IPv6 Operations Working Group J. Hoagland Internet-Draft Symantec Expires: January 8, 2009 S. Krishnan Ericsson D. Thaler Microsoft July 7, 2008 Security Concerns With Tunneling draft-ietf-v6ops-tunnel-security-concerns-00 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 8, 2009. Abstract A number of security concerns with tunnels are documented. The primary intent of this document is to raise the awareness regarding the security issues with tunnels as deployed today. Hoagland, et al. Expires January 8, 2009 [Page 1] Internet-Draft Tunneling Security Concerns July 2008 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Tunnels May Bypass Security . . . . . . . . . . . . . . . . . 3 2.1. Network Security Bypass . . . . . . . . . . . . . . . . . 3 2.2. IP Ingress and Egress Filtering Bypass . . . . . . . . . . 5 2.3. Source Routing After the Tunnel Client . . . . . . . . . . 6 3. Challenges in Inspecting and Filtering Content of Tunneled Data Packets . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.1. Inefficiency of Selective Network Filtering of All Tunneled Packets . . . . . . . . . . . . . . . . . . . . . 7 3.2. Problems with deep packet inspection of tunneled data packets . . . . . . . . . . . . . . . . . . . . . . . . . 8 4. Increased Exposure Due to Tunneling . . . . . . . . . . . . . 9 4.1. NAT Holes Increase Attack Surface . . . . . . . . . . . . 9 4.2. Exposure of a NAT Hole . . . . . . . . . . . . . . . . . . 11 4.3. Public Tunnels Widen Holes in Restricted NATs . . . . . . 12 5. Tunnel Address Concerns . . . . . . . . . . . . . . . . . . . 12 5.1. Feasibility of Guessing Tunnel Addresses . . . . . . . . . 12 5.2. Profiling Targets Based on Tunnel Address . . . . . . . . 13 6. Additional Security Concerns . . . . . . . . . . . . . . . . . 14 6.1. Attacks Facilitated By Changing Tunnel Server Setting . . 15 7. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 17 8. Security Considerations . . . . . . . . . . . . . . . . . . . 17 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 17 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 17 10.1. Normative References . . . . . . . . . . . . . . . . . . . 17 10.2. Informative References . . . . . . . . . . . . . . . . . . 18 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 18 Intellectual Property and Copyright Statements . . . . . . . . . . 20 Hoagland, et al. Expires January 8, 2009 [Page 2] Internet-Draft Tunneling Security Concerns July 2008 1. Introduction With NATs becoming increasingly more prevalent, there have recently been many tunneling protocols developed that go through NATs or firewalls by tunneling over UDP or TCP. For example, Teredo [RFC4380], L2TPv2 [RFC2661], and L2TPv3 [RFC3931] all tunnel IP packets over UDP. Similarly, many SSL VPN solutions that tunnel IP packets over HTTP (and hence over TCP) are deployed today. This document discusses security concerns with tunneling IP packets, and includes recommendations where relevant. The primary intent of this document is to provide information that can be used in any new or updated tunnel protocol specification. Secondarily, this document can help improve security deployments using tunnel protocols. 2. Tunnels May Bypass Security 2.1. Network Security Bypass 2.1.1. Problem Tunneled IP traffic may not receive the intended level of inspection or policy application by network-based security devices unless such devices are specifically tunnel-aware. This reduces defense in depth and may cause security gaps. This applies to all network-located devices and to any end-host based firewalls whose existing hooking mechanism(s) would not show them the IP packet stream after the tunnel client does decapsulation. 2.1.2. Discussion Evasion by tunneling is often a problem for network-based security devices such as network firewalls, intrusion detection and prevention systems, and router controls. To provide such functionality in the presence of tunnels, the developer of such devices must add support for detunneling for each new protocol. There is typically a significant lag between when the security developer recognizes that a tunnel will be used (or will be remotely usable) to a significant degree and when the detunneling can be implemented in a product update, the update tested and released, and customers begin using the update. Late changes in the protocol specification or in the way it is implemented can cause additional delays. This becomes a significant security concern when a delay in applied coverage is occurring frequently. Hoagland, et al. Expires January 8, 2009 [Page 3] Internet-Draft Tunneling Security Concerns July 2008 For example, for L2TP or Teredo, an unaware network security device would inspect or regulate the outer IP and the IP-based UDP layer as normal, but it would not recognize that there is an additional IP layer contained inside the UDP payload that it needs to apply the same controls as it would to a native packet. (Of course, if the device discards the packet due to something in the IP or UDP header, such as referring to an unknown protocol, the embedded packet is no