Internet Engineering Task Force (IETF) J. Abley Request for Comments: 6629 ICANN Category: Informational M. Bagnulo ISSN: 2070-1721 A. Garcia-Martinez UC3M June 2012 Considerations on the Application of the Level 3 Multihoming Shim Protocol for IPv6 (Shim6) Abstract This document discusses some considerations on the applicability of the level 3 multihoming Shim protocol for IPv6 (Shim6) and associated support protocols and mechanisms to provide site multihoming capabilities in IPv6. Status of This Memo This document is not an Internet Standards Track specification; it is published for informational purposes. This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are a candidate for any level of Internet Standard; see Section 2 of RFC 5741. Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at http://www.rfc-editor.org/info/rfc6629. Copyright Notice Copyright (c) 2012 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
Table of Contents 1. Introduction ....................................................3 2. Deployment Scenarios ............................................4 3. Addresses and Shim6 .............................................6 3.1. Protocol Version (IPv4 vs. IPv6) ...........................6 3.2. Prefix Lengths .............................................7 3.3. Address Generation and Configuration .......................7 3.4. Use of CGA vs. HBA .........................................7 4. Shim6 in Multihomed Nodes .......................................8 5. Shim6 Capabilities .............................................10 5.1. Fault Tolerance ...........................................10 5.1.1. Establishing Communications After an Outage ........10 5.1.2. Short-Lived and Long-Lived Communications ..........11 5.2. Load Balancing ............................................11 5.3. Traffic Engineering .......................................12 6. Application Considerations .....................................12 7. Interaction with Other Protocols and Mechanisms ................13 7.1. Shim6 and Mobile IPv6 .....................................13 7.1.1. Multihomed Home Network ............................14 7.1.2. Shim6 Between the HA and the MN ....................16 7.2. Shim6 and SEND ............................................16 7.3. Shim6, SCTP and MPTCP .....................................17 7.4. Shim6 and NEMO ............................................18 7.5. Shim6 and HIP .............................................18 7.6. Shim6 and Firewalls .......................................19 7.7. Shim6 and NPTv6 ...........................................20 8. Security Considerations ........................................23 8.1. Privacy Considerations ....................................24 9. Contributors ...................................................24 10. Acknowledgements ..............................................24 11. References ....................................................25 11.1. Normative References .....................................25 11.2. Informative References ...................................26
1. Introduction Site multihoming is an arrangement by which a site may use multiple paths to the rest of the Internet to provide better reliability for traffic passing in and out of the site than would be possible with a single path. Some of the motivations for operators to multihome their network are described in [RFC3582]. In IPv4, site multihoming is achieved by injecting the additional state required to allow session resilience over re-homing events [RFC4116] into the global Internet routing system (sometimes referred to as the Default-Free Zone, or DFZ). There is concern that this approach will not scale [RFC3221] [RFC4984]. Site multihoming in IPv6 can be achieved as in IPv4, thus facing similar scalability concerns. However, IPv6's large address space enables a different solution for site multihoming in IPv6: to assign multiple addresses to each host, one or more from each provider. Deploying site multihoming in this way does not impact the routing system. So such a site multihoming strategy may be extended to a large number of sites, and may be applied to small sites that would not be eligible for site multihoming based on the injection of routes to Provider Independent (PI) prefixes. A drawback of this multihoming approach is that it does not provide transport-layer stability across re-homing events. Shim6 provides layer-3 support for making re-homing events transparent to the transport layer by means of a shim approach. Once a Shim6 session has been established, the failure detection mechanism defined for Shim6 allows finding new, valid locator combinations in case of failure and using these locators to continue the communication. However, Shim6 does not provide failure protection to the communication establishment, so if a host within a multihomed site attempts to establish a communication with a remote host and selects an address that corresponds to a failed transit path, the communication will fail. State information relating to the multihoming of two endpoints exchanging unicast traffic is retained on the endpoints themselves, rather than in the network. Communications between Shim6-capable hosts and Shim6-incapable hosts proceed as normal, but without the benefit of transport-layer stability. The Shim6 approach is thought to have better scaling properties with respect to the state held in the DFZ than the PI approach. In order to successfully deploy Shim6 in a multihomed site, additional mechanisms may be required to solve issues, such as selecting the source address appropriate to the destination and to the outgoing provider, or to allow the network manager to perform traffic engineering. Such problems are not specific to Shim6, but are relevant to the hosts of any site that is connected to multiple
transit providers, and that receives an IPv6 prefix from each of the providers [RFC5220]. Some of these mechanisms are not defined today. However, note that once a Shim6 session has been established, Shim6 reduces the impact of these problems, because if a working path exists, Shim6 will find it. This note describes the applicability of the Level 3 multihoming (hereafter Shim6) protocol defined in [RFC5533] and the failure detection mechanisms defined in [RFC5534]. The terminology used in this document, including terms like locator and Upper-Layer Identifier (ULID), is defined in [RFC5533]. 2. Deployment Scenarios The goal of the Shim6 protocol is to support locator agility in established communications; different layer-3 endpoint addresses may be used to exchange packets belonging to the same transport-layer session, all the time presenting a consistent identifier pair to upper-layer protocols. In order to be useful, the Shim6 protocol requires that at least one of the peers have more than one address that could be used on the wire (as locators). In the event of communications failure between an active pair of addresses, the Shim6 protocol attempts to reestablish communication by trying different combinations of locators. While other multi-addressing scenarios are not precluded, the scenario in which the Shim6 protocol is expected to operate is that of a multihomed site that is connected to multiple transit providers, and that receives an IPv6 prefix from each of them. This configuration is intended to provide protection for the end-site in the event of a failure in some subset of the available transit providers, without requiring the end-site to acquire PI address space or requiring any particular cooperation between the transit providers.
