of sending CANCEL and ACK (for non 2xx responses) to the same port. If the alias connection no longer exists, the client SHOULD open a new connection to the resolved address and send the CANCEL or ACK there instead. The newly opened connection MAY be inserted into the alias table. 8.1 Client Behavior The proposed mechanism uses a new Via header field parameter. The "alias" parameter is included in a Via header field value to indicate that the client wants to create a transport layer alias. The client places its advertised address in the Via header field value (in the "sent-by" production). The implications of placing an "alias" parameter in the topmost Via header of a request must be understood by the client. Specifically, this means that the client MUST keep the connection open for as long as the resources on the host operating system allow it to, and that it MUST accept requests over this connection -- as opposed to a default listening port -- from its downstream peer. And furthermore, it MUST reuse the connection when subsequent requests in the same or different transactions are destined to the same resolved address. Note that RFC3261 states that a response should arrive over the same connection that was opened for a request. Mahy, et al. Expires February 22, 2007 [Page 10] Internet-Draft SIP Connection Reuse August 2006 Whether or not to allow an aliased connection ultimately depends on the recipient of the request; i.e., the client does not get any confirmation that its downstream peer created the alias, or indeed that it even supports this specification. Thus, clients MUST NOT assume that the acceptance of a request by a server automatically enables connection aliasing. They MUST continue receiving requests on their default port. Clients must be prepared for the case that the connection no longer exists when they are ready to send a subsequent request over it. This may happen if the peer ran out of operating system resources and had to close the connection. In such a case, a new connection MUST be opened to the resolved address and the alias table updated accordingly. For TLS connections, clients MUST authenticate the connection before forming an alias; Section 9.1 discusses the authentication steps in more detail. Once the server has been authenticated, the client MUST cache, in the alias table, the identity (or identities) of the server as they appear in the X.509 certificate subjectAlternativeName extension field. The client must also populate the destination IP address, port, and transport of the server in the alias table; these fields are retrieved from executing RFC3263 server resolution process on the next hop URI. And finally, the client must populate the alias descriptor field with the socket descriptor used to connect to the server. TCP connections, obviously, cannot be authenticated in a similar manner as TLS connections; nor can an identity be derived by inspecting fields of a certificate. For TCP connections, clients MUST use as the destination identity the URI used as input to the RFC3263 server resolution process. Once the alias table has been updated with a resolved address, and the client wants to send a new request in the direction of the server, it should reuse the connection only if all of the following conditions hold: 1. The client uses the RFC3263 resolution on a URI and arrives at a resolved address contained in the alias table, and 2. The URI used for RFC3263 server resolution matches one of the identities stored in the alias table row corresponding to that resolved address. 8.2 Server Behavior When a server receives a request whose topmost Via header contains an "alias" parameter, it signifies that the upstream client will leave the connection open beyond the transaction and dialog lifetime, and Mahy, et al. Expires February 22, 2007 [Page 11] Internet-Draft SIP Connection Reuse August 2006 that subsequent transactions and dialogs that are destined to a resolved address that matches the identifiers in the advertised address in the topmost Via header can reuse this connection. Whether or not to honor an aliased connection ultimately depends on the policies of the server. It MAY choose to honor it, and thereby send subsequent requests over the aliased connection. If the server chooses not to honor an aliased connection, it MUST allow the request to proceed as though the "alias" parameter was not present in the topmost Via header. This assures interoperability with RFC3261 server behavior. Clients should feel comfortable including the "alias" parameter without fear that the server will reject the SIP request because of its presence. Servers MUST be prepared to deal with the case that the aliased connection no longer exist when they are ready to send a subsequent request over it. This may happen if the peer ran out of operating system resources and had to close the connection. In such a case, a new connection MUST be opened to the resolved address and the alias table updated accordingly. If the Via sent-by contains a port, it MUST be used as a destination port. Otherwise the default port is the destination port. Servers must authenticate the connection before forming an alias. Section 9.2 discusses the authentication steps in more detail. The server, if it decides to accept the connection, MUST cache, in the alias table, the identity (or identities) of the client as they appear in the X.509 certificate subjectAlternativeName extension field. The server must also populate the