IPv6 Operations Working Group D. Thaler Internet-Draft Microsoft Expires: January 15, 2009 July 14, 2008 Teredo Extensions draft-thaler-v6ops-teredo-extensions-01.txt Status of this Memo By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. This Internet-Draft will expire on January 15, 2009. Abstract This document specifies a set of extensions to the Teredo protocol. These extensions provide additional capabilities to Teredo, including support for more types of Network Address Translations (NATs), and support for more efficient communication. Thaler Expires January 15, 2009 [Page 1] Internet-Draft Teredo Extensions July 2008 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 3. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.1. Symmetric NAT Support Extension . . . . . . . . . . . . . 8 3.2. UPnP-Enabled Symmetric NAT Extension . . . . . . . . . . . 10 3.3. Port-Preserving Symmetric NAT Extension . . . . . . . . . 11 3.4. Hairpinning Extension . . . . . . . . . . . . . . . . . . 12 3.5. Server Load Reduction Extension . . . . . . . . . . . . . 14 4. Message Syntax . . . . . . . . . . . . . . . . . . . . . . . . 14 4.1. Nonce Trailer . . . . . . . . . . . . . . . . . . . . . . 15 4.2. Alternate Address Trailer . . . . . . . . . . . . . . . . 15 4.3. Alternate Address Trailer . . . . . . . . . . . . . . . . 16 4.4. Random Port Trailer . . . . . . . . . . . . . . . . . . . 17 5. Protocol Details . . . . . . . . . . . . . . . . . . . . . . . 18 5.1. Symmetric NAT Support Extension . . . . . . . . . . . . . 18 5.1.1. Abstract Data Model . . . . . . . . . . . . . . . . . 18 5.1.2. Message Processing . . . . . . . . . . . . . . . . . . 18 5.2. UPnP-Enabled Symmetric NAT Extension . . . . . . . . . . . 20 5.2.1. Abstract Data Model . . . . . . . . . . . . . . . . . 20 5.2.2. Initialization . . . . . . . . . . . . . . . . . . . . 20 5.2.3. Message Processing . . . . . . . . . . . . . . . . . . 21 5.3. Port-Preserving Symmetric NAT Extension . . . . . . . . . 22 5.3.1. Abstract Data Model . . . . . . . . . . . . . . . . . 22 5.3.2. Timers . . . . . . . . . . . . . . . . . . . . . . . . 23 5.3.3. Message Processing . . . . . . . . . . . . . . . . . . 24 5.4. Hairpinning Extension . . . . . . . . . . . . . . . . . . 26 5.4.1. Abstract Data Model . . . . . . . . . . . . . . . . . 26 5.4.2. Initialization . . . . . . . . . . . . . . . . . . . . 27 5.4.3. Message Processing . . . . . . . . . . . . . . . . . . 27 5.5. Server Load Reduction Extension . . . . . . . . . . . . . 28 5.5.1. Abstract Data Model . . . . . . . . . . . . . . . . . 28 5.5.2. Timers . . . . . . . . . . . . . . . . . . . . . . . . 29 5.5.3. Message Processing . . . . . . . . . . . . . . . . . . 29 6. Protocol Examples . . . . . . . . . . . . . . . . . . . . . . 30 6.1. Symmetric NAT Support Extension . . . . . . . . . . . . . 30 6.2. UPnP-enabled Symmetric NAT Extension . . . . . . . . . . . 32 6.3. Port-Preserving Symmetric NAT Extension . . . . . . . . . 34 6.4. Hairpinning Extension . . . . . . . . . . . . . . . . . . 38 6.5. Server Load Reduction Extension . . . . . . . . . . . . . 40 7. Security Considerations . . . . . . . . . . . . . . . . . . . 41 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 42 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 42 9.1. Normative References . . . . . . . . . . . . . . . . . . . 42 9.2. Informative References . . . . . . . . . . . . . . . . . . 42 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 42 Intellectual Property and Copyright Statements . . . . . . . . . . 44 Thaler Expires January 15, 2009 [Page 2] Internet-Draft Teredo Extensions July 2008 1. Introduction This document specifies extensions to the Teredo protocol, as specified in [RFC4380]. These extensions provide additional capabilities to Teredo, including support for more types of Network Address Translations (NATs), and support for more efficient communication. 2. Terminology This document uses the following terminology, for consistency with [RFC4380]. Address-Restricted NAT: A Restricted NAT that accepts packets from an external host's IP address X and port Y if the internal host has sent a packet destined to IP address X regardless of the destination port. Address-Symmetric NAT: A symmetric NAT that has multiple external IP addresses assign different IP addresses and ports when communicating with different external hosts. Cone NAT: A NAT that maps all requests from the same internal IP address and port to the same external IP address and port. Furthermore, any external host can send a packet to the internal host by sending a packet to the mapped external address and port. Direct Bubble: A Teredo bubble that is sent directly to the IPv4 node whose Teredo address is contained in the Destination field of the IPv6 header, as specified in [RFC4380] section 2.8. The IPv4 Destination Address and UDP Destination Port fields contain a mapped address/port. Hairpinning: A feature available in some NATs where if a host is positioned behind a NAT and is assigned a given external (public) address and port by the NAT, hairpinning support in the NAT allows another host behind the same NAT to send a packet destined to the external address and port assigned to the first host, and the NAT automatically routes the packet back to the correct host. This term came to be because the packet arrives on, and is sent out to, the same NAT interface. Indirect Bubble: A Teredo bubble that is sent indirectly (via the destination's Teredo server) to another Teredo client, as specified in [RFC4380] section 5.2.4. Local Address/Port: The IPv4 address and UDP port from which a Teredo client sends Teredo packets. The local port is referred to as the Thaler Expires January 15, 2009 [Page 3] Internet-Draft Teredo Extensions July 2008 Teredo service port in [RFC4380]. The local address of a node may or may not be globally routable because the node can be located behind one or more NATs. Mapped Address/Port: A global IPv4 address and a UDP port that results from the translation of a node's own local address/port by one or more NATs. The node learns these values through the Teredo protocol specified in [RFC4380]. For symmetric NATs, the mapped address/port can be different for every peer that a node tries to communicate with. Network Address Translation (NAT): The process of converting between IP addresses used within an intranet or other private network and Internet IP addresses. Nonce: A time-variant counter used in the connection setup phase to prevent message replay and other types of attacks. Peer: A Teredo client with which another Teredo Client needs to communicate. Port-Preserving NAT: A NAT that translates a local address/port to a mapped address/port such that the mapped port has the same value as the local port, as long as that same mapped address/port has not already been used for a different local address/port. Port-Restricted NAT: A restricted NAT that accepts packets from an external host's IP address X and port Y only if the internal host has sent a packet destined to IP address X and port Y. Port-Symmetric NAT: A symmetric NAT that has only a single external IP address and hence only assigns different ports when communicating with different external hosts. Private Address: An IPv4 address that is not globally routable but is part of the private address space specified in [RFC1918] section 3. Public Address: An external global address used by a NAT. Restricted NAT: A NAT where all requests from the same internal IP address and port are mapped to the same external IP address and port. Unlike the cone NAT, an external host can send packets to an internal host (by sending a packet to the external mapped address and port) only if the internal host has first sent a packet to the external host. There are two kinds of restricted NATs: address-restricted NATs and port-restricted NATs. Symmetric NAT: A NAT where all requests from the same internal IP Thaler Expires January 15, 2009 [Page 4] Internet-Draft Teredo Extensions July 2008 address and port and to the same destination IP address and port, are mapped to the same external IP address and port. Requests from the same internal IP address and port to a different destination IP address and port may be mapped to a different external IP address and port. Furthermore, a symmetric NAT accepts packets received from an external host's IP address X and port Y only if some internal host has sent packets to IP address X and port Y. Teredo Bubble: A Teredo control message (specified in [RFC4380] section 2.8) that is used to create a mapping in a NAT. There are two types of Teredo bubbles: direct bubbles and indirect bubbles. Teredo Client: A node that has access to the IPv4 Internet and wants to gain access to the IPv6 Internet. Teredo IPv6 Address: An IPv6 address that starts with the prefix 2001:0000:/32 and is formed as specified in [RFC4380] section 2.14. Teredo Server: A node that has a globally routable address on the IPv4 Internet, and is used as a helper to provide IPv6 connectivity to Teredo clients. Teredo Server Address: The IPv4 address of the Teredo server selected by a specific Teredo client. UPnP-enabled NAT: A NAT that has the UPnP device control protocol enabled, as specified in [UPNPWANIP]. (Note that today, by default, most UPnP-capable NATs have the UPnP device control protocol disabled.) 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]. 