,------------------------------------. ,----------------. | Rest of the Internet +-------+ Remote Host R | `--+-----------+------------------+--' `----------------' | | | LR ... LR[m] ,---+----. ,---+----. ,----+---. | ISP | | ISP | ...... | ISP[n] | `---+----' `---+----' `----+---' | | | ,---+-----------+------------------+---. | Multi-Homed Site S assigned | | prefixes P, P, ..., P[n] | | | | ,--------. L = P:iid, | | | Host H | L = P:iid, ... | | `--------' L[n] = P[n]:iid[n] | `--------------------------------------' Figure 1 In the scenario illustrated in Figure 1, host H communicates with some remote host R. Each of the addresses L[i] configured on host H in the multihomed site S can be reached through provider ISP[i] only, since ISP[i] is solely responsible for advertising a covering prefix for P[i] to the rest of the Internet. The use of locator L[i] on H hence causes inbound traffic towards H to be routed through ISP[i]. Changing the locator from L[i] to L[j] will have the effect of re-routing inbound traffic to H from ISP[i] to ISP[j]. This is the central mechanism by which the Shim6 protocol aims to provide multihoming functionality: by changing locators, host H can change the upstream ISP used to route inbound packets towards itself. Regarding the outbound traffic to H, the path taken in this case depends on both the actual locator LR[j] used for R, and the administrative exit selection policy of site S. As discussed in Section 4, the site should deliver outgoing packets that have a source address derived from the prefix of ISP[i] to that particular provider, in order to prevent those packets from being filtered due to ingress filtering [RFC2827] being applied by the providers. It is worth noting that in a scenario such as the one depicted in Figure 1, the paths followed by inbound and outbound traffic are determined, to a large extent, by the locators in use for the communication. This is not a particular issue of Shim6, but it is common to any deployment in which hosts are configured with addresses received from different providers. Traffic Engineering in such sites will likely involve proper configuration of address selection policies in the hosts, by means of mechanisms such as the ones discussed in Section 4.
The Shim6 protocol has other potential applications beyond site multihoming. For example, since Shim6 is a host-based protocol, it can also be used to support host multihoming. In this case, a failure in communication between a multihomed host and some other remote host might be repaired by selecting a locator associated with a different interface. To allow nodes to benefit from the capabilities provided by Shim6, (discussed in Section 5) such as fault tolerance, nodes should be configured to initiate a Shim6 session with any peer node if they have multiple locators to use. Note that this configuration can be performed transparently to the applications, in the sense that applications do not need to be aware of the Shim6 functionality provided by the node; in particular, nodes are not forced to use the Shim6 API [RFC6316] to benefit from Shim6. The Shim6 session should be created after the two nodes have been communicating for some time, i.e., using the deferred context establishment facility provided by Shim6. Otherwise, the cost of the Shim6 4-way handshake used for establishing the session may exceed the benefits provided for short- lived communications (see Section 5.1.2). More advanced node configuration may involve configuring different delays for initiating the session for different applications, for example, based on a per- port configuration. Nodes being able to use a single locator for the communication should not initiate the creation of a Shim6 context, but should participate if another node initiates it. Note that Shim6-aware applications can override this behavior by means of the Shim6 API [RFC6316]. 3. Addresses and Shim6 3.1. Protocol Version (IPv4 vs. IPv6) The Shim6 protocol is defined only for IPv6. While some Shim6-like approaches have been suggested to support IPv4 addresses as a locator [SHIM6-ESD], it is not clear if such extensions are feasible. The Shim6 protocol, as specified for IPv6, incorporates cryptographic elements in the construction of locators (see [RFC3972] and [RFC5535]). Since IPv4 addresses are insufficiently large to contain addresses constructed in this fashion, direct use of Shim6 with IPv4 addresses is not possible. In addition, there are other factors to take into account when considering the support of IPv4 addresses, in particular, IPv4 locators. Using multiple IPv4 addresses in a single host in order to support the Shim6 style of multihoming would result in an increased IPv4 address consumption, which would be problematic considering that the IPv4 address space has been exhausted. Besides, Shim6 may
experience additional problems if locators become translated on the wire. Address translation is more likely to involve IPv4 addresses. IPv4 addresses can be translated to other IPv4 addresses (for example, a private IPv4 address into a public IPv4 address and vice versa) or to/from IPv6 addresses (for example, as defined by NAT64 [RFC6146]). When address translation occurs, a locator exchanged by Shim6 could be different from the address needed to reach the corresponding host, either because the translated version of the locator exchanged by Shim6 is not known or because the translation state no longer exists in the translator device. Besides, the translated locators will not be verifiable with the current Cryptographically Generated Address (CGA) and Hash-Based Address (HBA) verification mechanisms, which protect the locators as seen by the node for which they are configured. 3.2. Prefix Lengths The Shim6 protocol does not assume that all the prefixes assigned to the multihomed site have the same prefix length. However, the use of CGA [RFC3972] and HBA [RFC5535] involves encoding information in the lower 64 bits of the locators. This imposes the requirement that all interface addresses should be able to accommodate 64-bit interface identifiers on Shim6-capable hosts. Note that this is imposed by RFC 4291 [RFC4291]. 3.3. Address Generation and Configuration The security of the Shim6 protocol is based on the use of CGA and HBA addresses. The CGA and HBA generation process can use the information provided by the stateless auto-configuration mechanism defined in [RFC4862] with the additional considerations presented in [RFC3972] and [RFC5535]. Stateful address auto-configuration using DHCP [RFC3315] is not currently supported, because there is no defined mechanism to convey the CGA Parameter Data Structure and other relevant information from the DHCP server to the host. An analysis of the possible interactions between DHCPv6 and CGA can be found in [DHCPv6-CGA]. 3.4. Use of CGA vs. HBA The choice between CGA and HBA is a trade-off between flexibility and performance.