longer a concern.) In addition, if the tunnel does encryption, the network-based security device may not be able to do much, just as if IPsec end-to-end encryption were used without tunneling. Network security controls being not applied must be a concern to those that set them up, since those controls are supposed to adequately regulate all traffic. If network controls are being bypassed due to the use of tunneling, the burden of controls shifts to the tunnel client host. Since security administrators may not have full control over all the nodes on their network, they sometimes prefer to implement security controls on the network. One implication of the security control bypass is that defense in depth has been reduced, perhaps down to zero unless a 'local firewall' is in use as recommended in [RFC4380]. However, even if there are host-based security controls that recognize tunnels, security administrators may not have configured them with full security control parity, even if all controls that were maintained by the network are available on the host. Thus there may be gaps in desired coverage. Compounding this is that, unlike what would be the case for native IP, some network administrators will not even be aware that their hosts are globally reachable, if the tunnel provides connectivity to/ from the Internet; for example, they may not be expecting this for hosts with [RFC1918] addresses behind a NAT. In addition, Section 3.2 discusses how it may not be efficient to find all tunneled traffic for network devices to examine. 2.1.3. Recommendations Security administrators who do not consider tunneling an acceptable risk should disable tunnel functionality unless their network-based security controls adequately recognize the tunneled traffic. However, there may be an awareness gap. Thus, due to the possible negative security consequences, we recommend that explicit user action be required to enable tunneling, at least for the first time. For example, when Teredo is being enabled or when it is going to be used for the first time, there could be a descriptive warning about the possible reduction of defense in depth that will occur. In Hoagland, et al. Expires January 8, 2009 [Page 4] Internet-Draft Tunneling Security Concerns July 2008 addition, Teredo client functionality should be easy to disable on the host and through a central management facility if one is provided. (We could find no pre-existing mechanism for tunneling protocols to use that could automate their functionality being disabled unless all network-based security controls were aware of it. A separate type of consent request packet would be needed.) To minimize security exposure due to tunnels, we recommend that a tunnel be an interface of last resort, independent of IP version. Specifically, we suggest that when both native and tunneled access to a remote host is available, that the native-based access be used in preference to tunneled access. This should also promote greater efficiency and reliability. Note that although Rule 7 of [RFC3484] section 6 will prefer native connectivity over tunnels, this rule is only a tie-breaker when a choice is not made by earlier rules; hence tunneling mechanisms that are tied to a particular range of IP address space will be decided based on the prefix precedence. For example, using the prefix policy mechanism of [RFC3484] section 2.1, Teredo might have a precedence of 5 so that native IPv4 is preferred over Teredo. 2.2. IP Ingress and Egress Filtering Bypass 2.2.1. Problem IP addresses inside tunnels are not subject to ingress and egress filtering in the network they tunnel over, unless extraordinary measures are taken. Only the tunnel endpoints can do such filtering. 2.2.2. Discussion Ingress filtering (sanity-checking incoming destination addresses) and egress filtering (sanity-checking outgoing source addresses) are done to mitigate attacks and to make it easier to identify the source of a packet and are considered to be a good practice. This is most naturally (and in the general case, by requirement) done at network boundaries. Tunneled IP traffic bypassing this network control is a specific case of Section 2.1, but is illustrative. 2.2.3. Recommendations The recommendations in Section 2.1.3 can help here. For this problem specifically, there are three locations in which ingress and egress filtering can be done. Hoagland, et al. Expires January 8, 2009 [Page 5] Internet-Draft Tunneling Security Concerns July 2008 Network based: Network-based devices (e.g., routers) could be updated to find all tunneled packets and to apply ingress and egress controls equally to tunneled IP addresses. Tunnel server based: Tunnel servers can apply ingress and egress controls to tunneled IP addresses passing through them to and from tunnel clients. Host based: Tunnel clients could make an effort to conduct ingress and egress filtering. Protocols that embed an IPv4 address in a tunneled IPv6 address directly between peers often do filtering based on checking the correpondence. Protocols that accept tunneled packets directly from a server or relay do filtering the same way as it would be done on a native link with traffic from a router. To do host-based filtering with protocols that allow both other hosts and a router over a common tunnel (e.g., 6to4 [RFC3056], Teredo, ISATAP [RFC4214], and 6over4 [RFC2529]), hosts would need to be able to know the outer IP address of all routers from which it could receive traffic. This might be learned via Secure Neighbor Discovery (SEND) [RFC3971] or some other mechanism (e.g., [RFC4214] section 8.3.2). 