3. Overview The Teredo protocol (as specified in [RFC4380]) enables nodes located behind one or more IPv4 NATs to obtain IPv6 connectivity by tunneling packets over UDP. When a node behind a NAT needs to communicate with a peer (i.e., another node) that is behind a NAT, there are four sets of IPv4 address/port pairs of interest: o The node's own IPv4 address/port. o The external IPv4 address/port to which the node's NAT translates. Thaler Expires January 15, 2009 [Page 5] Internet-Draft Teredo Extensions July 2008 o The peer's own IPv4 address/port. o The external IPv4 address/port to which the peer's NAT translates. When the node sends a packet to a peer, the node needs to send it from the node's own IPv4 address/port, destined to the peer's external IPv4 address/port. By the time it arrives at the peer (i.e., after passing through both NATs), the peer will see the same packet as coming from the node's external IPv4 address/port, destined to the peer's own IPv4 address/port. In this document, the term local address/port refers to a Teredo client's own IPv4 address/port; and mapped address/port refers to the external IPv4 address/port to which its NAT translates the local address/port. That is, the mapped address/port is what the IPv4 Internet sees the Teredo client as. A Teredo client running on a node communicates with a Teredo server to discover its mapped address/port. The mapped address/port, along with the Teredo server address, are used to generate an IPv6 address known as a Teredo IPv6 address. This allows any peer that gets the node's IPv6 address to easily determine the external IPv4 address/ port to which to send IPv6 packets encapsulated in IPv4 UDP messages. This document specifies extensions to the Teredo protocol. These Teredo extensions are independent of each other and can be implemented in isolation, except that the UPnP-Symmetric NAT Extension and the Port-Preserving Symmetric NAT Extension both require the Symmetric NAT Support Extension to be implemented. An implementation of this specification can support any combination of the Teredo extensions, subject to the above-mentioned restriction. The following matrix outlines the connectivity improvements of some of the extensions outlined in this document. Thaler Expires January 15, 2009 [Page 6] Internet-Draft Teredo Extensions July 2008 Destination NAT | | | | | | Port- | | | | | | UPnP | UPnP | pres. | | | | Addr. | Port | Port | Port | Port- | Port- | Addr. Source NAT | Cone | rest. | rest. | rest. | symm. | symm. | symm. | symm. -----------+-------+-------+-------+-------+-------+-------+-------+------- Cone | Yes | Yes | Yes | Yes | SNS | SNS | SNS | SNS -----------+-------+-------+-------+-------+-------+-------+-------+------- Address | Yes | Yes | Yes | Yes | SNS | SNS | SNS | No restricted | | | | | | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- Port | Yes | Yes | Yes | Yes | No | SNS + | No | No restricted | | | | | | PP | | -----------+-------+-------+-------+-------+-------+-------+-------+------- UPnP Port- | Yes | Yes | Yes | Yes | SNS + | No | No | No restricted | | | | | UPnP | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- UPnP Port | SNS | SNS | No | SNS + | SNS + | No | No | No symmetric | | | | UPnP | UPnP | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- Port- | | | SNS | | | SNS | | preserving | SNS | SNS | + | No | No | + | No | No Port- | | | PP | | | PP | | symmetric | | | | | | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- Port- | SNS | SNS | No | No | No | No | No | No symmetric | | | | | | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- Address- | SNS | No | No | No | No | No | No | No symmetric | | | | | | | | -----------+-------+-------+-------+-------+-------+-------+-------+------- Yes = Supported by . SNS = Supported with the Symmetric NAT Support Extension. SNS+UPNP = Supported with the Symmetric NAT Support Extension and UPnP Symmetric NAT Extension. SNS+PP = Supported with the Symmetric NAT Support Extension and Port- Preserving Symmetric NAT Extension. No = No connectivity. Matrix of Connectivity Improvements for Teredo Extensions Figure 1 Thaler Expires January 15, 2009 [Page 7] Internet-Draft Teredo Extensions July 2008 3.1. Symmetric NAT Support Extension The qualification procedure (as specified in [RFC4380] section 5.2.1) is a process that allows a Teredo client to determine the type of NAT that it is behind, in addition to its mapped address/port as seen by its Teredo server. However, [RFC4380] section 5.2.1 suggests that if the client learns it is behind a symmetric NAT, the Teredo client should go into an "offline state" where it is not able to use Teredo. The primary reason for doing so is that it is not easy for Teredo clients to connect to each other if either or both of them are positioned behind a symmetric NAT. Because of the way a symmetric NAT works, a peer sees a different mapped address/port in the IPv4/ UDP headers of packets coming from a Teredo client than the node's Teredo server sees (and hence appears in the node's Teredo IPv6 address). Consequently, a symmetric NAT does not allow incoming packets from a peer that are addressed to the mapped address/port embedded in the node's Teredo IPv6 address. Thus, the incoming packets are dropped and communication with Teredo client behind symmetric NATs is not established. With the Symmetric NAT Support Extension, Teredo clients begin to use Teredo even after they detect that they are positioned behind a symmetric NAT. Consider the topology shown in Figure Figure 2. Teredo Client B uses Teredo Server 2 to learn that its mapped address/port is 137.107.0.1: 8192, and constructs a Teredo IPv6 address, as specified in [RFC4380] section 4. Hence, CE49:7601 is the hexadecimal value of the address of Teredo Server 2 (206.73.118.1), the mapped port is exclusive-OR'ed with 0xFFFF to form DFFF, and the Mapped Address is exclusive-OR'ed with 0xFFFFFFFF to form 7C94:FFFE. Teredo Client A uses Teredo Server 1 to learn that its mapped address/port is 157.60.0.1:4096 and, with this extension, constructs a Teredo IPv6 address (as specified in [RFC4380] section 4) even though it learns that it is behind a symmetric NAT. Hence, 4136:E378 is the hexadecimal value of the address of Teredo Server 1 (65.54.227.120), the mapped port is exclusive-OR'ed with 0xFFFF to form EFFF, and the Mapped Address is exclusive-OR'ed with 0xFFFFFFFF to form 62C3:FFFE. The Symmetric NAT Support Extension enables a Teredo client positioned behind a symmetric NAT to communicate with Teredo peers positioned behind a cone or address-restricted NATs as follows, depending on what side initiates the communication. Thaler Expires January 15, 2009 [Page 8] Internet-Draft Teredo Extensions July 2008 -------------------------------------------- / \ < IPv6 Internet > \ / -|----------------------------------------|- | | +----------+ +----------+ | Teredo | | Teredo | | Server 1 | | Server 2 | +----------+ +----------+ 65.54.227.120| 206.73.118.1| -|----------------------------------------|- / \ < IPv4 Internet > \ / -|----------------------------------------|- 157.60.0.1| 137.107.0.1| UDP port 4096| UDP port 8192| +---------+ +----------+ |Symmetric| |Other type| | NAT | | of NAT | +---------+ +----------+ | | +-----------------+ +-----------------+ | Teredo client A | | Teredo client B | +-----------------+ +-----------------+ 3FFE:831F:4136:E378:8000:EFFF:62C3:FFFE 3FFE:831F:CE49:7601:0:DFFF:7C94:FFFE Teredo Address Teredo Address Symmetric NAT example Figure 2 In the first case, assume a Teredo Client B (B) positioned behind a cone or address-restricted NATs, initiates communication with Teredo Client A (A) positioned behind a symmetric NAT. B sends an indirect bubble via A's server (Teredo Server 1) to A, and A responds with a direct bubble. This direct bubble reaches B, as it is positioned behind a cone/address-restricted NAT. However, the mapped address/ port in the IPv4/UDP headers of the direct bubble are different from the mapped address/port embedded in A's Teredo IPv6 address. B therefore remembers the mapped address/port of the direct bubble and uses them for future communication with A, and thus communication is established. In the second case, assume A positioned behind a symmetric NAT initiates communication with B behind a cone or address-restricted NAT. A sends an indirect bubble to B via B's server (Teredo Server Thaler Expires January 15, 2009 [Page 9] Internet-Draft Teredo Extensions July 2008 2), and B responds with a direct bubble. This direct bubble is dropped by A's symmetric NAT because the direct bubble is addressed to the mapped address/port embedded in A's Teredo IPv6 address. However, communication can be established by having B respond with an indirect bubble via A's server (Teredo Server 1). Now the scenario is similar to the first case and communication will be established. 3.2. UPnP-Enabled Symmetric NAT Extension The UPnP-enabled Symmetric NAT Extension is dependent on the Symmetric NAT Support Extension. Only if Teredo clients have been enabled to acquire a Teredo IPv6 address in spite of being behind a symmetric NAT, will this extension help in traversing UPnP-enabled Symmetric NATs. The Symmetric NAT Support Extension enables communication between Teredo clients behind symmetric NATs with Teredo clients behind cone NATs or address-restricted NATs. However, clients behind symmetric NATs can still not communicate with clients behind port-restricted NATs or symmetric NATs. Referring again to Figure Figure 2 (see section Section 3.1), assume Teredo Client A is positioned behind a symmetric NAT and initiates communication with Client B, which is positioned behind a port- restricted NAT. Client A sends a direct bubble and an indirect bubble to Client B via Client B's server (Teredo Server 2). As per the characteristics of the symmetric NAT, the IPv4 source of the direct bubble contains a different mapped address and/or port than the one embedded in the Teredo server. This direct bubble is dropped because Client B's NAT does not have state to let it pass through, and Client B does not learn the mapped address/port used in the IPv4/ UDP headers. In response to the indirect bubble from Client A, Client B sends a direct bubble destined to the mapped address/port embedded in Client A's Teredo IPv6 address. This direct bubble is dropped because Client A's NAT does not have state to accept packets destined to that mapped address/port. The direct bubble does, however, cause Client B's NAT to set up outgoing state for the mapped address/port embedded in Client A's Teredo IPv6 address. As described in section Section 3.1, Client B also sends an indirect bubble that elicits a direct bubble from Client A. Unlike the case in section Section 3.1, however, the direct bubble from Client A is dropped as Client B's NAT does not have state for the mapped address/ port that Client A's NAT uses. Note Client B's NAT is port- restricted and hence requires both the mapped address and port to be the same as in its outgoing state, whereas in section Section 3.1, Client A's NAT was a cone or address-restricted NAT which only required the mapped address (but not port) to be the same. Thus, Thaler Expires January 15, 2009 [Page 10] Internet-Draft Teredo Extensions July 2008 communication between Client A and Client B fails. If Client B were behind a symmetric NAT, the problem is further complicated by Client B's NAT using a different outgoing mapped address/port than the one embedded in Client B's Teredo IPv6 address. If a Teredo client is separated from the global Internet by a single UPnP-enabled symmetric or port-restricted NAT, it can communicate with other Teredo clients that are positioned behind a single UPnP- enabled symmetric or port-restricted NAT as follows: Teredo clients, before communicating with the Teredo server during the qualification procedure, use UPnP to reserve a local address/port to mapped address/port translation. Therefore, during the qualification procedure, the Teredo server reflects back the reserved mapped address/port, which then is included in the Teredo IPv6 address. The mapping created by UPnP allows the NAT to forward packets destined for the mapped address/port to the local address/ port, independent of the source of the packets. Thus, a Teredo client, positioned behind a UPnP-enabled symmetric NAT, can receive a direct bubble sent by any Teredo peer. The Teredo client compares the peer's mapped address/port as seen in the IPv4/ UDP headers with the mapped address/port in the peer's Teredo IPv6 address. If the two mappings are different, the packet was sent by another Teredo client positioned behind a symmetric NAT. The Symmetric NAT Support Extension suggested that the Teredo client use the peer's mapped address/port seen in the IPv4/UDP headers for future communication. However, since symmetric NAT-to-symmetric NAT communication would not have been possible anyway, the Teredo client sends back a direct bubble to the mapped port/address embedded in the peer's Teredo IPv6 address. If the peer is also situated behind a UPnP-enabled NAT, the direct bubble will make it through and communication will be established. 