The use of HBA is more efficient in the sense that addresses require less computation than CGA, involving only hash operations for both the generation and the verification of locator sets. However, the locators of an HBA set are determined during the generation process and cannot be subsequently changed; the addition of new locators to that initial set is not supported. Therefore, a node using an HBA as a ULID for a Shim6 session can only use the locators associated to that HBA for the considered Shim6 session. If the node wants to use a new set of locators, it has to generate a new HBA including the prefixes of the new locators (which will result with very high probability in different addresses to those of the previous set). New sessions initiated with a ULID belonging to the new HBA address set could use the new locators. The use of CGA is more computationally expensive, involving public- key cryptography in the verification of locator sets. However, CGAs are more flexible in the sense that they support the dynamic modification of locator sets. Therefore, CGAs are well suited to support dynamic environments such as mobile hosts, where the locator set must be changed frequently. HBAs are better suited for sites where the prefix set remains relatively stable. It should be noted that since HBAs are defined as a CGA extension, it is possible to generate an address that incorporates the strengths of both HBA and CGA, i.e., that a single address can be used as an HBA, enabling computationally-cheap validation amongst a fixed set of addresses, and also as a CGA, enabling dynamic manipulation of the locator set. For additional details, see [RFC5535]. 4. Shim6 in Multihomed Nodes Shim6 multihomed nodes are likely to experience problems related to the attachment to different provision domains. Note that these problems are not specific to Shim6. [RFC6418] discusses the problems associated with nodes with multiple interfaces, which may involve difficulties in o managing the configuration associated with different providers. o finding the appropriate DNS server to resolve a query and to match DNS answers to providers. o routing the packets to the right provider. o selecting the source address appropriate to the destination and to the outgoing provider.
o performing session management appropriately. Some of these problems may also arise in single-interface hosts connected to multiple networks, for example, in configurations in which a customer network receives multiple Provider Aggregatable prefixes. These problems are relevant to other solutions supporting multihoming, such as Stream Control Transmission Protocol (SCTP) [RFC4960], Multipath TCP (MPTCP) [RFC6182], or Host Identity Protocol (HIP) [RFC4423]. Note also that single-homed nodes implementing Shim6 to improve communications with other nodes having multiple addresses will not experience these problems. The compatibility of Shim6 with configurations or mechanisms developed to solve any multihoming problem has to be carefully considered on a case-by-case basis. However, the interaction of Shim6 with some of the solutions discussed in [IPv6NAT] is commented on in the next paragraphs. In order to configure source and destination address selection, tools such as DHCPv6 can be used to disseminate an [RFC3484] policy table to a host [6MAN]. The impact to Shim6 using this solution, which disseminates the policy table to the hosts, is the following: Shim6 selects the ULID pair to use in communication according to the mechanism described in [RFC3484]. In case different locator pairs need to be explored, nodes also use the rules defined by [RFC3484] to identify valid pairs, and to establish an order among them, as described in [RFC5534]. When a locator has been selected by a host to be used as the source address for a Shim6 session, Shim6 has no means to enforce an appropriate path for that source address in either the host or the network. For IPv6 nodes, the next-hop router to use for a given set of destinations can be configured through Extensions to Router Advertisements, through Default Router Preference and More-Specific Routes [RFC4191], the use of a DHCPv6 option, or the use of a routing protocol. It is also possible to rely on routers that consider source addresses in their forwarding decisions in addition to the usual destination-based forwarding. All these solutions are compatible with Shim6 operation. Note that an improper matching of source address and egress provider may result in packets being dropped if the provider performs ingress filtering [RFC2827], i.e., dropping packets that come from customer networks with source addresses not belonging to the prefix assigned to them to prevent address spoofing.
For some particular configurations, i.e., for a walled-garden or closed service, the node may need to identify the most appropriate DNS server to resolve a particular query. For an analysis of this problem, the reader is referred to [IPv6NAT]. Finally, note that Shim6 is built to handle communication problems, so it may recover from the misconfiguration (or lack) of some of the mechanisms used to handle the aforementioned problems. For example, if any notification is received from the router dropping the packets with legitimate source addresses as a result of ingress filtering, the affected locator could be associated with a low preference (or not be used at all). But even if such a notification is not received, or not processed by the Shim6 layer, the defective source address or next-hop selection will be treated as a communication failure. Therefore, Shim6 re-homing could finally select a working path in which packets are not filtered, if this path exists. This behavior results from the powerful end-to-end resilience properties exhibited by the REAchability Protocol (REAP) [RFC5534]. 5. Shim6 Capabilities 5.1. Fault Tolerance 5.1.1. Establishing Communications After an Outage If a host within a multihomed site attempts to establish a communication with a remote host and selects a locator that corresponds to a failed transit path, bidirectional communication between the two hosts will not succeed. In order to establish a new communication, the initiating host must try different combinations of (source, destination) locator pairs until it finds a pair that works. The mechanism for this default address selection is described in [RFC3484]. As a result of the use of this mechanism, some failures may not be recovered, even if a valid alternative path exists between two communicating hosts. For example, assuming a failure in ISP (see Figure 1), and host H initiating a communication with host R, the source address selection algorithm described in [RFC3484] may result in the selection of the source address corresponding to ISP for every destination address being tried by the application. However, note that if R is the node initiating the communication, it will find a valid path provided that the application at R tries every available address for H. Since a Shim6 context is normally established between two hosts only after initial communication has been set up, there is no opportunity for Shim6 to participate in the discovery of a suitable, initial (source, destination) locator pair. The same consideration holds for referrals, as described in Section 6.
5.1.2. Short-Lived and Long-Lived Communications The Shim6 context establishment operation requires a 4-way packet exchange, and involves some overhead on the participating hosts in memory and CPU. For short-lived communications between two hosts, the benefit of establishing a Shim6 context might not exceed the cost, perhaps because the protocols concerned are fault tolerant and can arrange their own recovery (e.g., DNS) or because the frequency of re-homing events is sufficiently low that the probability of such a failure occurring during a short-lived exchange is not considered significant. It is anticipated that the exchange of Shim6 context will provide the most benefit for exchanges between hosts that are long-lived. For this reason, the default behavior of Shim6-capable hosts is expected to employ deferred context-establishment. Deferred context setup ensures that session-establishment time will not be increased by the use of Shim6. This default behavior can be overridden by applications that prefer immediate context establishment, regardless of transaction longevity, by using [RFC6316]. Note that all the above considerations refer to the lifetime of the interaction between the peers, and not the lifetime of a particular connection (e.g., TCP connection). In other words, the Shim6 context is established between ULID pairs and it affects all the communication between these ULIDs. So, two nodes with multiple short-lived communications using the same ULID pair would benefit as much from the Shim6 features as two nodes having a single long-lived communication. One example of such a scenario would be a web-client software downloading web content from a server over multiple TCP connections. Each TCP connection is short-lived, but the communication/contact between the two ULID could be long-lived. 5.2. Load Balancing The Shim6 protocol does not support load balancing within a single context: all packets associated with a particular context are exchanged using a single locator pair per direction, with the exception of forked contexts, which are created upon explicit requests from the upper-layer protocol. It may be possible to extend the Shim6 protocol to use multiple locator pairs in a single context, but the impact of such an extension on upper-layer protocols (e.g., on TCP congestion control) should be considered carefully.