2.3. Source Routing After the Tunnel Client 2.3.1. Problem If the encapsulated IP packet specifies source routing beyond the recipient tunnel client, the host may forward the IP packet to the specified next hop. This may be unexpected and contrary to administrator wishes and may have bypassed network-based source routing controls. 2.3.2. Discussion A detailed discussion of issues related to source routing can be found in [RFC5095]. 2.3.3. Recommendations Tunnel clients should by default discard tunneled IP packets that specify additional routing, as recommended in [RFC5095], though they may also allow the user to configure what source routing types are allowed. All pre-existing source routing controls should be upgraded Hoagland, et al. Expires January 8, 2009 [Page 6] Internet-Draft Tunneling Security Concerns July 2008 to apply these controls to tunneled IP packets as well. 3. Challenges in Inspecting and Filtering Content of Tunneled Data Packets 3.1. Inefficiency of Selective Network Filtering of All Tunneled Packets 3.1.1. Problem There is no mechanism to both efficiently and immediately filter all tunneled packets. This limits the ability to prevent tunnel use on a network. 3.1.2. Discussion Given concerns about tunnel security or a network's lack of preparedness for tunnels, a network administrator may wish to simply block all use of tunnels. He or she may wish to do so using network controls; this could be either due to not having confidence in the ability to disable it on all hosts attached to the network or due to wanting an extra layer of prevention. One simple method to do that is easy to employ for many tunnel protocols is to block outbound packets to the UDP or TCP port used (e.g., UDP port 3544 for Teredo, UDP port 1701 for L2TP, etc.). This prevents a tunnel client from establishing a new tunnel. However, existing tunnels will not necessarily be affected if the blocked port is used only for initial setup. In addition, if the blocking is applied on the outside of the client's NAT, the NAT will retain the port mapping for the client and the client may or may not continue to use the IP address assigned to its tunnel. It is not known if blocking all traffic to a given outbound port will interfere with non-tunneled traffic. Another simple alternative, if the tunnel server addresses are well- known, is to filter out all traffic to/from such addresses. The other approach is to find all packets to block in the same way as would be done for inspecting all packets (Section 3.2). However, this faces the difficulties in terms of efficiency of filtering as was present there. 3.1.3. Recommendations Tunneling over UDP or TCP (including HTTP) to reach the Internet is not recommended as a solution for managed networks. (Windows, for Hoagland, et al. Expires January 8, 2009 [Page 7] Internet-Draft Tunneling Security Concerns July 2008 example, disables Teredo by default if it detects that it is within a managed network.) Administrators of such networks may wish to filter all tunneled traffic at the boundaries of their networks. It is sufficient to filter out the tunneled connection requests (if they can be identified) to stop further tunneled traffic. The easiest mechanism for this would be to filter out outgoing traffic sent to the destination port defined by the tunneling protocol, and incoming traffic with that source port. 3.2. Problems with deep packet inspection of tunneled data packets 3.2.1. Problem There is no efficient mechanism for network-based devices to inspect the contents of all tunneled data packets, the way they can for native packets. This makes it difficult to apply the same controls as they do to native IP. 3.2.2. Discussion Some tunnel protocols are easy to identify, such as if all data packets are encapsulated using a well-known UDP or TCP port that is unique to the protocol. Other protocols, however, either use dynamic ports for data traffic, or else share ports with other protocols (e.g., tunnels over HTTP). The implication of this is that network-based devices that wish to passively inspect (and perhaps selectively apply policy to) all encapsulated traffic must inspect all TCP or UDP packets (or at least all packets not part of a session that is known not to be a tunnel). This is imperfect since a heuristic must then be applied to determine if a packet is indeed part of a tunnel. This may be too slow to make use of in practice, especially if it means that all TCP or UDP packets must be taken off of the device's "fast path". One heuristic that can be used on packets to determine if they are tunnel-related