3.3. Port-Preserving Symmetric NAT Extension The Port-Preserving Symmetric NAT Extension is dependent on the Symmetric NAT Support Extension (section Section 3.1). Only if Teredo clients have been enabled to acquire a Teredo IPv6 address in spite of being behind a symmetric NAT will this extension help in traversing port-preserving symmetric NATs. The Symmetric NAT Support Extension enables communication between Teredo clients behind symmetric NATs with Teredo clients behind cone NATs or address-restricted NATs. However, clients behind symmetric NATs can still not communicate with clients behind port-restricted or symmetric NATs, as described in section Section 3.2. Note that the Port-Preserving Symmetric NAT Extension described here is independent Thaler Expires January 15, 2009 [Page 11] Internet-Draft Teredo Extensions July 2008 of the UPnP-enabled Symmetric NAT Extension, described in section Section 3.2. If a Teredo client is positioned behind a port-preserving symmetric NAT, the client can communicate with other Teredo clients positioned behind a port-restricted NAT or a port-preserving symmetric NAT as follows. Teredo clients compare the mapped port learned during the qualification procedure with their local port to determine if they are positioned behind a port-preserving NAT. If both the mapped port and the local port have the same value, the Teredo client is positioned behind a port-preserving NAT. At the end of the qualification procedure, the Teredo client also knows if it is positioned behind a symmetric NAT, as described in section Section 3.1. Teredo clients positioned behind port-preserving symmetric NATs can also listen on randomly chosen local ports. If the randomly chosen local port has not been used by the symmetric NAT as a mapped port in a prior port-mapping, the NAT uses the same port number as the mapped port. Thus, the challenge is to get the first direct bubble sent out from the random port to be destined to a valid destination address and port. When the mapped address/port is embedded in the destination's Teredo IPv6 address, this is easy. The communication setup is more complicated when the destination Teredo client is also positioned behind a port-preserving symmetric NAT. In such a case, both Teredo clients need to send their first direct bubbles to the correct destination mapped address/port. Thus the protocol messages, which communicate one Teredo client's random port number to the other Teredo client, must be exchanged indirectly (via Teredo servers). When one Teredo client has access to the other Teredo client's random port number, it can send a direct bubble destined to the mapped address embedded in the destination's Teredo IPv6 address, and the mapped port can be the same as the destination's random port number. If both NATs are port-preserving, port-preserved mappings are created on both NATs and the second direct bubble succeeds in reaching the destination. 3.4. Hairpinning Extension Hairpinning support in a NAT routes packets that are sent from a private (local) address destined to a public (mapped) address of the NAT, back to the another private (local) destination address behind the same NAT. If hairpinning support is not available in a NAT, two Teredo clients behind the same NAT are not able to communicate with each other, as specified in [RFC4380] section 8.3. Thaler Expires January 15, 2009 [Page 12] Internet-Draft Teredo Extensions July 2008 The Hairpinning Extension enables two clients behind the same NAT to talk to each other when the NAT does not support hairpinning. This process is illustrated in the following diagram. -------------------------------------------- / \ < IPv6 Internet > \ / -|----------------------------------------|- | | +----------+ +----------+ | Teredo | | Teredo | | Server 1 | | Server 2 | +----------+ +----------+ 206.73.118.1| 65.54.227.120| -|----------------------------------------|- / \ < IPv4 Internet > \ / --------------------|----------------------- | NAT +-------+ without | NAT | hairpinning | E | support +-------+ | +------------------+---------------------+ 192.168.1.0| 192.168.1.1| UDP port 4095| UDP port 4096| +---------+ +----------+ | NAT | | NAT | | F | | G | +---------+ +----------+ | | +-----------------+ +-----------------+ | Teredo client A | | Teredo client B | +-----------------+ +-----------------+ 3FFE:831F:CE49:7601:0:DFFF:7C94:FFFE 3FFE:831F:4136:E378:8000:EFFF:62C3:FFFE Teredo Address Teredo Address Hairpinning example Figure 3 The Teredo Client A (A) includes, as part of its indirect bubble sent to Teredo Client B (B), its local address/port. B, upon receiving the indirect bubble, tries to establish communication by sending direct bubbles to the mapped address/port of A, and also to the local Thaler Expires January 15, 2009 [Page 13] Internet-Draft Teredo Extensions July 2008 address/port of B. If a Teredo client is part of a multi-NAT hierarchy and the NAT to which the Teredo client is connected supports the UPnP protocol (as specified in [UPNPWANIP]), the Teredo client can use UPnP to determine the mapped address/port assigned to it by the NAT. This information can be included along with the local address/port when sending the indirect bubble. The destination Teredo client now tries to establish a connection by sending direct bubbles to the mapped address/port in the Teredo IPv6 address, to the local address/port included in the bubble, and also to the mapped address/port included in the bubble. 3.5. Server Load Reduction Extension If communication between a Teredo client and a Teredo peer was successfully established but at a later stage was silent for a while, for efficiency it is best to refresh the mapping state in the NATs that are positioned between them. To refresh the communication between itself and a Teredo peer, a Teredo client needs to solicit a direct bubble response from the Teredo peer. An indirect bubble is sent to solicit a direct bubble response from a Teredo peer, as specified in [RFC4380] section 5.2.4. However, these indirect bubbles increase the load on the Teredo server. The Server Load Reduction Extension allows Teredo clients to send direct bubbles most of the time instead of sending indirect bubbles all of the time in the following way: 1. When a Teredo client tries to refresh its communication with a Teredo peer, it uses a direct bubble instead of an indirect bubble. However, because direct bubbles do not normally solicit a response, the direct bubble format is extended to be able to solicit a response. 2. When a Teredo client receives a direct bubble that is soliciting a response, the Teredo client responds with a direct bubble. 3. If attempts to re-establish communication with the help of direct bubbles fail, the Teredo client starts over the process of establishing communication with the Teredo peer, as specified in [RFC4380] section 5.2.4. 4. Message Syntax All Teredo messages are transported over the User Datagram Protocol (UDP), as specified in [RFC4380] section 3. Thaler Expires January 15, 2009 [Page 14] Internet-Draft Teredo Extensions July 2008 4.1. Nonce Trailer The Nonce Trailer is used by the Symmetric NAT Support Extension (and therefore the UPnP-enabled Symmetric NAT Extension and Port- Preserving Symmetric NAT Extension also) and the Hairpinning Extension. The Nonce Trailer can be present in both indirect and direct bubbles sent by the Teredo client. The nonce in the Nonce Trailer helps authenticate a Teredo client positioned behind a Symmetric NAT. 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ... | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Type (1 byte): The Trailer Option type. This field MUST be set to 0x01. Length (1 byte): The length in bytes of the rest of the option. This field MUST be set to 0x04. Nonce (4 bytes): The Nonce value. 4.2. Alternate Address Trailer The Alternate Address Trailer is used by the Hairpinning Extension. The Alternate Address Trailer MUST NOT be present in any packets other than direct bubbles sent by a Teredo client. The Alternate Address Trailer provides another Teredo client positioned behind the same NAT with more address options that it can use to connect. 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | Alternate Address/Port List (variable) | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Type (1 byte): The Trailer Option type. This field MUST be set to 0x03. Length (1 byte): The length in bytes of the rest of the option. The Thaler Expires January 15, 2009 [Page 15] Internet-Draft Teredo Extensions July 2008 value of this field MUST be in the range 8 to 26 (i.e., 2 bytes for the Reserved field, and 6 bytes for each entry in the Alternate Address/Port List). This allows for a minimum of one address/port mapping and a maximum of four address/port mappings to be advertised. It SHOULD be at most 14 as a maximum of two address/port mappings can be determined by Teredo: one local address/port and one obtained using UPnP. Since the length of the alternate address/port is 6 bytes, the valid range of values is only 8, 14, 20 and 26. Reserved (2 bytes): This field MUST be set to 0x0000 and ignored on receipt. Alternate Address/Port List (variable): An array of additional address/port pairs that can be used by other Teredo clients to communicate with the sender. Each alternate address/port entry MUST be formatted as follows: 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | IPv4 Address | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Port | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ IPv4 Address (4 bytes): An IPv4 address in network byte order. This field MUST contain a valid unicast address. Port (2 bytes): A port number in network byte order. This field MUST NOT be zero. 4.3. Alternate Address Trailer The Neighbor Discovery Option Trailer is used by the Server Load Reduction Extension because it allows direct bubbles to encode an IPv6 Neighbor Solicitation ([RFC4861] section 4.3), in addition to an IPv6 Neighbor Advertisement ([RFC4861] section 4.4), which prevents packets from being relayed indirectly through a Teredo server. The Neighbor Discovery Option Trailer allows the receiver to differentiate between a direct bubble which is soliciting a response versus a regular direct bubble. This allows Teredo clients to use direct bubbles to refresh inactive connections instead of using indirect bubbles. Thaler Expires January 15, 2009 [Page 16] Internet-Draft Teredo Extensions July 2008 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | DiscoveryType | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ... | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Type (1 byte): The Trailer Option type. This field MUST be set to 0x04. Length (1 byte): The length in bytes of the rest of the option. This field MUST be set to 0x04. DiscoveryType (1 byte): This field MUST be set to one of the following values: TeredoDiscoverySolicitation (0x00): The receiver is requested to respond with a direct bubble of DiscoveryType TeredoDiscoveryAdvertisement. TeredoDiscoveryAdvertisement (0x01): The direct bubble is in response to a direct bubble or an indirect bubbles containing DiscoveryType TeredoDiscoverySolicitation. Reserved (3 bytes): This field MUST be set to 0x000000 on transmission and ignored on receipt. 