When many contexts are considered together in aggregation, e.g., on a single host that participates in many simultaneous contexts or in a site full of hosts, some degree of load sharing should occur naturally due to the selection of different locator pairs in each context. However, there is no mechanism defined to ensure that this natural load sharing is arranged to provide a statistical balance between transit providers. Note that the use of transport-layer solutions enhanced with mechanisms to allow the use of multiple paths for a transport session are more amenable for achieving load-balancing. One such solution is MPTCP [RFC6182]. 5.3. Traffic Engineering For sites with prefixes obtained from different providers, the paths followed by inbound and outbound traffic are determined to a large extent by the locators selected for each communication. This is not a particular issue of Shim6, but it is common to any deployment in which hosts are configured with addresses received from different providers. Traffic engineering in such sites will likely involve proper configuration of the address selection policies defined by [RFC3484]. The Shim6 protocol provides some lightweight traffic engineering capabilities in the form of the Locator Preferences option, which allows a host to inform a remote host of local preferences for locator selection. In this way, the host can influence the incoming path for the communication. This mechanism is only available after a Shim6 context has been established, and it is a host-based capability rather than a site-based capability. There is no defined mechanism that would allow use of the Locator Preferences option amongst a site full of hosts to be managed centrally by the administrator of the site. 6. Application Considerations Shim6 provides multihoming support without forcing changes in the applications running on the host. The fact that an address has been generated according to the CGA or HBA specification does not require any specific action from the application, e.g., it can obtain remote CGA or HBA addresses as a result of a getaddrinfo() call to trigger a DNS Request. The storage of CGA or HBA addresses in DNS does not require any modification to this protocol, since they are recorded using AAAA records. Moreover, neither the ULID/locator management [RFC5533] nor the failure detection and recovery [RFC5534] functions require application awareness.
However, a specific API [RFC6316] has been developed for those applications that might require additional capabilities in ULID/ locator management, such as the locator pair in use for a given context, or the set of local or remote locators available for it. This API can also be used to disable Shim6 operation when required. It is worth noting that callbacks can benefit naturally from Shim6 support. In a callback, an application in B retrieves IP_A, the IP address of a peer A, and B uses IP_A to establish a new communication with A. As long as the address exchanged, IP_A, is the ULID for the initial communication between A and B, and B uses the same address as in the initial communication, and this initial communication is alive (or the context has not been deleted), the new communication could use the locators exchanged by Shim6 for the first communication. In this case, communication could proceed even if the ULID of A is not reachable. However, Shim6 does not provide specific protection to current applications when they use referrals. A referral is the exchange of the IP address IP_A of a party A by party B to party C, so that party C could use IP_A to communicate with party A. In a normal case, the ULID IP_A would be the only information sent by B to C as a referral. But if IP_A is no longer valid as the locator in A, C could have trouble establishing a communication with A. Increased failure protection for referrals could be obtained if B exchanged the whole list of alternative locators of A; although, in this case the application protocol should be modified. Note that B could send to C the current locator of A, instead of the ULID of A, as a way of using the most recent reachability information about A. While in this case no modification of the application protocol is required, some concerns arise: host A may not accept one of its locators as ULID for initiating a communication, and if a CGA are used, the locator may not be a CGA so a Shim6 context among A and C could not be created. 7. Interaction with Other Protocols and Mechanisms In this section we discuss the interaction between Shim6 and other protocols and mechanisms. Before starting the discussion, it is worth noting that at the time of this writing, there is a lack of experience with the combination of Shim6 and these protocols and mechanisms. Therefore, the conclusions stated should be reviewed as real experience is gained in the use of Shim6. 7.1. Shim6 and Mobile IPv6 Here, we consider some scenarios in which the Shim6 protocol and the Mobile IPv6 (MIPv6) protocol [RFC6275] might be used simultaneously.
7.1.1. Multihomed Home Network In this case, the Home Network of the Mobile Node (MN) is multihomed. This implies the availability of multiple Home Network prefixes, resulting in multiple Home Addresses (HoAs) for each MN. Since the MN is a node within a multihomed site, it seems reasonable to expect that the MN should be able to benefit from the multihoming capabilities provided by the Shim6 protocol. Moreover, the MN needs to be able to obtain the multihoming benefits, even when it is roaming away from the Home Network: if the MN is away from the Home Network while the Home Network suffers a failure in a transit path, the MN should be able to continue communicating using alternate paths to reach the Home Network. The resulting scenario is the following: +------------------------------------+ | Internet | +------------------------------------+ | | +----+ +----+ |ISP1| |ISP2| +----+ +----+ | | +------------------------------------+ | Multihomed Home Network | | Prefixes: P1 and P2 | | | | Home Agent | | // | +------------------//----------------+ // // +-----+ | MN | HoA1, HoA2 +-----+ Figure 2 So, in this configuration, the Shim6 protocol is used to provide multiple communication paths to all the nodes within the multihomed sites (including the mobile nodes), and the MIPv6 protocol is used to support mobility of the multihomed site's mobile nodes.