or not is as follows. For each known tunnel protocol, attempt parsing the packet as if it were a packet of that protocol, destined to the local host (i.e., where the local host has the destination address in the inner IP header, if any). If all syntax checks pass, up to and including the inner IP header (if the tunnel doesn't use encryption), then treat the packet as if it is a tunneled packet of that protocol. It is possible that non-tunnel packets will match as tunneled using Hoagland, et al. Expires January 8, 2009 [Page 8] Internet-Draft Tunneling Security Concerns July 2008 this heuristic, but tunneled packets (of the known types of tunnels) should not escape inspection, absent implementation bugs. For some protocols, it may be possible to monitor setup exchanges to know to expect that data will be exchanged on certain ports later. (Note that this does not necessarily apply to Teredo, for example, since communicating with another Teredo client behind a cone NAT does not require such signaling. However, deprecation of the cone bit as discussed in [RFC4380] means this technique may indeed work with Teredo.) 3.2.3. Recommendations As illustrated above, it should be clear that inspecting the contents of tunneled data packets is highly complex and often impractical. For this reason, if a network wishes to monitor IP traffic, tunneling is not recommended. For example, to provide an IPv6 transition solution, the network should provide native IPv6 connectivity or a tunnel solution (e.g., ISATAP or 6over4) that encapsulates data packets between hosts and a router within the network. 4. Increased Exposure Due to Tunneling 4.1. NAT Holes Increase Attack Surface 4.1.1. Problem If the tunnel allows inbound access from the public Internet, the opening created in a NAT due to a tunnel client increases its Internet attack surface area. If vulnerabilities are present, this increased exposure can be used by attackers and their programs. If the tunnel allows inbound access only from a private network (e.g., a remote network to which one has VPN'ed), the opening created in the NAT still increases its attack surface area, although not as much as in the public Internet case. 4.1.2. Discussion When a tunnel is active, a mapped port is maintained on the NAT through which remote hosts can send packets and perhaps establish connections. The following sequence is intended to sketch out the processing on the tunnel client host that can be reached through this; the actual processing for a given host may be somewhat different. Hoagland, et al. Expires January 8, 2009 [Page 9] Internet-Draft Tunneling Security Concerns July 2008 1. Link-layer protocol processing 2. (Outer) IP host firewall processing 3. (Outer) IP processing by stack 4. UDP/TCP processing by stack 5. Tunnel client processing 6. (Inner) IP host firewall processing 7. (Inner) IP processing by stack 8. Various upper layer processing may follow The inner firewall (and other security) processing may or may not be present, but if it is, some of the inner IP processing may be filtered. (For example, [RFC4380] section 7.1 recommends that an IPv6 host firewall be used on all Teredo clients.) (By the virtue of the tunnel being active, we can infer that the firewall is unlikely to do any filtering based on the outer IP.) Any of this processing may expose vulnerabilities an attacker can exploit; similarly these may expose information to an attacker. Thus, even if firewall filtering is in place (as is prudent) and filters all incoming packets, the exposed area is larger than if a native IP Internet connection were in place, due to the processing that takes place before the inner IP is reached (specifically, the UDP/TCP processing, the tunnel client processing, and additional IP processing, especially if one is IPv4 and the other is IPv6). One possibility is that a layer 3 targeted worm makes use of a vulnerability in the exposed processing. While the main benefit to worms from tunneling is targeting at layer 3 reaching the end host, even a throughly firewalled host could be subject to a worm that spreads with a single UDP packet if the right remote code vulnerability is present. 4.1.3. Recommendations This problem seems inherent in tunneling being active on a host, so the solution seems to be to minimize tunneling use. For example, it can be active only when it is really needed and only for as long as needed. So, the tunnel interface can be initially not configured and only used when it is entirely the last resort. The interface should then be deactivated (ideally, automatically) again Hoagland, et al. Expires January 8, 2009 [Page 10] Internet-Draft Tunneling Security Concerns July 2008 as soon as possible. Note however that the hole will remain in the NAT for some amount of time after this, so some processing of incoming packets is inevitable (unless the client's native IP address behind the NAT is changed). 