destination IP address, port and transport in the alias table from the topmost Via header (using the ";received" parameter for the destination IP address). If the port number is omitted, a default port number of 5061 is to be used. And finally, the server must populate the alias descriptor field with the socket descriptor used to accept the connection from the client (see Section 5 for the contents of the alias table.) Once the alias table has been updated, and the server wants to send a request in the direction of the client, it should reuse the connection only if all of the following conditions hold: 1. The server, which acts as a client for this transaction, uses the RFC3263 resolution process on a URI and arrives at a resolved address contained in the alias table, and Mahy, et al. Expires February 22, 2007 [Page 12] Internet-Draft SIP Connection Reuse August 2006 2. The URI used for RFC3263 server resolution matches one of the identities stored in the alias table row corresponding to that resolved address. 9. Security Considerations This document presents requirements and a mechanism for reusing existing connections easily. Unauthenticated connection reuse would present many opportunities for rampant abuse and hijacking. Authenticating connection aliases is essential to prevent connection hijacking. For example, a program run by a malicious user of a multiuser system could attempt to hijack SIP requests destined for the well-known SIP port from a large relay proxy. 9.1 Authenticating TLS Connections: Client View When a TLS client establishes a connection with a server, it is presented with the server's X.509 certificate. Authentication proceeds as described in Section 5 of [7]. 9.2 Authenticating TLS Connections: Server View A TLS server conformant to this specification MUST ask for a client certificate; if the client possesses a certificate, it will be presented to the server for mutual authentication, and authentication proceeds as described in Section 6 of [7]. If the client does not present a certificate, the server MUST proceed as if the "alias" parameter was not present in the topmost Via. In this case, the alias table MUST not be updated. 9.3 Security Considerations for the TCP Transport The mechanism for reusing TLS connections MUST NOT be used to reuse TCP connections because there isn't any way to perform the authentication step. Instead, it is RECOMMENDED that TCP peers that want to avail of connection reuse do so such that each peer actively opens up a TCP connection in the direction of the other (as depicted in Figure 2). This manner of opening connections, while still not secure, is at least much more apparent and direct than using the connection reuse mechanism over TCP in an unauthenticated fashion. Connection reuse over TCP is inherently insecure. Because the nature of the aliasing mechanism is such that it redirects requests destined for one port at a host to another port, service hi-jacking can result if adequate care is not taken to ensure that the redirected port is indeed authorized to receive the requests that would normally have gone to another, authorized port. Consider the following scenario to Mahy, et al. Expires February 22, 2007 [Page 13] Internet-Draft SIP Connection Reuse August 2006 understand the service hi-jacking attack that can be mounted when using connection reuse over TCP. A TCP server receives a request with the "alias" parameter as follows (the "received" parameter is added by the server after getting the request): Via: SIP/2.0/TCP uac.example.com;branch=z9hG4bKa7c8dze;alias; received=192.0.4.33 From the server's perspective, its alias table is updated such that whenever a request is destined to 192.0.4.33, port 5060, it will instead be sent to the peer at the end of the aliased connection. The security attack can now be mounted as follows: assume a malware program is running on a multi-user computer. The malware program knows that a user on the computer runs a SIP user agent, but the SIP user agent is currently not active (possibly by scanning ports on the local machine to seek a busy port 5060). Note that the malware program does not need to wait until the legitimate user agent was not running, however, doing so increases the chances that the server will not reject the malware program's request. Once the malware program decides that a legitimate user agent is not running, it sends sends a request to the server with an "alias" parameter. The server believes it is accepting a request from a legitimate user agent and sends subsequent requests to the aliased connection. The SIP service on the computer has now effectively been hi-jacked for the default port. The malware program does not need administrative privileges to execute, and in fact, can masquerade as any user (legitimate or not) of the computer. Later on, when the legitimate user agent is started, it may also send a request with an "alias" parameter to the server, which may detect that it now has two aliased connections. Making matters much worse, it cannot determine which of the two is the legitimate one and may well reject the request from the legitimate user. In another form of this attack, the legitimate user agent may not support connection aliasing, but the malware program may use the mechanism to usurp the SIP service on the computer. In yet another form of an attack, the malware program uses the aliasing mechanism to shortcut registering with a proxy to receive requests. In this case, it sends a