The proposed protocol architecture would be the following: +--------------+ | Application | +--------------+ | Transport | +--------------+ | IP | | +----------+ | | | IPSec | | | +----------+<--ULIDs | | Shim6 | | | +----------+<--HoAs | | MIPv6 | | | +----------+<--CoAs | | +--------------+ Figure 3 In this architecture, the upper-layer protocols and IPSec would use ULIDs of the Shim6 protocol (see Section 16.1 in [RFC5533] for more detail on the interaction between Shim6 and IPsec). Only the HoAs will be presented by the upper layers to the Shim6 layer as potential ULIDs. Two Shim6 entities will exchange their own available HoAs as locators. Therefore, Shim6 provides failover between different HoAs and allows preservation of established communications when an outage affects the path through the ISP that has delegated the HoA used for initiating the communication (similar to the case of a host within a multihomed site). The Care-of Addresses (CoAs) are not presented to the Shim6 layer and are not included in the local locator set in this case. The CoAs are managed by the MIPv6 layer, which binds each HoA to a CoA. For example, if a single CoA, CoA1, is available for the MN in the foreign link to which it is attached, every HoA should have a bind to CoA1. So, in this case, the upper-layer protocols select a ULID pair for the communication. The Shim6 protocol translates the ULID pair to an alternative locator in case that is needed. Both the ULIDs and the alternative locators are HoAs. Next, the MIPv6 layer maps the selected HoA to the corresponding CoA, which is the actual address included in the wire. The Shim6 context is established between the MN and the Correspondent Node (CN), and it would allow the communication to use all the available HoAs to provide fault tolerance. The MIPv6 protocol is used between the MN and the Home Agent (HA) in the case of the bidirectional tunnel mode, and between the MN and the CN in case of the Route Optimization (RO) mode.
7.1.2. Shim6 between the HA and the MN Another scenario where a Shim6-MIPv6 interaction may be useful is the case where a Shim6 context is established between the MN and the HA in order to provide fault tolerance capabilities to the bidirectional tunnel between them. Consider the case where the HA has multiple addresses (whether because the Home Network is multihomed or because the HA has multiple interfaces) and/or the MN has multiple addresses (whether because the visited network is multihomed or because the MN has multiple interfaces). In this case, if a failure affects the address pair that is being used to run the tunnel between the MN and HA, additional mechanisms need to be used to preserve the communication. One possibility would be to use MIPv6 capabilities, by simply changing the CoA used as the tunnel endpoint. However, MIPv6 lacks the failure detection mechanisms that would allow the MN and/or the HA to detect the failure and trigger the usage of an alternative address. Shim6 provides such a failure detection protocol, so one possibility would be re-using the failure detection function from the Shim6 failure detection protocol in MIPv6. In this case, the Shim6 protocol wouldn't be used to create Shim6 context and provide fault tolerance, but just its failure detection functionality would be re-used. The other possibility would be to use the Shim6 protocol to create a Shim6 context between the HA and the MN, so that the Shim6 detects any failure and re-homes the communication in a transparent fashion to MIPv6. In this case, the Shim6 protocol would be associated with the tunnel interface. 7.2. Shim6 and SEND Secure Neighbor Discovery (SEND) [RFC3971] uses CGAs to prove address ownership for Neighbor Discovery [RFC4861]. The Shim6 protocol can use either CGAs or HBAs to protect locator sets included in Shim6 contexts. It is expected that some hosts will need to participate in both SEND and Shim6 simultaneously. In the case that both the SEND and Shim6 protocols are using the CGA technique to generate addresses, there is no conflict; the host will generate addresses for both purposes as CGAs, and since it will be in control of the associated private key, the same CGA can be used for the different protocols.
In the case that a Shim6-capable host is using HBAs to protect its locator sets, the host will need to generate an address that is both a valid CGA and a valid HBA, as defined in [RFC5535]. In this case, the CGA Parameter Data Structure containing a valid public key and the Multi-Prefix extension are included as inputs to the hash function. 7.3. Shim6, SCTP, and MPTCP Both the SCTP [RFC4960] and MPTCP [RFC6182] protocols provide a reliable, stream-based communications channel between two hosts that provides a superset of the capabilities of TCP. One notable feature of these two protocols is that they allow the exchange of endpoint addresses between hosts in order to recover from the failure of a particular endpoint pair, or to benefit from multipath communication in the MPTCP case, in a manner that is conceptually similar to locator selection in Shim6. SCTP and MPTCP are transport-layer protocols, higher in the protocol stack than Shim6; hence, there is no fundamental incompatibility that would prevent a Shim6-capable host from communicating using SCTP or MPTCP. However, since either SCTP or MPTCP, and Shim6 aim to exchange addressing information between hosts in order to meet the same generic goal, it is possible that their simultaneous use might result in unexpected behavior, e.g., lead to race conditions. The capabilities of these transport protocols with respect to path maintenance of a reliable, connection-oriented stream protocol are more extensive than the more general layer-3 locator agility provided by Shim6. Therefore, it is recommended that Shim6 not be used for SCTP or MPTCP sessions, and that path maintenance be provided solely by SCTP or MPTCP. There are at least two ways to enforce this behavior. One option is to make the stack, and in particular the Shim6 sublayer, aware of the use of SCTP or MPTCP, and in this case refrain from creating a Shim6 context. The other option is that the upper transport layer indicates, using a Shim6-capable API like the one proposed in [RFC6316], that no Shim6 context must be created for this particular communication. In general, the issues described here may also arise for protocols that handle different addresses for two communicating nodes at a higher level than the network layer to improve reliability, performance, congestion control, etc.