4.4. Random Port Trailer The Random Port Trailer is used by the Port-Preserving Symmetric NAT Extension in both indirect and direct bubbles. 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type | Length | Random Port | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Type (1 byte): The Trailer Option type. MUST be set to 0x05. Length (1 byte): The length in bytes of the rest of the option. This field MUST be set to 0x02. Random Port (2 bytes): The random port number on which the sender is listening. This field MUST be specified in network byte order. Thaler Expires January 15, 2009 [Page 17] Internet-Draft Teredo Extensions July 2008 5. Protocol Details 5.1. Symmetric NAT Support Extension [RFC4380] section 5.2.1 advises that no Teredo IPv6 address be configured if the Teredo client is positioned behind a symmetric NAT. For Teredo clients positioned behind symmetric NATs, the mapped address/port used by its NAT when communicating with a Teredo peer is different from the mapped address/port embedded in the Teredo client's Teredo IPv6 address. The Symmetric NAT Support Extension provides a solution to this problem. In addition, [RFC4380] section 5.2.9 specifies a direct IPv6 connectivity test to determine that the mapped address/port in the Teredo IPv6 address of a peer is not spoofed. It does this through the use of a nonce in ICMPv6 Echo Request and Response messages (which are defined in [RFC2463] section 4). However, the direct IPv6 connectivity test is limited only to communication between Teredo IPv6 addresses and non-Teredo IPv6 addresses. In the following extension, we introduce the use of a nonce in direct and indirect bubbles and provide a mechanism to verify that the mapped address/ port are not spoofed. This extension is optional; an implementation SHOULD support it. 5.1.1. Abstract Data Model This section describes a conceptual model of possible data organization that an implementation maintains to participate in this protocol. The described organization is provided to facilitate the explanation of how the protocol behaves. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document. In addition to the state specified in [RFC4380] section 5.2, the following are also required: Peer Entry: The following additional state is required on a per-peer basis: o Nonce Sent: The value of the nonce sent in the last indirect bubble sent to the Teredo peer. o Nonce Received: The value of the nonce received in the last indirect bubble received from the Teredo peer. 5.1.2. Message Processing Except as specified in the following sections, the rules for message processing are as specified in [RFC4380]. Thaler Expires January 15, 2009 [Page 18] Internet-Draft Teredo Extensions July 2008 5.1.2.1. Sending an Indirect Bubble The rules for when indirect bubbles are sent to a Teredo peer are specified in [RFC4380] section 5.2.6. When a Teredo client sends an indirect bubble, it MUST generate a random 4-byte value, and include it in the Nonce field of a Nonce Trailer (section 2.2.1) appended to the indirect bubble, and also store it in the Nonce Sent field of its Peer Entry for that Teredo peer. 5.1.2.2. Sending a Direct Bubble The rules for when direct bubbles are sent to a Teredo peer are specified in [RFC4380] section 5.2.6. When a Teredo client sends a direct bubble to a peer after receiving an indirect bubble with a Nonce Trailer, it MUST include in the direct bubble a Nonce Trailer with the same nonce value. If the Teredo client is about to send a direct bubble before it has received an indirect bubble from the Teredo peer, the Teredo client MUST NOT include a Nonce Trailer. 5.1.2.3. Receiving an Indirect Bubble The rules for processing an indirect bubble are specified in [RFC4380] section 5.2.3. In addition, when a Teredo client receives an indirect bubble containing a Nonce Trailer, the Teredo client MUST store the nonce in the Nonce Received field of its Peer Entry for that Teredo peer. If an indirect bubble is received without a Nonce Trailer, and the Nonce Received field in the Peer Entry is non-zero, the Nonce Received field SHOULD be set to zero. 5.1.2.4. Receiving a Direct Bubble If the mapped address/port of the direct bubble matches the mapped address/port embedded in the source Teredo IPv6 address, the direct bubble MUST be accepted, as specified in [RFC4380] section 5.2.3. In addition, if the mapped address/port does not match the embedded address/port but the direct bubble contains a Nonce Trailer with a nonce that matches the Nonce Sent field of the Teredo peer, the direct bubble MUST be accepted. If neither of the above conditions are true, the direct bubble MUST be dropped. If the direct bubble is accepted, the Teredo client MUST record the mapped address/port from which the direct bubble is received in the mapped address/port fields of the Teredo peer, as specified in Thaler Expires January 15, 2009 [Page 19] Internet-Draft Teredo Extensions July 2008 [RFC4380] section 5.2. 5.2. UPnP-Enabled Symmetric NAT Extension The UPnP-enabled Symmetric NAT Extension is optional; an implementation SHOULD support it. This extension has the Symmetric NAT Support Extension (as specified in section Section 5.1) as a dependency. Any node that implements this extension MUST also implement the Symmetric NAT Support Extension. 5.2.1. Abstract Data Model This section describes a conceptual model of possible data organization that an implementation maintains to participate in this protocol. The described organization is provided to facilitate the explanation of how the protocol behaves. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document. This extension extends the abstract data model in section Section 5.1.1 by adding the following additional fields. UPnP-Enabled NAT flag: This is a Boolean value, set to TRUE if the NAT positioned in front of the Teredo client is UPnP enabled. UPnP-Mapped Address/Port: The mapped address/port assigned via UPnP to the Teredo client by the UPnP-enabled NAT behind which the Teredo client is positioned. Note that this field has a valid value only if the NAT to which the Teredo client is connected is UPnP enabled. Also note that if the Teredo client is positioned behind a single NAT only (as opposed to a series of nested NATs), this value is the same as the mapped address/port embedded in its Teredo IPv6 address. Symmetric NAT flag: This is a Boolean value, set to TRUE if the Teredo client is positioned behind a symmetric NAT. Peer Entry: The following state needs to be added on a per-peer basis: Symmetric Peer flag: This is a Boolean value and is TRUE if the Teredo peer is positioned behind a symmetric NAT. 5.2.2. Initialization Prior to beginning the qualification procedure, the Teredo client MUST invoke the AddPortMapping function, as specified in [UPNPWANIP] section 2.4.16, with the following parameters: Thaler Expires January 15, 2009 [Page 20] Internet-Draft Teredo Extensions July 2008 o NewRemoteHost: "" (empty string) o NewExternalPort: Local Port value o NewProtocol: UDP o NewInternalPort: Local Port value o NewInternalClient: Local Address value o NewEnabled: TRUE o NewPortMappingDescription: "TEREDO" o NewLeaseDuration: 0 The successful completion of the AddPortMapping function indicates that the NAT has created a port mapping from the external port of the NAT to the internal port of the Teredo client node. The parameters are specified so that any external host should be able to send packets to the Teredo client by sending packets to the mapped address/port. The Teredo client MUST set its UPnP-Enabled NAT flag based on whether the AddPortMapping function succeeded or failed. During the qualification procedure (as specified in [RFC4380] section 5.2.1) when the Teredo client receives a response from the secondary Teredo server, the Teredo client MUST compare the mapped address/port learned from the secondary Teredo server with the mapped address/port associated with the Teredo server. If either the mapped address or mapped port value is different, the Symmetric NAT flag MUST be set to TRUE. After the qualification procedure, the mapped address/port learned from the Teredo server MUST be compared to the UPnP-Mapped Address/ Port. If both are the same, the Teredo client is positioned behind a single NAT and the UPnP-Mapped Address/Port MUST be zeroed out. 5.2.3. Message Processing Except as specified in the following sections, the rules for message processing are as specified in [RFC4380] section 5.2.3. 5.2.3.1. Receiving a Direct Bubble Except as indicated below, the rules for handling a direct bubble are as specified in section Section 5.1.2.4. A Teredo client positioned behind a UPnP-enabled NAT (port-restricted NAT as well as symmetric NAT) will receive all packets sent to the mapped address/port embedded in its Teredo IPv6 address. Thus when a Teredo client receives a direct bubble, it MUST compare the mapped address/port from which the packet was received with the mapped address/port embedded in the Teredo IPv6 address in the source address field of the IPv6 header. If the two are not the same, then it indicates that the Teredo peer is positioned behind a symmetric Thaler Expires January 15, 2009 [Page 21] Internet-Draft Teredo Extensions July 2008 NAT and it MUST set the Symmetric Peer flag in its Peer Entry. 5.2.3.2. Sending a Direct Bubble The rules for sending a direct bubble are specified in [RFC4380] section 5.2.6. In addition, section Section 5.1.2.2 specifies that direct bubbles be sent to the mapped address/port stored in the Peer Entry. This criteria is further refined as follows. If the Teredo client sending the direct bubble meets all of the following criteria: o The Symmetric NAT flag is set to TRUE. o The UPnP-Enabled NAT flag is set to TRUE. o The UPnP-Mapped Address/Port are set to zero. o The peer'sSymmetric Peer flag is set to TRUE. then the Teredo client MUST send the direct bubble to the mapped address/port embedded in the peer's Teredo IPv6 address. This is because Symmetric-to-Symmetric and Port-Restricted-to- Symmetric NAT communication between the Teredo client and the peer would have failed anyway. However, by taking a chance that the peer may also be positioned behind a UPnP-enabled NAT just like the Teredo client itself, the Teredo client can try sending the direct bubble to the mapped address/port in the peer's Teredo IPv6 address and hope for the packet to go through. If the packet does go through, communication is established. 