4.2. Exposure of a NAT Hole 4.2.1. Problem Attackers are more likely to know about a tunnel client's NAT hole than a typical hole in the NAT. If they know about the hole, they could try to use it. 4.2.2. Discussion There are at least three reasons why an attacker may be more likely to learn of the tunnel client's exposed port than a typical NAT exposed port: 1. The NAT mapping for a tunnel is typically held open for a significant period of time, and kept stable. This increases the chance of it being discovered. 2. In some protocols (e.g., Teredo), the external IP address and port are contained in the client's address that is used end-to- end and possibly even advertised in a name resolution system. While the tunnel protocol itself might only distribute this address in IP headers, peers, routers, and other on-path nodes still see the client's IP address. Although this point does not apply directly to protocols (e.g., L2TP) that do not construct the inner IP address based on the outer IP address, the inner IP address is still known to peers, routers, etc. and can still be reached by attackers without knowing the external IP address or port. 3. The tunnel protocol contains more messages that are exchanged and with more parties (e.g., due to a longer path length) than without using the tunnel, offering more chance for visibility into the port and address in use. 4.2.3. Recommendations The recommendations from Section 4.1 seem to apply here as well: minimize tunnel use. Hoagland, et al. Expires January 8, 2009 [Page 11] Internet-Draft Tunneling Security Concerns July 2008 4.3. Public Tunnels Widen Holes in Restricted NATs 4.3.1. Problem Tunnels that allow inbound connectivity from the Internet (e.g., Teredo, tunnel brokers, etc) essentially turn a restricted or symmetric NAT into an unrestricted one, for all tunnel client ports. This eliminates NAT filtering for such ports and may eliminate the need for an attacker to spoof an address. 4.3.2. Discussion Restricted, port restricted, and symmetric NATs [RFC3489] limit the source of incoming packets to just those that are a previous destination. This poses a problem for tunnels that intend to allow inbound connectivity from the Internet. Various protocols (e.g., Teredo, STUN [RFC3489], etc.) provide a facility for peers, upon request, to become a previous destination. This works by sending a "bubble" packet via a server, which is passed to the client, and then sent by the client (through the NAT) to the originator. This removes any NAT-based barrier to attackers sending packets in through the client's service port. In particular, an attacker would no longer need to either be an actual previous destination or to forge its addresses as a previous destination. When forging, the attacker would have had to learn of a previous destination and then would face more challenges in seeing any returned traffic. 4.3.3. Recommendations Minimizing public tunnel use (see Section 4.1.3) would lower the attack opportunity related to this exposure. 5. Tunnel Address Concerns 5.1. Feasibility of Guessing Tunnel Addresses 5.1.1. Problem For some types of tunneling protocols, it may be feasible to guess IP addresses assigned to tunnels, either when looking for a specific client or when looking for an arbitrary client. This is in contrast to native IPv6 addresses in general, but is no worse than for native IPv4 addresses today. Hoagland, et al. Expires January 8, 2009 [Page 12] Internet-Draft Tunneling Security Concerns July 2008 For example, some protocols (e.g., 6to4 and Teredo) use well-defined address ranges. As another example, using well-known public servers for Teredo or tunnel brokers also implies using a well known address range. 5.2. Profiling Targets Based on Tunnel Address 5.2.1. Problem An attacker encountering an address associated with a particular tunneling protocol or well-known tunnel server has the opportunity to infer certain relevant pieces of information that can be used to profile the host before sending any packets. This can reduce the attacker's footprint and increase the attacker's efficiency. 5.2.2. Discussion The tunnel address reveals some information about the nature of the client. o That a host has a tunnel address associated with a given proocol means that the client is running on some platform for which there exists a tunnel client implementation of that protocol. In addition, if some platforms have that protocol installed by default and where the host's default rules for using it make it susceptible to being in use, then it is more likely to be running on such a platform than on one where it is not used by default. For example, as of this writing, seeing a Teredo address suggests that the host it is on is running Windows Vista. o Similarly, the use of an address associated with a particular tunnel server also suggests some information. Tunnel client software is often deployed, installed, and/or configured using some degree of automation. It seems likely that the majority of the time the tunnel server that results from the initial configuration will go unchanged from the initial setting. Moreover, the server that is configured for use may be associated with a particular means of installation, which often suggests the platform. For example, if the server