7.4. Shim6 and NEMO The Network Mobility (NEMO) [RFC3963] protocol extensions to MIPv6 allow a Mobile Network to communicate through a bidirectional tunnel via a Mobile Router (MR) to a NEMO-compliant HA located in a Home Network. If either or both the MR or HA are multihomed, then an established Shim6 context preserves the integrity of the bidirectional tunnel between them in the event that a transit failure occurs in the connecting path. Once the tunnel between MR and HA is established, hosts within the Mobile Network that are Shim6-capable can establish contexts with remote hosts in order to receive the same multihoming benefits as any host located within the Home Network. 7.5. Shim6 and HIP Shim6 and HIP [RFC4423] are architecturally similar in the sense that both solutions allow two hosts to use different locators to support communications between stable ULIDs. The signaling exchange to establish the demultiplexing context on the hosts is very similar for both protocols. However, there are a few key differences. First, Shim6 avoids defining a new namespace for ULIDs, preferring instead to use a routable locator as a ULID, while HIP uses public keys and hashes thereof as ULIDs. The use of a routable locator as the ULID better supports deferred context establishment, application callbacks, and application referrals, and avoids management and resolution costs of a new namespace, but requires additional security mechanisms to securely bind the ULID with the locators. Second, Shim6 uses an explicit context header on data packets for which the ULIDs differ from the locators in use (this header is only needed after a failure/re-homing event occurs), while HIP may compress this context-tag function into the Encapsulating Security Payload (ESP) Security Parameter Index (SPI) field [RFC5201]. Third, HIP as presently defined requires the use of public-key operations in its signaling exchange and ESP encryption in the data plane, while the use of Shim6 requires neither (if only HBA addresses are used). By default, HIP provides data protection, while this is a non-goal for Shim6. Shim6 aimed to provide a solution to a specific problem, multihoming, which minimizes deployment disruption, while HIP is considered more of an experimental approach intended to solve several more general problems (mobility, multihoming, and loss of end-to-end addressing transparency) through an explicit identifier/locator split.
Communicating hosts that are willing to run HIP (perhaps extended with Shim6's failure detection protocol) likely have no reason to also run Shim6. In this sense, HIP may be viewed as a possible long- term evolution or extension of the Shim6 architecture, or one possible implementation of the Extended Shim6 Design (ESD) [SHIM6-ESD]. 7.6. Shim6 and Firewalls The ability of Shim6 to divert the communication to different paths may be affected by certain firewall configurations. For example, consider a deployment in which one of the peers of a Shim6 session is protected by a firewall (i.e., all the paths to the locators of that peer traverse the firewall). The firewall implements the Simple Security model [RFC4864], in which incoming packets are checked against a state resulting from outgoing traffic, either associated with the locator of the internal node ('endpoint independent filtering') or to both the locators of the internal and external nodes ('address dependent filtering' or 'address and port dependent filtering'). If the external node changes the locator associated with the internal node, the packet will be discarded by the firewall. In addition, if the firewall implements 'address dependent filtering' or 'address and port dependent filtering', any change by the external node in the locator used to identify itself will also result in the packet being discarded by the firewall. This issue could be mitigated by making the firewalls aware of the different locators that could be associated with a given communication. If the firewall is implemented in the communication node itself, the firewall could inspect the Shim6 control packet exchange to obtain this information, or the Shim6 software module could explicitly inform the firewall software module. For firewalls located outside the node, the Shim6 control packet exchange can be used to associate the alternate locators to the communication state, although it may not work for topologies in which both directions for the communication do not traverse the firewall, or in which the firewall is not traversed after a locator change. The detail of any of such mechanisms is out of the scope of this document. However, note that a failure in using the alternative locators does not impact the communication between the nodes as long as the path between them defined by the initial locator pair remains available. In this case, data packets flow between the communicating nodes as for any non-Shim6 communication.
7.7. Shim6 and NPTv6 Address translation techniques such as Network Prefix Translation (NPTv6) [RFC6296] may be used until workable solutions to avoid renumbering or facilitate multihoming are developed [RFC5902]. We now consider the impact of NPTv6 in Shim6 operation. Some of the considerations stated in this section may also be applicable to other types of IPv6 NAT. The main purpose of Shim6 is to provide locator agility below transport protocols. To prevent the risk of redirection attacks by abusing the locator exchange facilities provided by Shim6, the protocol is built upon the cryptographic properties of CGA and HBA addresses. When the CGA address of a node is used as the local ULID, the locators configured in the node can be signed with the private key associated with the CGA. A peer receiving a Shim6 message performs a hash of the CGA Parameter Data Structure information received, including a public key, to assure that this key is bound to the CGA address, and then checks the signature protecting the locators. When an HBA address of a node is used as the local ULID, the HBA address securely chains the ULID and other locators of the node by means of a hash. For both the CGA and the HBA, the locators can be exchanged at the four-way handshake used to establish the Shim6 context, or once the context has been established by means of an Update Request message. When a node behind an NPTv6 communicates, the NAT device translates the address assigned to this internal node to an address of its address pool. This operation results in a mismatch between the address seen by external hosts and the address configured in the internal node, which is the locator that would be conveyed in a Shim6 locator exchange and is also the address for which the security defined in the CGA and HBA specifications are provided. Then, the validation processes performed by an external node may prevent the creation of the Shim6 context, or may allow the context to be created but render the alternative locator of the internal host unusable. However, note that the failure in creating a Shim6 context, or in using the alternative locators, does not impact the communication between the nodes as long as the path between them defined by the initial locator pair remains available. Data packets flow between the communicating nodes as for any non-Shim6 communication. Not creating the Shim6 context, or not being able to convey the local locators to the peer node, affect the added value provided by Shim6, i.e., the ability to preserve the communication in case any of the locators fail. Therefore, using Shim6 with NPTv6 does not provide less functionality than using IPv6 in the same scenario.