5.3. Port-Preserving Symmetric NAT Extension The Port-Preserving Symmetric NAT Extension is optional; an implementation SHOULD support it. This extension has the Symmetric NAT Support Extension (as specified in section Section 5.1) as a dependency. Any node that implements this extension MUST also implement the Symmetric NAT Support Extension. 5.3.1. Abstract Data Model This section describes a conceptual model of possible data organization that an implementation maintains to participate in this protocol. The described organization is provided to facilitate the explanation of how the protocol behaves. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document. The Port-Preserving Symmetric NAT Extension extends the abstract data model in section Section 5.1.1 by adding the following additional fields. Thaler Expires January 15, 2009 [Page 22] Internet-Draft Teredo Extensions July 2008 Port-Preserving NAT flag: This is a Boolean value, set to TRUE if the Teredo client is positioned behind a port-preserving NAT. Symmetric NAT flag: This is a Boolean value, set to TRUE if the Teredo client is positioned behind a symmetric NAT. Peer Entry: The following fields need to be added on a per-peer basis: o Random Port: This field contains the value of the random port that the Teredo client is using for communication with the peer. Set to zero by default. o Peer Random Port: This field contains the value of the random port that the peer is using for communication with this Teredo client. Set to zero by default. o Direct Receive on Primary Port: This is a Boolean value, set to TRUE if a packet is received from the Teredo peer on the primary local port. Set to FALSE by default. o Direct Receive on Random Port: This is a Boolean value, set to TRUE if a packet is received from the Teredo peer on the Random Port. Set to FALSE by default. o Connection Refresh Count: This field contains the number of direct bubbles that have been sent to the peer since the last time data was communicated between the two peers. o Last Data Packet Sent Timestamp: This field contains the time stamp of the last data packet sent to the peer. This time stamp is different from the field that stores the data and time of last transmission to the peer (as specified in [RFC4380] section 5.2) because the RFC-defined field is also updated every time a direct bubble is sent. 5.3.2. Timers The Port-Preserving Symmetric NAT Extension requires the following timer: Refresh timer: A timer to refresh peer connections on which no data has been sent for a while. When the refresh timer expires, the Teredo client MUST go through its list of peers and for each peer to which the Teredo client is communicating through the random port, the Teredo client MUST check the Last Data Packet Sent Timestamp to determine if data has been sent to the peer in the last 30 seconds, and check the Connection Refresh Count field to determine if the count has reached the maximum allowed value of 20. If both checks are false, the Teredo client MUST send a direct bubble (as specified in section Section 5.3.3.3) to the peer and increment the Connection Refresh Count. This direct bubble is sent as an attempt to keep the port mappings on all the Thaler Expires January 15, 2009 [Page 23] Internet-Draft Teredo Extensions July 2008 intermediate NATs alive while the application/user may be temporarily inactive. If on the other hand, data has been sent to the peer in the last 30 seconds, the Connection Refresh Count MUST be reset to zero. The refresh timer MUST then be rescheduled to expire in 30 seconds. 5.3.3. Message Processing 5.3.3.1. Sending a Data Packet On receiving a data packet to be transmitted to the Teredo Peer (in addition to the rules specified in [RFC4380] section 5.2.4), the Teredo client MUST update the Last Data Packet Sent Timestamp when the packet is actually sent. 5.3.3.2. Sending an Indirect Bubble The rules for sending an indirect bubble are as specified in section Section 5.1.2.1 and [RFC4380] section 5.2.6. In addition to those rules, the Teredo client MUST do the following: o If the Symmetric NAT flag is set, and the Teredo peer is not marked as "trusted" (as specified in [RFC4380] section 5.2) and the Random Port is zero, the Teredo client MUST first select a random port number to use, store it in the Random Port field of the Peer Entry, and begin listening on that port. o If the Random Port value is non-zero, the Teredo client MUST append a Random Port Trailer to the indirect bubble. 5.3.3.3. Sending a Direct Bubble The rules for when direct bubbles are sent to a Teredo peer are as specified in [RFC4380] section 5.2.6. In addition, section Section 5.1.2.2 defines rules for enabling communication for clients positioned behind a symmetric NAT. In addition to the rules defined in both the above mentioned sections, the following rules apply also. If the Symmetric NAT flag is set, and the Teredo peer is not marked as "trusted" (as specified in [RFC4380] section 5.2) the Teredo client MUST send a direct bubble destined to the mapped address/port embedded in the Teredo IPv6 address of the Teredo peer. (This direct bubble will contain the Nonce Trailer (section 2.2.1).) If the peer Random Port field is non-zero, the Teredo client MUST send another direct bubble from its own random port, destined to the peer random port. The IPv4 destination address MUST be the mapped address embedded in the Teredo IPv6 address. In addition, the Teredo client MUST include the Random Port Trailer (section 2.2.5). Thaler Expires January 15, 2009 [Page 24] Internet-Draft Teredo Extensions July 2008 5.3.3.4. Receiving an Indirect Bubble The rules for processing an indirect bubble are as specified in section Section 5.1.2.3 and [RFC4380] section 5.2.3. In addition to these rules, if the incoming indirect bubble has a Random Port Trailer, the following additional processing MUST be done. If the Peer Random Port field of the Peer Entry is zero, the Teredo client MUST store the port from the Random Port Trailer in the Peer Random Port field of the Peer Entry. If the Peer Random Port field is non-zero and if either the Peer Random Port field and the new advertised port have the same value, or if active data has been exchanged between the two Teredo clients in the last 30 seconds (that is, "time of last transmission" or "time of last reception," as specified in [RFC4380] section 5.2, is set to a time that is less than 30 seconds ago), the new advertised port value MUST be ignored. If the Peer Random Port field is non-zero and the new advertised port value is different from the Peer Random Port value, and it has been more than 30 seconds since the last exchange of data packets between the two Teredo clients, (that is, "time of last transmission" and "time of last reception" are set to a time that is more than 30 seconds ago), the Teredo client SHOULD store the new advertised port value in the Peer Random Port field, clear the Random Port field, and stop listening on the old random port. This allows communication to be re-established if either side changes the random port that it is using. 5.3.3.5. Receiving a Direct Bubble The rules for handling direct bubbles are specified in section Section 5.1.2.4 and [RFC4380] section 5.2.3. The rules for whether to accept a direct bubble are extended as follows: o If the direct bubble is received on the primary port and the Teredo peer is not "trusted," the status field of the Teredo client MUST be changed to "trusted" and the Direct Receive on Primary flag MUST be set to TRUE. The mapped address/port from which the direct bubble was received MUST be recorded in the mapped address/port fields of the Teredo peer, as specified in [RFC4380] section 5.2. The Teredo client MUST then set the Random Port field in the Peer Entry to zero and stop listening on the old random port. o If the direct bubble is received on the primary port, the Teredo peer is "trusted," and the Direct Receive on Primary flag is set to TRUE, the Teredo client MUST compare the mapped address/port of the direct bubble with the mapped address/port of the Peer Entry. Thaler Expires January 15, 2009 [Page 25] Internet-Draft Teredo Extensions July 2008 If both mappings are the same, the direct bubble MUST be accepted. If the mappings are different and it has been more than 30 seconds since the last packet exchange with the Teredo peer (that is, "time of last transmission" and "time of last reception," as defined in [RFC4380] section 5.2, are set to a time that is more than 30 seconds ago), the mapping on the Teredo peer's NAT has changed and communication needs to be re-established. This MUST be done by changing the status of the peer to "not-trusted", setting the Direct Receive on Primary flag to FALSE, and sending an indirect bubble to the Teredo peer via its Teredo server. o If the direct bubble is received on the primary port, the Teredo peer is "trusted," the Direct Receive on Primary flag is set to FALSE, and the Direct Receive on Random Port flag is set to TRUE, the mapped address/port from which the direct bubble is received MUST be stored in the mapped address/port fields of the Peer Entry. The Direct Receive on Primary flag MUST be set to TRUE. The Teredo client MUST then set the Random Port field in the Peer Entry to zero and stop listening on the old random port. Finally, the Direct Receive on Random Port flag MUST be set to FALSE. o If the direct bubble is received on the random port and the Teredo peer is not "trusted," the status field of the Teredo client MUST be changed to "trusted" and the Direct Receive on Random Port flag MUST be set to TRUE. The mapped address/port from which the direct bubble was received MUST be recorded in the mapped address/ port fields of the Teredo Peer Entry, as specified in [RFC4380] section 5.2. o If the direct bubble is received on the random port, the Teredo peer is "trusted," the Direct Receive on Primary Port flag is FALSE, and the Direct Receive on Random Port flag is set to TRUE, the Teredo client MUST compare the mapped address/port in the direct bubble with the mapped address/port in the Peer Entry. If the two mappings are the same, the direct bubble MUST be accepted. If the mappings are different, it implies that the NAT had deleted the mapping and when it reassigned the mapping, a different external port was chosen. In this instance the Teredo client SHOULD set the Random Port field to zero, stop listening on the old random port, and send an indirect bubble to the Teredo peer as specified in section Section 5.3.3.2. 5.4. Hairpinning Extension This extension is optional; an implementation SHOULD support it. 5.4.1. Abstract Data Model This section describes a conceptual model of possible data organization that an implementation maintains to participate in this protocol. The described organization is provided to facilitate the Thaler Expires January 15, 2009 [Page 26] Internet-Draft Teredo Extensions July 2008 explanation of how the protocol behaves. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document. In addition to the state specified in [RFC4380] section 5.2, the following are also required: UPnP Mapped Address/Port: The mapped address/port assigned via UPnP to the Teredo client by the UPnP-enabled NAT behind which the Teredo client is positioned. This field has a valid value only if the NAT to which the Teredo client is connected is UPnP-enabled. In addition, if the Teredo client is positioned behind a single NAT only (as opposed to a series of nested NATs), this value will be the same as the mapped address/port embedded in its Teredo IPv6 address. Peer Entry: Per-peer state is extended beyond what is described in [RFC4380] by including the following: o Alternate Address/Port list: The list of alternate address/port pairs advertised by the peer. 5.4.2. Initialization Behavior is as specified in [RFC4380], with the following additions. Prior to beginning the qualification procedure, the Teredo client MUST invoke the AddPortMapping function (as specified in [UPNPWANIP] section 2.4.16) with the parameters specified in section Section 5.2.2. If successful, it indicates that the NAT has created a port mapping from the external port of the NAT to the internal port of the Teredo client node. If the AddPortMapping function is successful, the Teredo client MUST store the mapping assigned by the NAT in its UPnP Mapped Address/Port state. After the qualification procedure, the mapped address/port learned from the Teredo server MUST be compared to the UPnP Mapped Address/ Port. If both are the same, the Teredo client is positioned behind a single NAT and the UPnP Mapped Address/Port MUST be zeroed out. 5.4.3. Message Processing 5.4.3.1. Sending an Indirect Bubble The rules for when indirect bubbles are sent to a Teredo peer are as specified in [RFC4380] section 5.2.6. If communication between a Teredo client and a Teredo peer has not been established, the Teredo client MUST include the Alternate Address Trailer in the indirect bubble. The Alternate Address Trailer MUST include the node's local address/port in the Alternate Address/Port list. If the UPnP Mapped Thaler Expires January 15, 2009 [Page 27] Internet-Draft Teredo Extensions July 2008 Address/Port is non-zero, the Alternate Address Trailer MUST also include it in the list. Hairpinning requires "direct IPv6 connectivity tests" (as specified in [RFC4380] section 5.2.9) to succeed before it can accept packets from an IPv4 address and port not embedded in the Teredo IPv6 address. Hence the indirect bubble MUST also include a Nonce Trailer. 5.4.3.2. Receiving an Indirect Bubble The rules for processing indirect bubbles are as specified in [RFC4380] section 5.2.3. In addition to those rules, when a Teredo client receives an indirect bubble with the Alternate Address Trailer, it SHOULD first verify that the Alternate Address Trailer is correctly formed (as specified in section Section 4.2), and drop the bubble if not. Otherwise, it MUST set the Alternate Address/Port list in its Peer Entry to the list in the trailer. The Teredo client, besides sending direct bubbles to the mapped address/port embedded in the Teredo IPv6 address (as specified in [RFC4380] section 5.2.6), MUST also send a direct bubble to each mapped address/port advertised in the Alternate Address Trailer. In each of the direct bubbles, the Teredo client MUST include a Nonce Trailer with the nonce value received in the indirect bubble. 5.4.3.3. Receiving a Direct Bubble If the mapped address/port of the direct bubble matches the mapped address/port embedded in the source Teredo IPv6 address, the direct bubble MUST be accepted, as specified in [RFC4380] section 5.2.3. If the mapped address/port does not match the embedded address/port, but the direct bubble contains a Nonce Trailer with a nonce that matches the Nonce Sent field of the Teredo peer, the direct bubble MUST be accepted. If neither of the above rules match, the direct bubble MUST be dropped. 5.5. Server Load Reduction Extension This extension is optional; an implementation SHOULD support it. 5.5.1. Abstract Data Model This section describes a conceptual model of possible data organization that an implementation maintains to participate in this Thaler Expires January 15, 2009 [Page 28] Internet-Draft Teredo Extensions July 2008 protocol. The described organization is provided to facilitate the explanation of how the protocol behaves. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document. In addition to the state specified in [RFC4380] section 5.2, the following are also required: Peer Entry: The following state needs to be added on a per-peer basis: o Count of Solicitations Transmitted: The number of Solicitation packets sent. 5.5.2. Timers Retransmission Timer: A timer used to retransmit Teredo Neighbor Solicitation packets. When the retransmission timer expires, the Teredo client MUST retransmit a direct bubble with a Neighbor Discovery Option Trailer, and increment the Count of Solicitations Transmitted. If the count is less than three, it MUST then reset the timer to expire in two seconds. Otherwise (if the count is now three), it MUST send an indirect bubble to the Teredo peer to re-establish connectivity as if no communication between the Teredo client and the Teredo peer had been established. 5.5.3. Message Processing Except as specified below, processing is the same as specified in [RFC4380]. 5.5.3.1. Sending a Data Packet Upon receiving a data packet to be transmitted to the Teredo peer, the Teredo client MUST determine whether data has been exchanged between the Teredo client and peer in either direction in the last 30 seconds (using the state as specified in [RFC4380] section 5.2). If not, the Teredo client MUST send a direct bubble with a Neighbor Discovery Option Trailer having the DiscoveryType field set to TeredoDiscoverySolicitation. The Count of Solicitations Transmitted field MUST be set to 1. The retransmission timer MUST be set to expire in two seconds. 5.5.3.2. Receiving a Direct Bubble The rules for processing direct bubbles are as specified in [RFC4380] section 5.2.3. In addition to those rules, upon receiving a direct Thaler Expires January 15, 2009 [Page 29] Internet-Draft Teredo Extensions July 2008 bubble containing a Neighbor Discovery Option Trailer with DiscoveryType field set to TeredoDiscoverySolicitation, the Teredo client MUST respond with a direct bubble with the Neighbor Discovery Option Trailer having the DiscoveryType field set to TeredoDiscoveryAdvertisement. 6. Protocol Examples The following sections describe several operations as used in common scenarios to illustrate the function of Teredo Extensions. 6.1. Symmetric NAT Support Extension The following protocol example illustrates the use of the Symmetric NAT Support Extension. In Figure Figure 2 (section Section 3.1), assume Teredo Client A, which is positioned behind a port-symmetric NAT, wants to communicate with Teredo Client B, which is positioned behind an address- restricted NAT. The qualification procedure where the Teredo client determines that it is positioned behind a symmetric NAT is exactly the same as that specified in [RFC4380] section 5.2.1. Because of the Symmetric NAT Extension, Client A continues to configure a Teredo IPv6 address even after determining that the Teredo client is positioned behind a symmetric NAT. Next the following packet exchange helps Teredo Client A (A) establish communication with Teredo Client B (B). Thaler Expires January 15, 2009 [Page 30] Internet-Draft Teredo Extensions July 2008 Teredo Client A's Client B's Teredo Client Teredo Teredo Client A NAT Server Server NAT B | | | | | | | | | Direct Bubble to B | | | 1 |--------------------------------------------------->| | | | | | | | |Indirect Bubble to B via B's Teredo Server| | | 2 |----------------------------------------->|----------------->| | | | | | | | | | Direct Bubble to A | | | | |<--------------------------------------------------| 3 | | | | | | | | |Indirect Bubble to A via A's Teredo Server| |<-----------------|<-----------------------------------------| 4 | | | | | | | | | Direct Bubble to B | | | 5 |------------------------------------------------------------>| | | | | | | |Indirect Bubble to B via B's Teredo Server| | | 6 |----------------------------------------->|----------------->| | | | | | | | | | Direct Bubble to A | | | |<------------------------------------------------------------| 7 | | | | | | Port-Symmetric NAT to Address-Restricted NAT Packet Exchange 1. A sends a direct bubble (Packet 1) destined to the mapped address/port embedded in B's Teredo IPv6 address. The mapped port in the source field of the packet assigned by client A's NAT is different from the mapped port embedded in A's Teredo IPv6 address. This is characteristic of the port-symmetric NAT positioned in front of A. The mapped address in the source field of the packet is the same as the mapped address embedded in the Teredo IPv6 address of A. 2. The abovementioned direct bubble is dropped by B's NAT because it has not seen an outgoing packet destined to A's mapped IPv4 address. 3. A sends an indirect bubble (Packet 2) destined to B via client B's Teredo server. 4. The above-mentioned indirect bubble is received by B. B then responds with the following packets. The first packet sent by B is a direct bubble (Packet 3) destined to the mapped address/ port embedded in A's Teredo IPv6 address. 5. The above-mentioned direct bubble is dropped by A's NAT because the NAT has not seen any outgoing packet sourced from the mapped address/port embedded in A's Teredo IPv6 address and destined to the mapped address/port embedded in B's Teredo IPv6 address. Thaler Expires January 15, 2009 [Page 31] Internet-Draft Teredo Extensions July 2008 6. B also sends an indirect bubble (Packet 4) destined to A via A's Teredo Server. 7. The abovementioned indirect bubble is successfully received by A. A responds to the indirect bubble with its own direct bubble (Packet 5). This direct bubble is exactly the same as the first direct bubble (Packet 1) sent by A. 8. This time around the abovementioned direct bubble is accepted by B's NAT because it has seen an outgoing packet (Packet 3) sourced from the mapped address/port embedded in B's Teredo IPv6 address and destined to the mapped address/port embedded in A's Teredo IPv6 address. It is important to remember that A's NAT is port-symmetric and hence varies only the mapped port while the mapped address remains the same. And B's NAT is address- restricted and cares only about prior communication with the IPv4 address, and not the specific port. At this point, communication in one direction is now possible (B to A, but not vice versa). 9. After receiving the direct bubble, B remembers the new mapped address/port that was in the source fields of the direct bubble and uses those for future communication with A instead of the mapped address/port embedded in A's Teredo IPv6 address. 