request to the edge proxy (who may also substitute as the inbound proxy with access to a location service for that domain). In the request is a bogus request URI that will cause the edge proxy to fail the request, however, the edge proxy keeps the connection open and any subsequent requests destined to that host on the default port are instead sent to the malware Mahy, et al. Expires February 22, 2007 [Page 14] Internet-Draft SIP Connection Reuse August 2006 program. Registration is thus not needed in order to receive incoming requests. HTTP Digest is useful to mitigate only a subset of these attacks over TCP. For instance, HTTP Digest helps in authenticating a user agent to a proxy server before the alias table is updated. However, HTTP Digest is of no help when one proxy desires to enter an aliasing agreement with another downstream proxy. 10. Support for Virtual Servers Virtual servers present special considerations for connection reuse. Under the name-based virtual server scheme, one SIP proxy may host many virtual domains. If adequate defenses are not put in place, a connection opened to a downstream server on behalf of one domain can be usurped by a malicious user in another domain. The Destination Identity column in the alias table has been added to aid in such defenses. If an implementation does not support virtual SIP servers, it MAY omit caching the identities in the alias table; however, if an implementation supports virtual SIP servers, then it MUST cache the identities in the alias table. 10.1 Virtual Servers and TLS Connections To understand the specific problem associated with hijacking a TLS connection when virtual servers are used, consider a proxy P1 that hosts two domains: atlanta.example.com and chicago.example.org. Also assume that the physical IP address of P1 is 192.168.0.1. Incoming requests to all the domains that P1 hosts arrive on port 5061. A user, bob@atlanta.example.com, sends an instant message to a user Alice in a domain not hosted by P1. Alice's proxy establishes an alias to P1, thereby resulting in the following alias table (note: to illustrate the connection hijacking problem associated with virtual servers, the alias table below does not contain the Destination Identity column). Destination Destination Destination Alias IP Address Port Transport Descriptor ... 192.168.0.1 5061 TLS 25 Figure 7: Alias table at Alice's Proxy. At some later time, a user hosted by another virtual domain in P1, bob@chicago.example.org, sends an instant message to Alice. Alice's proxy will get the network identifiers from the topmost Via, and note that they are already in the alias table. Thinking that the newly Mahy, et al. Expires February 22, 2007 [Page 15] Internet-Draft SIP Connection Reuse August 2006 arrived request is intended to replace the old (possibly stale) alias, it may update its alias table with the new descriptor. Some time after that, Alice wants to send an instant message to sips:bob@atlanta.example.com. When RFC3263 resolution is done on sips:atlanta.example.com, the resolved address will match an entry in the alias table. But that entry is now aliased to a connection with bob@chicago.example.org. The end result of all this is that an instant message intended for bob@atlanta.example.com ends up in the inbox of bob@chicago.example.org. It is to alleviate this very problem that the identities from the X.509 certificates are stored in the alias table and used to determine whether or not to reuse a connection. Saving the identities in the alias table mitigates this problem because Alice's proxy will actually form two aliased connections to P1: one row in the table will contain the resolved address of P1 but with an identity corresponding to atlanta.example.com and a second row will contain the same resolved address but with an identity corresponding to chicago.example.org. Now, when Alice's proxy wants to send a request in the backwards direction, it will match the URI used to do RFC3263 resolution to the appropriate identity before reusing the connection. 10.2 Virtual Servers and TCP Connections The same problem described in Section 10.1 occurs for the TCP transport as well, however, the mechanics associated are somewhat different from the TLS case since connection reuse across TCP connections follows the model depicted in Figure 2 (i.e., one connection in each direction). To illustrate the problem nonetheless, consider the same proxy and virtual domains of described in Section 10.1. Alice is interested in sending instant messages to two of her colleagues, sip:bob@atlanta.example.com and sip:bob@chicago.example.org. Unbeknownst to her, atlanta.example.com and chicago.example.org resolve to the same network identifiers when RFC3263 resolution is done on these URIs (they are hosted on P1, a proxy that supports multiple virtual domains at one physical address: 192.168.0.1). Alice sends an instant message to her colleague, sip:bob@atlanta.example.com. This act causes her proxy to open up a TCP connection to P1 and update its alias table as follows: Mahy, et al. Expires February 22, 2007 [Page 16] Internet-Draft SIP Connection Reuse August 2006 Destination Destination Destination Alias IP Address Port Transport Descriptor ... 