We now illustrate some cases that may occur when combining Shim6 and NPTv6. The following discussion does not aim to be exhaustive in the cases that may arise, but just aims to provide some examples of possible situations. We assume a scenario in which host A is located behind a NPTv6 device for its locator IP_A1, but it is connected to the public IPv6 Internet for its locator IP_A2. Once translated, locator IP_A1 appears to external nodes as IP_T. Node A communicates with node B, with public addresses IP_B1 and IP_B2. +-----+ | A | +-----+ IP_A1 | | IP_A2 | | | +-----+ | | +--------+ | | NPTv6 | | +--------+ | IP_T | | | | +--------------------------+ | Internet | +--------------------------+ | | IP_B1 | | IP_B2 +-----+ | B | +-----+ Figure 4 We first discuss some issues related with the four-way handshake used to establish the Shim6 context. When the locator information is included in the Shim6 exchange, either in the I2 or R2 messages, the receiver is required to validate the ULID of the peer node by performing the CGA or HBA address validation procedure. In case the validation fails, the message containing the information is silently discarded. In the scenario depicted in Figure 4, some of the cases that may occur are: o Node A initiates the exchange, with IP_B1 as the destination address and IP_A1 as the source address, which is a CGA. Node A includes IP_A2 as an alternative locator in the I2 message. Node B sees IP_T as the ULID for A, so when it validates the CGA with the information contained in I2, the validation fails because the CGA Parameter Data Structure contains information bound to IP_A1. Therefore, B silently discards the received I2 message. Without
receiving a valid I2 message, B does not create the Shim6 context. Without receiving the R2 message, A also does not create the Shim6 context. However, data communication can proceed as long as the path between IP_A1 and IP_B1 is valid. A similar case occurs if IP_A1 and IP_A2 form an HBA, instead of using CGAs for securing the communication. o Node A initiates the exchange with IP_B1 as the destination address and IP_A2 (its public address) as the source address, which is a CGA. Node A includes IP_A1 as an alternative locator in the I2 message. In this case, B can successfully validate IP_A2 as a CGA. Regarding the validation of IP_A1 as an alternative locator for A, the Shim6 specification [RFC5533] indicates that it should perform this check when the I2 message is received, but it may perform it later on, provided that the check is performed before using it as a locator. In case the validation is performed when I2 is received, the I2 message would be silently discarded, with the same result as for the previous case. In case the validation is performed later, the Shim6 context would be established in both nodes A and B, but B could not send to IP_A1, and packets sent by A from IP_A1 will not be received by B. Note that in this case both IP_B1 and IP_B2 could be used by A and B, as long as the locator for A is IP_A2, so limited locator agility may be achieved. o Node B initiates the exchange with IP_B1 as the source address, and IP_A2 as the destination address, which is a CGA. This case is similar to the previous one, although it is the R2 message sent by A that cannot be validated. While A can create a context with B, B cannot do the same for A. Data communication using IP_B1 and IP_A2 can proceed. However, A may try to use IP_B2 as an alternative locator, but the data packets sent carrying the Shim6 Extension Header will not be associated by B to any established context, so they will be discarded. The same occurs for packets sent by A with IP_A1 as the source address. We can also consider the case in which node A does not exchange its own locators in the Shim6 establishment exchange. For example, a Shim6 context can be established between CGA IP_A2 and IP_B1. B can convey locator IP_B2 in the four-way handshake, and validation will be correctly done by A. Later on, A may send an Update Request message to inform B about its locator IP_A1. Validation for this message will fail in B, and B will send a Shim6 Error message to A. Neither A nor B will use IP_A1 as a locator. However, IP_A2, IP_B1, and IP_B2 can still be used as valid locators for the communication.
Finally, note that modification of the Shim6 control packets by the NPTv6 would not be able to generate a valid signature when a CGA is being used or a Parameter Data Structure binding the translated locator to the other locators of a node when an HBA is being used. Therefore, the same failure cases described before would remain. 8. Security Considerations This section considers the applicability of the Shim6 protocol from a security perspective, i.e., which security features can be expected by applications and users of the Shim6 protocol. First of all, it should be noted that the Shim6 protocol is not a security protocol, unlike HIP, for instance. This means that, as opposed to HIP, it is an explicit non-goal of the Shim6 protocol to provide enhanced security for the communications that use the Shim6 protocol. The goal of the Shim6 protocol design, in terms of security, is not to introduce new vulnerabilities that were not present in the current non-Shim6 enabled communications. In particular, it is an explicit non-goal of Shim6 protocol security to provide protection from on-path attackers. On-path attackers are able to sniff and spoof packets in the current Internet, and they are able to do the same in Shim6 communications (as long as the communication flows through the path on which they are located). Summarizing, the Shim6 protocol does not provide data packet protection from on-path attackers. However, the Shim6 protocol does use several security techniques. The goal of these security measures is to protect the Shim6 signaling protocol from new attacks resulting from the adoption of the Shim6 protocol. In particular, the use of HBA/CGA prevents on-path and off-path attackers from injecting new locators into the locator set of a Shim6 context, thus preventing redirection attacks [RFC4218]. Moreover, the usage of probes before re-homing to a different locator as a destination address prevents flooding attacks from off-path attackers. Note that for nodes using CGA addresses, security depends on the secure handling of the private key associated with the signature and validation of locators. In particular, any address configuration method must assure that the private key remains secret, as discussed in Section 3.3. In addition, the usage of a 4-way handshake for establishing the Shim6 context protects against DoS attacks, so hosts implementing the Shim6 protocol should not be more vulnerable to DoS attacks than regular IPv6 hosts. Finally, many Shim6 signaling messages contain a Context Tag, meaning that only attackers that know the Context Tag can forge them. As a
consequence, only on-path attackers can generate false Shim6 signaling packets for an established context. The impact of these attacks would be limited since they would not be able to add additional locators to the locator set (because of the HBA/CGA protection). In general, the possible attacks have similar effects to the ones that an on-path attacker can launch on any regular IPv6 communication. The residual threats are described in the Security Considerations of the Shim6 protocol specification [RFC5533]. 8.1. Privacy Considerations The Shim6 protocol is designed to provide some basic privacy features. In particular, HBAs are generated in such a way that the different addresses assigned to a host cannot be trivially linked together as belonging to the same host, since there is nothing in common in the addresses themselves. Similar features are provided when the CGA protection is used. This means that it is not trivial to determine that a set of addresses is assigned to a single Shim6 host. However, the Shim6 protocol does exchange the locator set in clear text, and it also uses a fixed Context Tag when using different locators in a given context. This implies that an attacker observing the Shim6 context establishment exchange or seeing different payload packets exchanged through different locators, but with the same Context Tag, can determine the set of addresses assigned to a host. However, this requires that the attacker be located along the path and can capture the Shim6 signaling packets. 9. Contributors The analysis on the interaction between the Shim6 protocol and the other protocols presented in this note benefited from the advice of various people including Tom Henderson, Erik Nordmark, Hesham Soliman, Vijay Devarpalli, John Loughney, and Dave Thaler. 10. Acknowledgements Joe Abley's work was supported, in part, by the US National Science Foundation (research grant SCI-0427144) and DNS-OARC. Marcelo Bagnulo worked on this document while visiting Ericsson Research Laboratory Nomadiclab. Alberto Garcia-Martinez was supported, in part, by the eeCONTET project (TEC2011-29688-C02-02, granted by the Spanish Science and Innovation Ministry).