10. A then times out and resends an indirect bubble (Packet 6) and in response, B sends a direct bubble (Packet 7). This direct bubble is destined to the new learned mapped address/port and hence A's NAT permits the direct bubble through. Communication is now possible in the other direction (client A to B). 6.2. UPnP-enabled Symmetric NAT Extension The following protocol example illustrates the use of the UPnP- Enabled Symmetric NAT Extension in addition to the Symmetric NAT Support Extension. Assume Teredo Client A, which is positioned behind a UPnP-enabled port-symmetric NAT and wants to communicate with Teredo Client B, which is also positioned behind a UPnP-Enabled port-symmetric NAT. Before both clients start their qualification procedure, they use UPnP to reserve port mappings on their respective NATs. The UPnP operations succeed for both the clients and the clients hence know that they are positioned behind UPnP-enabled NATs. After the qualification procedure, both clients have valid Teredo IPv6 addresses because they both support the Symmetric NAT Support Extension. Also, after the qualification procedure both clients will compare their mapped address/port determined through UPnP with the mapped address/port determined through the qualification procedure. Because both will be the same, the clients will zero out their UPnP mapped address/port values and conclude that they are each located Thaler Expires January 15, 2009 [Page 32] Internet-Draft Teredo Extensions July 2008 behind a single UPnP-enabled NAT. The following packet exchange shows Teredo client A (A) establishing communication with Teredo client B (B). Teredo Client A's Client B's Teredo Client Teredo Teredo Client A NAT Server Server NAT B | | | | | | | | | Direct Bubble to B | | | 1 |------------------------------------------------------------>| | | | | | | |Indirect Bubble to B via B's Teredo Server| | | 2 |----------------------------------------->|----------------->| | | | | | | | | | Direct Bubble to A | | | |<------------------------------------------------------------| 3 | | | | | | UPnP-enabled Symmetric NAT Packet Exchange 1. A sends a direct bubble (Packet 1) to the mapped address/port embedded in B's Teredo IPv6 address. Because A's NAT is a symmetric NAT, the UDP source port field in the packet assigned by A's NAT is different from the mapped port embedded in A's Teredo IPv6 address, but the IPv4 source address of the packet is the same as the mapped address embedded in A's Teredo IPv6 address. 2. The above-mentioned direct bubble is received by B because it is destined for the UPnP mapped address/port of B and hence is let through by the NAT. At this point, B deduces that A is positioned behind a symmetric NAT because the mapped address/port from which the direct bubble is received is different from the mapped address/port that is embedded in A's Teredo IPv6 address. B also knows that itself is positioned behind a UPnP-enabled symmetric NAT. Hence, instead of storing the mapped address/port from which the direct bubble was received, it stores the mapped address/port embedded in A's Teredo IPv6 Address for further communication with A. At this point, communication in one direction is now possible (B to A, but not vice versa). 3. A also sends an indirect bubble (Packet 2) destined to B via B's Teredo Server. 4. The above indirect bubble is received by B. B then responds with a direct bubble (Packet 3) destined to the mapped address/port embedded in A's Teredo IPv6 address, as in step 2. 5. Because A's NAT is also UPnP-enabled, the above-mentioned direct bubble is received by A. A also notices that B is positioned behind a Symmetric NAT because the mapped address/port from which the packet is received is different from the mapped address/port Thaler Expires January 15, 2009 [Page 33] Internet-Draft Teredo Extensions July 2008 embedded in B's Teredo IPv6 address. Because A knows it is positioned behind a UPnP-enabled symmetric NAT, it records the mapped address/port embedded in B's Teredo IPv6 address for further communication with B. At this point, communication is now possible in the other direction (A to B). 6.3. Port-Preserving Symmetric NAT Extension The following protocol example illustrates the use of the Port- Preserving Symmetric NAT Extension. Assume Teredo Client A (A), which is positioned behind a port- preserving symmetric NAT, wants to communicate with Teredo Client B (B), which is also positioned behind a port-preserving symmetric NAT. The following packet exchange explains the configuration setup and communication setup between the two clients. Thaler Expires January 15, 2009 [Page 34] Internet-Draft Teredo Extensions July 2008 Teredo Client A's Client B's Teredo Client Teredo Teredo Client A NAT Server Server NAT B | | | | | | | | | Direct Bubble to B | | | 1 |--------------------------------------------------->| | | | | | | | |Indirect Bubble to B via B's Teredo Server| | | 2 |----------------------------------------->|----------------->| | | | | | | | | | Direct Bubble to A | | | | |<--------------------------------------------------| 3 | | | | | | | | | Direct Bubble to A | | | | |<--------------------------------------------------| 4 | | | | | | | | |Indirect Bubble to A via A's Teredo Server| |<-----------------|<-----------------------------------------| 5 | | | | | | | | | Direct Bubble to B | | | 6 |--------------------------------------------------->| | | | | | | | | | | Direct Bubble to B | | | 7 |------------------------------------------------------------>| | | | | | | |Indirect Bubble to B via B's Teredo Server| | | 8 |----------------------------------------->|----------------->| | | | | | | | | | Direct Bubble to A | | | |<------------------------------------------------------------| 9 | | | | | | Port-Preserving Symmetric NAT Packet Exchange 1. During the qualification procedure, when the clients receive a response from the Teredo server, they compare the Port value in the Origin indication with the Local Port value. If both values match, the clients set the Port-Preserving NAT flag to TRUE. 2. When the response is received from the secondary Teredo server, the mapped address/port value in the Origin indication is compared with the mapped address/port value learned from the response received from the primary server. If the mappings are different, the Symmetric NAT flag is set to TRUE. 3. It is assumed that for both clients A and B, the Port-Preserving NAT flag and the Symmetric NAT flag are set to TRUE at the end of the qualification procedure. 4. Before A sends packets to B, A checks to see if it is positioned behind a port-preserving NAT and a symmetric NAT, which in the example, it is. A also checks to see if the peer is "trusted," Thaler Expires January 15, 2009 [Page 35] Internet-Draft Teredo Extensions July 2008 but it currently is not. Next, A checks if the Random Port is set to non-zero. Since it is still zero, A allocates a new random port, begins listening on it, and stores the value in the Random Port field. 5. A sends a direct bubble (Packet 1) from the primary port to the mapped address/port embedded in B's Teredo IPv6 Address. This direct bubble does not have a Nonce Trailer or a Random Port Trailer attached to the end. 6. The abovementioned direct bubble is dropped by B's NAT because it has not seen an outgoing packet destined to A's mapped address. 7. A sends an indirect bubble (Packet 2) destined to B via client B's Teredo server. This indirect bubble contains two trailers: the Nonce Trailer containing a random nonce, and the Random Port Trailer containing the random port value from the Peer Entry. The nonce used in the Nonce Trailer is also stored in the Nonce Sent field of the Peer Entry. 8. The abovementioned indirect bubble is received by B. B adds the Teredo peer to its peer list. B saves the nonce value from the Nonce Trailer in the Nonce Advertised field of the Peer Entry. B stores the port value from the Random Port Trailer in the Peer Random Port field in the Peer Entry. 9. B responds by sending the following packets. The first packet sent by B is a direct bubble (Packet 3) destined to the mapped address/port embedded in A's Teredo IPv6 Address. This packet is sent from the primary port. It includes the Nonce Trailer with the nonce from the Nonce Advertised field of the Peer Entry. 10. The abovementioned direct bubble is dropped by A's NAT because the NAT has not seen any outgoing packet sourced from the mapped address/port embedded in A's Teredo IPv6 Address and destined to the mapped address/port embedded in B's Teredo IPv6 Address. 11. B then checks if it is positioned behind a port-restricted NAT or a symmetric NAT. It also checks if the peer has already advertised a random port. In this case, B is positioned behind a port-preserving symmetric NAT and the peer has advertised a random port; hence it needs to use a random port. It checks if its Random Port field is set to non-zero. Since it is still zero, B allocates a new random port, begins listening on it, and stores it in the Random Port entry of the Peer Entry. B then sends a direct bubble (Packet 4) destined to the mapped address embedded in A's Teredo IPv6 address and the port stored in the Peer Random Port field of the Peer Entry. The direct bubble is sent from its own random port. 12. The above direct bubble is dropped by A's NAT because the NAT has not seen any outgoing packet sourced from the mapped address embedded in A's Teredo IPv6 address and random port advertised by A. Thaler Expires January 15, 2009 [Page 36] Internet-Draft Teredo Extensions July 2008 13. B also sends an indirect bubble (Packet 5) destined to A via A's Teredo server. This indirect bubble includes a Nonce Trailer and a Random Port Trailer. The Nonce Trailer includes a new randomly generated nonce that is also stored in the Nonce Sent field of the Peer Entry. The Random Port Trailer includes the value in the Random Port field of the Peer Entry. 14. The abovementioned indirect bubble is successfully received by A. A parses the trailers and stores the nonce contained in the Nonce Trailer in the Nonce Received field of the Peer Entry. A stores the port advertised in the Random Port Trailer in the Random Port field of the Peer Entry. 15. A responds with the following packets in response to the indirect bubble received. The first packet is a direct bubble (Packet 6) sent from the primary port and is destined to the mapped address/port embedded in B's Teredo IPv6 Address. 16. The abovementioned direct bubble again is dropped by B's NAT because the NAT has not seen an outgoing packet with the same 4-tuple as the incoming packet. 17. The next packet is also a direct bubble (Packet 7) and this one is sent from A's random port. The packet is destined to the mapped address embedded in B's Teredo IPv6 address and the Peer Random Port stored in the Peer Entry. 