field in a Teredo address is one of the IPv4 addressees to which teredo.ipv6.microsoft.com resolves, it suggests that the host is running Windows. o The external IPv4 address of a NAT can of course be readily associated with a particular organization or at least an ISP, and hence putting this address in an IPv6 address reveals this information. However, this is no different than using a native IP address, and hence is not new with tunneling. Hoagland, et al. Expires January 8, 2009 [Page 13] Internet-Draft Tunneling Security Concerns July 2008 o It is also possible that external client port numbers may be more often associated with particular client software or the platform on which it is running. The usefulness of this for platform determination is, however, reduced by the different NAT port number assignment behaviors. In addition, the same observations would apply to use of UDP or TCP over native IP as well, and hence this is not new with tunneling. The platform, tunnel client software, or organization information can be used by an attacker to target attacks more carefully. For example, an attacker may decide to attack an address only if it is likely to be associated with a particular platform or tunnel client software with a known vulnerability. (This is similar to the ability to guess some platforms based on the OUI in the EUI-64 portion of an IPv6 address generated from a MAC address, since some platforms are commonly used with interface cards from particular vendors.) In Teredo as defined in [RFC4380], the cone bit tells the attacker whether a bubble is needed to proceed a connection. It may also have some value in terms of profiling to the extent that it reveals the security posture of the network. If the cone bit is set, the attacker may decide it is fruitful to port scan the embedded external IPv4 address and others associated with the same organization, looking for open ports. As such, this cone bit is deprecated in Windows Vista. 5.2.3. Recommendations If installation programs randomized the server setting, that would reduce the extent to which they can be profiled. Similarly, administrators can choose to change the default setting to reduce the degree to which they can be profiled ahead of time. Randomizing the tunnel client port in use would mitigate any profiling that can be done based on the external port, especially if multiple different Teredo clients did this. Further discussion on randomizing ports can be found at [I-D.ietf-tsvwg-port-randomization]. It is recommended that tunnel protocols minimize the propagation knowledge about whether the NAT is a cone NAT. For example, the cone bit for Teredo should be deprecated. 6. Additional Security Concerns Hoagland, et al. Expires January 8, 2009 [Page 14] Internet-Draft Tunneling Security Concerns July 2008 6.1. Attacks Facilitated By Changing Tunnel Server Setting 6.1.1. Problem If an attacker could either change a tunnel client's server setting or change the IP addresses to which a configured host name resolves (e.g., by intercepting DNS queries), this would allow them to become a man-in-the-middle at least monitor peer communication and at worst pretend to represent the remote peer. 6.1.2. Discussion A client's server has good visibility into the client's communication with IP peers. If the server were switched to one that records this information and makes it available to third parties (e.g., advertisers, competitors, spouses, etc.) then sensitive information is being disclosed, especially if the client's host prefers the tunnel over native IP. Assuming the server provides good service, the user would not have reason to suspect the change. Full interception of IP traffic could also be arranged (including pharming) which would allow any number of deception or monitoring attacks including phishing. We illustrate this with an example phishing attack scenario. It is often assumed that the tunnel server is a trusted entity. It may be possible for malware or a malicious user to quietly change the Teredo client's server setting and have the user be unaware their trust has been misplaced for an indefinite period of time. However, malware or a malicious user can do much worse than this, so this is not a significant concern. Hence it is only important that an attacker on the network cannot change the client's server setting. 1. A phisher sets up a malicious tunnel server (or tampers with a legitimate one). This server, for the most part, provides correct service. 2. An attacker by some means and switches the host's tunnel server setting, or spoofs a DNS reply, to point to the above server. If neither DNS nor the tunnel setup is secured (i.e., if the client does not authenticate the information), then the attacker's tunnel server is seen as legitimate. 3. A user on the victim host types their bank's URL into his/her browser. 4. The bank's hostname resolves to one or more IP addresses and the tunnel is selected for socket connection for whatever reason Hoagland, et al. Expires January 8, 2009 [Page 15] Internet-Draft Tunneling Security Concerns July 2008 (e.g., the tunnel provides IPv6 connectivity and the bank has an IPv6 address). 