Shinta Sugimoto reviewed this document and provided comments and text. Brian Carpenter, Jari Arkko, Joel Halpern, Iljitsch van Beijnum, Sam Xia, Carsten Bormann, Dan Wing, Stephen Farrell, and Stewart Bryant reviewed this document and provided comments. 11. References 11.1. Normative References [RFC2827] Ferguson, P. and D. Senie, "Network Ingress Filtering: Defeating Denial of Service Attacks which employ IP Source Address Spoofing", BCP 38, RFC 2827, May 2000. [RFC3315] Droms, R., Ed., Bound, J., Volz, B., Lemon, T., Perkins, C., and M. Carney, "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", RFC 3315, July 2003. [RFC3484] Draves, R., "Default Address Selection for Internet Protocol version 6 (IPv6)", RFC 3484, February 2003. [RFC3963] Devarapalli, V., Wakikawa, R., Petrescu, A., and P. Thubert, "Network Mobility (NEMO) Basic Support Protocol", RFC 3963, January 2005. [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, "SEcure Neighbor Discovery (SEND)", RFC 3971, March 2005. [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", RFC 3972, March 2005. [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing Architecture", RFC 4291, February 2006. [RFC4423] Moskowitz, R. and P. Nikander, "Host Identity Protocol (HIP) Architecture", RFC 4423, May 2006. [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, September 2007. [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless Address Autoconfiguration", RFC 4862, September 2007. [RFC4960] Stewart, R., Ed., "Stream Control Transmission Protocol", RFC 4960, September 2007.
[RFC5201] Moskowitz, R., Nikander, P., Jokela, P., Ed., and T. Henderson, "Host Identity Protocol", RFC 5201, April 2008. [RFC5533] Nordmark, E. and M. Bagnulo, "Shim6: Level 3 Multihoming Shim Protocol for IPv6", RFC 5533, June 2009. [RFC5534] Arkko, J. and I. van Beijnum, "Failure Detection and Locator Pair Exploration Protocol for IPv6 Multihoming", RFC 5534, June 2009. [RFC5535] Bagnulo, M., "Hash-Based Addresses (HBA)", RFC 5535, June 2009. [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful NAT64: Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers", RFC 6146, April 2011. [RFC6182] Ford, A., Raiciu, C., Handley, M., Barre, S., and J. Iyengar, "Architectural Guidelines for Multipath TCP Development", RFC 6182, March 2011. [RFC6275] Perkins, C., Ed., Johnson, D., and J. Arkko, "Mobility Support in IPv6", RFC 6275, July 2011. [RFC6316] Komu, M., Bagnulo, M., Slavov, K., and S. Sugimoto, Ed., "Sockets Application Program Interface (API) for Multihoming Shim", RFC 6316, July 2011. 11.2. Informative References [6MAN] Matsumoto, A., Fujisaki, T., Kato, J., and T. Chown, "Distributing Address Selection Policy using DHCPv6", Work in Progress, February 2012. [DHCPv6-CGA] Jiang, S. and S. Shen, "Analysis of Possible DHCPv6 and CGA Interactions", Work in Progress, March 2012. [IPv6NAT] Matsushima, S., Okimoto, T., Troan, O., Miles, D., and D. Wing, "IPv6 Multihoming without Network Address Translation", Work in Progress, February 2012. [RFC3221] Huston, G., "Commentary on Inter-Domain Routing in the Internet", RFC 3221, December 2001. [RFC3582] Abley, J., Black, B., and V. Gill, "Goals for IPv6 Site- Multihoming Architectures", RFC 3582, August 2003.
[RFC4116] Abley, J., Lindqvist, K., Davies, E., Black, B., and V. Gill, "IPv4 Multihoming Practices and Limitations", RFC 4116, July 2005. [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and More-Specific Routes", RFC 4191, November 2005. [RFC4218] Nordmark, E. and T. Li, "Threats Relating to IPv6 Multihoming Solutions", RFC 4218, October 2005. [RFC4864] Van de Velde, G., Hain, T., Droms, R., Carpenter, B., and E. Klein, "Local Network Protection for IPv6", RFC 4864, May 2007. [RFC4984] Meyer, D., Ed., Zhang, L., Ed., and K. Fall, Ed., "Report from the IAB Workshop on Routing and Addressing", RFC 4984, September 2007. [RFC5220] Matsumoto, A., Fujisaki, T., Hiromi, R., and K. Kanayama, "Problem Statement for Default Address Selection in Multi-Prefix Environments: Operational Issues of RFC 3484 Default Rules", RFC 5220, July 2008. [RFC5902] Thaler, D., Zhang, L., and G. Lebovitz, "IAB Thoughts on IPv6 Network Address Translation", RFC 5902, July 2010. [RFC6296] Wasserman, M. and F. Baker, "IPv6-to-IPv6 Network Prefix Translation", RFC 6296, June 2011. [RFC6418] Blanchet, M. and P. Seite, "Multiple Interfaces and Provisioning Domains Problem Statement", RFC 6418, November 2011. [SHIM6-ESD] Nordmark, E., "Extended Shim6 Design for ID/loc split and Traffic Engineering", Work in Progress, February 2008.
Authors' Addresses Joe Abley ICANN 12025 Waterfront Drive Suite 300 Los Angeles, CA 90094 USA Phone: +1 519 670 9327 EMail: firstname.lastname@example.org Marcelo Bagnulo U. Carlos III de Madrid Av. Universidad 30 Leganes, Madrid 28911 Spain Phone: +34 91 6248814 EMail: email@example.com URI: http://www.it.uc3m.es/ Alberto Garcia-Martinez U. Carlos III de Madrid Av. Universidad 30 Leganes, Madrid 28911 Spain Phone: +34 91 6248782 EMail: firstname.lastname@example.org URI: http://www.it.uc3m.es/