18. Because both NATs are port-preserving NATs and the random ports have not been used for any other mapping, the abovementioned direct bubble is received by B because B's NAT has seen an outgoing packet (Packet 4) with the same address/port pairs. B stores the address/port from which the direct bubble was received in the mapped address/port fields of the Peer Entry. It changes the status of the peer to "trusted" and sets the Direct Receive on Random Port field to TRUE. At this point, communication in one direction is now possible (B to A, but not vice versa). 19. Because A still considers B to be "not-trusted," it times out and retransmits an indirect bubble (Packet 8). This packet contains a new nonce as part of the Nonce Trailer and also contains the value of the random port as part of the Random Port Trailer. 20. B receives the abovementioned indirect bubble. The processing of this indirect bubble is similar to the processing of Packet 2. Since B received a direct bubble on its random port, it does not respond with a direct bubble from its primary port. Instead, it responds with a direct bubble (Packet 9) sent from its random port, which is similar to Packet 4 mentioned above. 21. A receives the direct bubble sent by B. A stores the mapped address/port from which the direct bubble was received in mapped address/port fields in the Peer Entry. A changes the status of B to "trusted" and sets the Direct Receive on the Random Port field to TRUE. At this point, the communication is now possible Thaler Expires January 15, 2009 [Page 37] Internet-Draft Teredo Extensions July 2008 in the other direction (A to B). 6.4. Hairpinning Extension The following protocol example illustrates the use of the Hairpinning Extension. In Figure 3 Figure 3 (section Section 3.4), Teredo Client A (A) and Teredo Client B (B) are positioned behind different immediate NATs in a two-layer NAT topology; that is, the outermost NAT (say NAT E) is common to both A and B but the immediate NATs that they are connected to are different (say A is connected to NAT F while B is connected to NAT G). Further let's assume that the immediate NATs that A and B are connected to are UPnP-enabled (NAT F and NAT G are UPnP-enabled). We assume that NAT E does not support hairpinning; that is, the NAT does not relay packets originating from the private address space and destined for the public address of the NAT, back to the private address of the NAT. Before starting the qualification procedure, both A and B use UPnP to reserve port mappings on their respective NATs. They observe that the UPnP operation succeeds and both clients obtain valid UPnP Mapped Address/Port values. Next, both client A and client B implement the qualification procedure where they determine their mapped address/port values, as specified in [RFC4380] section 5.2.1. A and B both compare their UPnP Mapped Address/Port values with the mapped address/port values obtained through the qualification procedure. Because both A and B are part of a two-layer NAT topology, these values will be different. Hence both A and B continue to hold on to their UPnP Mapped Address/Port. The following packet exchange shows client A establishing communication with client B. Thaler Expires January 15, 2009 [Page 38] Internet-Draft Teredo Extensions July 2008 Teredo Teredo Client A's Client B's Client NAT Client NAT NAT Teredo Teredo A F B G E Server Server | | | | | | | | | Direct Bubble to B | | | | 1 |-------------------------------------->| | | | | | | | | | | Indirect Bubble to B via B's Teredo Server | 2 |----------------------------------------------------------->| | | |<----------------------------------------| | | | | | | | | | | Direct Bubble to A | | | 3 | | |------------------->| | | | | | | | | | | | | Direct | | | | | | |Bubble to A| | | | 4 | | |---------->| | | | | | | | | | | | | | Direct | | | | | | |Bubble to A| | | | 5 | | |---------->| | | | |<-----------------------------| | | | | | | | | | | | | | Indirect Bubble to A | | 6 | | |---------------------------->| | |<-----------------------------------------------| | | | | | | | | |Direct Bubble to B| | | | | 7 |----------------->| | | | | | | | | | | | Hairpinning-based Packet Exchange 1. A sends a direct bubble (Packet 1) to the mapped address/port embedded in B's Teredo IPv6 address. 2. The abovementioned direct bubble is dropped by NAT E, because it does not support Hairpinning. 3. A sends out an indirect bubble (Packet 2) destined to B via B's Teredo Server. In this indirect bubble, A includes the Alternate Address Trailer which includes both the local address/ port and the UPnP mapped address/port. 4. The abovementioned indirect bubble is received by B. After parsing the Alternate Address Trailer, B has a total of three addresses to communicate with: two from the Alternate Address Trailer and one from the mapped address/port embedded in A's Teredo IPv6 address. B then responds with the following packets. The first packet sent by B is a direct bubble (Packet 3) destined to the mapped address/port embedded in A's Teredo IPv6 address. Thaler Expires January 15, 2009 [Page 39] Internet-Draft Teredo Extensions July 2008 5. The abovementioned direct bubble will be dropped by the NAT E because it does not support Hairpinning. 6. Since the local address/port was the first mapping in the Alternate Address Trailer, the second direct bubble (Packet 4) sent by B is destined to the local address/port. 7. The abovementioned direct bubble is dropped because A and B are positioned behind different NATs and hence have their own private address space. A's local address is not reachable from B. 8. The next direct bubble (Packet 5) is sent by B destined to A's UPnP mapped address/port, which is the second mapping in the Alternate Address Trailer sent by A. 9. The abovementioned direct bubble is received by A because A's UPnP-mapped address is reachable from B. A stores the source address from which the direct bubble was received in the mapped address/port fields of the Peer Entry, as defined in [RFC4380] section 5.2. Also, the mapped address status field (as specified in [RFC4380] section 5.2.3) is changed to "trusted." At this point, communication in one direction is now possible (A to B, but not vice versa). 10. B also sends an indirect bubble (Packet 6) to A via A's Teredo server. As part of the indirect bubble, B also includes an Alternate Address Trailer, which contains the local address/port and the UPnP mapped address/port of B. 11. The abovementioned indirect bubble is received by A. After parsing the Alternate Address Trailer, A adds the two addresses in the Alternate Address Trailer to the Alternate Address List in the Peer Entry. Since the peer's mapping is "trusted" (point 9), A responds with only one direct bubble (Packet 7) that is sent to the mapped address/port stored in the Peer Entry. 12. The abovementioned direct bubble is received by B. B records the mapped address/port from which the direct bubble was received in the mapped address/port field in its Peer Entry, and change the status of the mapped address to "trusted." At this point, communication is now possible in the other direction (B to A). 6.5. Server Load Reduction Extension The following protocol example illustrates the use of the Server Load Reduction Extension. Assume Teredo client A (A) has established communication with Teredo Client B (B). Also assume that at some later point when no data packets have been exchanged between both clients for more than 30 seconds, the communication needs to be re-established as A wants to send a data packet to B. The following packet exchange helps A re-establish communication with Thaler Expires January 15, 2009 [Page 40] Internet-Draft Teredo Extensions July 2008 B. Teredo Client A's Client B's Teredo Client Teredo Teredo Client A NAT Server Server NAT B | | | | | | | | | Direct Bubble to B | | | 1 |------------------------------------------------------------>| | | | | | | | | | Direct Bubble to A | | | |<------------------------------------------------------------| 2 | | | | | | Server Load Reduction Packet Exchange 1. A sends a direct bubble (Packet 1) with the Neighbor Discovery Option Trailer, with the DiscoveryType field set to TeredoDiscoverySolicitation. 2. If the mapping on either of the NATs has not expired, the direct bubble is received by B. B parses the Neighbor Discovery Option and because the DiscoveryType was set to TeredoDiscoverySolicitation, B responds with a direct bubble (Packet 2). B's direct bubble also contains the Neighbor Discovery Option and the DiscoveryType is set to TeredoDiscoveryAdvertisement. 3. The abovementioned direct bubble is received by A and at this point, communication between the Teredo clients is re- established. 7. Security Considerations Security considerations are the same as those specified in [RFC4380] section 7. In addition, the Hairpinning Extension introduces the possibility of an amplification attack if a malicious user could advertise a large number of port mappings in the Alternate Address Trailer, resulting in a large number of direct bubbles sent in response. Because of this, section Section 4.2 explicitly limits the number of addresses that a Teredo client will accept. Because the nonce in the Nonce Trailer is used (as specified in section Section 5.1.2.4) to prevent spoofing of bubbles that would result in directing traffic to the wrong place, it is important that the nonce be random so that attackers cannot predict its value. Thaler Expires January 15, 2009 [Page 41] Internet-Draft Teredo Extensions July 2008 8. IANA Considerations [RFC Editor: please remove this section prior to publication.] This document has no IANA Actions. 9. References 9.1. Normative References [RFC1918] Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and E. Lear, "Address Allocation for Private Internets", BCP 5, RFC 1918, February 1996. [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through Network Address Translations (NATs)", RFC 4380, February 2006. [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, September 2007. [UPNPWANIP] UPnP Forum, "Internet Gateway Device (IGD) V 1.0", November 2001, . 9.2. Informative References [RFC2463] Conta, A. and S. Deering, "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification", RFC 2463, December 1998. Thaler Expires January 15, 2009 [Page 42] Internet-Draft Teredo Extensions July 2008 Author's Address Dave Thaler Microsoft Corporation One Microsoft Way Redmond, WA 98052 USA Phone: +1 425 703 8835 Email: dthaler@microsoft.com Thaler Expires January 15, 2009 [Page 43] Internet-Draft Teredo Extensions July 2008 Full Copyright Statement Copyright (C) The IETF Trust (2008). This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights. This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST 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. Intellectual Property The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79. Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf-ipr@ietf.org. Thaler Expires January 15, 2009 [Page 44]