5. The tunnel client uses the server for help in connecting to the the bank's IP address. Some tunneling protocols use a separate channel for signaling vs data, but this still allows the server to place itself in the data path by an appropriate signal to the client. For example, in Teredo, the client sends a ping request through a server which is expected to come back through a data relay, and a malicious server can simply send it back itself to indicate that is a data relay for the communication. 6. The rest works pretty much like any normal phishing transaction, except that the attacker acts as a tunnel server (or data relay, for protocols such as Teredo) and a host with the bank's IP address. This pharming type attack is not unique to tunneling. Switching DNS server settings to a malicious DNS server could have similar effect. 6.1.3. Recommendations In general, anti-phishing and anti-fraud provisions should help with aspects of this, as well as software that specifically monitors for tunnel server changes. Perhaps the best way to mitigate tunnel-specific attacks is to have the client either authenticate the tunnel server, or at least the means by which the tunnel server's IP address is determined. For example, SSL VPNs use https URLs and hence authenticate the server as being the expected one. Another mechanism, when IPv6 Router Advertisements are sent over the tunnel (e.g., in Teredo), is to use SEcure Neighbor Discovery (SEND) to verify that the client trusts the server. On the host, it should require an appropriate level of privilege in order to change the tunnel server setting (as well as other non- tunnel-specific settings such as the DNS server setting, etc.). Making it easy to see the current tunnel server setting (e.g., not requiring privilege for this) should help detection of changes. The scope of the attack can also be reduced by limiting tunneling use in general but especially in preferring native IPv4 to tunneled IPv6; this is because it is reasonable to expect that banks and similar web sites will continue to be accessible over IPv4 for as long as a significant fraction of their customers are still IPv4-only. Hoagland, et al. Expires January 8, 2009 [Page 16] Internet-Draft Tunneling Security Concerns July 2008 7. Acknowledgments The authors would like to thank Remi Denis-Courmont, Fred Templin, Jordi Palet Martinez, James Woodyatt and Christian Huitema for reviewing earlier versions of the document and providing comments to make this document better. The authors would also like to thank Alfred Hines for a careful review of the document. 8. Security Considerations This entire document discusses security concerns with tunnels. 9. IANA Considerations There are no actions for IANA in this document. 10. References 10.1. 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. [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 Domains without Explicit Tunnels", RFC 2529, March 1999. [RFC2661] Townsley, W., Valencia, A., Rubens, A., Pall, G., Zorn, G., and B. Palter, "Layer Two Tunneling Protocol "L2TP"", RFC 2661, August 1999. [RFC3056] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via IPv4 Clouds", RFC 3056, February 2001. [RFC3484] Draves, R., "Default Address Selection for Internet Protocol version 6 (IPv6)", RFC 3484, February 2003. [RFC3489] Rosenberg, J., Weinberger, J., Huitema, C., and R. Mahy, "STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)", RFC 3489, March 2003. [RFC3931] Lau, J., Townsley, M., and I. Goyret, "Layer Two Tunneling Protocol - Version 3 (L2TPv3)", RFC 3931, March 2005. Hoagland, et al. Expires January 8, 2009 [Page 17] Internet-Draft Tunneling Security Concerns July 2008 [RFC3971] Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure Neighbor Discovery (SEND)", RFC 3971, March 2005. [RFC4214] Templin, F., Gleeson, T., Talwar, M., and D. Thaler, "Intra-Site Automatic Tunnel Addressing Protocol (ISATAP)", RFC 4214, October 2005. [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. [RFC5095] Abley, J., Savola, P., and G. Neville-Neil, "Deprecation of Type 0 Routing Headers in IPv6", RFC 5095, December 2007. 10.2. Informative References [I-D.ietf-tsvwg-port-randomization] Larsen, M. and F. Gont, "Port Randomization", draft-ietf-tsvwg-port-randomization-01 (work in progress), February 2008. Authors' Addresses James Hoagland Symantec Corporation 350 Ellis St. Mountain View, CA 94043 US Email: Jim_Hoagland@symantec.com URI: http://symantec.com/ Suresh Krishnan Ericsson 8400 Decarie Blvd. Town of Mount Royal, QC Canada Phone: +1 514 345 7900 x42871 Email: suresh.krishnan@ericsson.com Hoagland, et al. Expires January 8, 2009 [Page 18] Internet-Draft Tunneling Security Concerns July 2008 Dave Thaler Microsoft Corporation One Microsoft Way Redmond, WA 98052 USA Phone: +1 425 703 8835 Email: dthaler@microsoft.com Hoagland, et al. Expires January 8, 2009 [Page 19] Internet-Draft Tunneling Security Concerns 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. Hoagland, et al. Expires January 8, 2009 [Page 20]