192.168.0.1 5060 TCP 25 Figure 8: Alias table at Alice's Proxy. At some later time, Alice sends an instant message to her colleague, sip:bob@chicago.example.org. RFC3263 resolution on sip: chicago.example.org will result in a set of network identifiers that are already present in the alias table of Alice's proxy. Thus, Alice's proxy will not open up a new TCP connection, and instead, use the same connection to erroneously deliver the message to the wrong recipient. Unlike the case of TLS connections, TCP connections do not have certificates that can be used to further choose the right aliased connection. Instead of certificates, domain names must be used for TCP as follows: When Alice's proxy sends the request to a URI, a RFC3263 resolution is done on the URI to derive a resolved address. In addition to saving the network identifiers in the alias table, the proxy must also save the URI in the Destination Identity column. That is, if a request was sent to sip:bob@atlanta.example.com, then the alias table will be populated as follows: Destination Destination Destination Destination Alias IP Address Port Transport Identity Descriptor ... 192.168.10.2 5060 TCP atlanta.example.com 32 Figure 9: Alias table at Alice's Proxy with a TCP Destination Identity column. Subsequent requests sent over the aliased connection must ensure that the URI in the Destination Identity column matches the URI where the request is being sent. 11. Connection Reuse and SRV Interaction Connection reuse has an interaction with the DNS SRV load balancing mechanism. To understand the interaction, consider the following figure: Mahy, et al. Expires February 22, 2007 [Page 17] Internet-Draft SIP Connection Reuse August 2006 /+---- S1 +-------+/ | Proxy |------- S2 +-------+\ \+---- S3 Figure 8: Load balancing. Here, the proxy uses DNS SRV to load balance across the three servers, S1, S2, and S3. Using the connect reuse mechanism specified in this document, over time the proxy will maintain a distinct aliased connection to each of the servers. However, once this is done, subsequent traffic is load balanced across the three downstream servers in the normal manner. 12. IANA Considerations This specification defines a new Via header field parameter called "alias" in the "Header Field Parameters and Parameter Values" sub- registry as per the registry created by [6]. The required information is: Header Field Parameter Name Predefined Values Reference ___________________________________________________________________ Via alias No RFCXXXX RFC XXXX [NOTE TO RFC-EDITOR: Please replace with final RFC number of this specification.] 13. Acknowledgments Thanks to Jon Peterson for helpful answers about certificate behavior with SIP, Jonathan Rosenberg for his initial support of this concept, and Cullen Jennings for providing a sounding board for this idea. Other members of the SIP WG that contributed to this document include Jeroen van Bemmel, Keith Drage, Matthew Gardiner, Rajnish Jain, and Rocky Wang. 14. References 14.1 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] Bradner, S., "Key words for use in RFCs to Indicate Requirement Mahy, et al. Expires February 22, 2007 [Page 18] Internet-Draft SIP Connection Reuse August 2006 Levels", RFC 2119, March 1997. [3] Dierks, T. and C. Allen, "The TLS Protocol Version 1.0", RFC 2246, January 1999. [4] Rosenberg, J. and H. Schulzrinne, "Session Initiation Protocol (SIP): Locating SIP Servers", RFC 3263, June 2002. [5] Crocker, D. and P. Overell, "Augmented BNF for Syntax Specifications: ABNF", RFC 4234, October 2005. [6] Camarillo, G., "The Internet Assigned Numbers Authority (IANA) Header Field Paramater Registry for the Session Initiation Protocol (SIP)", BCP 98, RFC 3968, December 2004. [7] Gurbani, V., Jeffrey, A., and S. Lawrence, "Domain Certificates in the Session Initiation Protocol (SIP)", draft-gurbani-sip-domain-certs-03 (work in progress), August 2006. 14.2 Informational References [8] Jennings, C. and R. Mahy, "Managing Client Initiated Connections in the Session Initiation Protocol (SIP)", draft-ietf-sip-outbound-04.txt (work in progress), June 2006. [9] Rescorla, E., "SSL and TLS: Designing and Building Secure Systems", Addison-Wesley Publishing , 2001. [10] Rosenberg, J., "The Session Initiation Protocol (SIP) UPDATE Method", RFC 3311, September 2002. [11] Rosenberg, J., Peterson, J., Schulzrinne, H., and H. Camarillo, "Best Current Practices for Third Party Call Control (3pcc) in the Session Initiation Protocol (SIP)", RFC 3725, April 2004. [12] Sparks, R., "The Session Initiation Protocol (SIP) Refer Method", RFC 3515, April 2003. [13] Roach, A., "The Session Initiation Protocol (SIP)-Specific Event Notification", RFC 3265, June 2002. [14] Stewart, R., Xie, Q., Morneault, K., Sharp, C., Schwarzbauer, H., Taylor, T., Rytina, I., Kalla, M., Zhang, L., and V. Paxson, "The Session Initiation Protocol (SIP)-Specific Event Notification", RFC 2960, October 2000. Mahy, et al. Expires February 22, 2007 [Page 19] Internet-Draft SIP Connection Reuse August 2006 Authors' Addresses Rohan Mahy Plantronics Email: rohan@ekabal.com Vijay K. Gurbani (editor) Lucent Technologies, Inc./Bell Laboratories Email: vkg at acm dot org Brett Tate BroadSoft Email: brett@broadsoft.com Mahy, et al. Expires February 22, 2007 [Page 20] Internet-Draft SIP Connection Reuse August 2006 Intellectual Property Statement 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. Disclaimer of Validity 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. Copyright Statement Copyright (C) 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. Acknowledgment Funding for the RFC Editor function is currently provided by the Internet Society. Mahy, et al. Expires February 22, 2007 [Page 21]