view Side-By-Side changes
SHIM6 WG E. Nordmark Internet-Draft Sun Microsystems Expires:September 5,November 16, 2006 M. Bagnulo UC3MMarch 4,May 15, 2006 Level 3 multihoming shim protocoldraft-ietf-shim6-proto-04.txtdraft-ietf-shim6-proto-05.txt Status of this Memo By submitting this Internet-Draft, each author represents that any applicable patent or other IPR claims of which he or she is aware have been or will be disclosed, and any of which he or she becomes aware will be disclosed, in accordance with Section 6 of BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt. The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. This Internet-Draft will expire onSeptember 5,November 16, 2006. Copyright Notice Copyright (C) The Internet Society (2006). Abstract The SHIM6 protocol is a layer 3 shim for providing locator agility below the transport protocols, so that multihoming can be provided for IPv6 with failover and load sharing properties, without assuming that a multihomed site will have a provider independent IPv6 address prefix which is announced in the global IPv6 routing table. The hosts in a site which has multiple provider allocated IPv6 address prefixes, will use the shim6 protocol specified in this document to Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 1] Internet-Draft Shim6 ProtocolMarchMay 2006 setup state with peer hosts, so that the state can later be used to failover to a different locator pair, should the original one stop working. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 51.11.1. Goals . . . . . . . . . . . . . . . . . . . . . . . . . . 51.21.2. Non-Goals . . . . . . . . . . . . . . . . . . . . . . . . 61.31.3. Locators as Upper-layer Identifiers . . . . . . . . . . . 61.41.4. IP Multicast . . . . . . . . . . . . . . . . . . . . . . 71.51.5. Renumbering Implications . . . . . . . . . . . . . . . . 81.61.6. Placement of the shim . . . . . . . . . . . . . . . . . . 91.71.7. Traffic Engineering . . . . . . . . . . . . . . . . . .11. 10 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 122.12.1. Definitions . . . . . . . . . . . . . . . . . . . . . . . 122.22.2. Notational Conventions . . . . . . . . . . . . . . . . . 15 3. Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . 16 4. Protocol Overview . . . . . . . . . . . . . . . . . . . . . . 174.14.1. Context Tags . . . . . . . . . . . . . . . . . . . . . . 194.24.2. Context Forking . . . . . . . . . . . . . . . . . . . . . 194.34.3. API Extensions . . . . . . . . . . . . . . . . . . . . . 204.44.4. Securing shim6 . . . . . . . . . . . . . . . . . . . . . 204.54.5. Overview of Shim Control Messages . . . . . . . . . . . . 214.64.6. Extension Header Order . . . . . . . . . . . . . . . . . 22 5. Message Formats . . . . . . . . . . . . . . . . . . . . . . . 245.15.1. Common shim6 Message Format . . . . . . . . . . . . . . . 245.25.2. Payload Extension Header Format . . . . . . . . . . . . . 245.35.3. Common Shim6 Control header . . . . . . . . . . . . . . . 255.45.4. I1 Message Format . . . . . . . . . . . . . . . . . . . . 275.55.5. R1 Message Format . . . . . . . . . . . . . . . . . . . . 285.65.6. I2 Message Format . . . . . . . . . . . . . . . . . . .30 5.7. 29 5.7. R2 Message Format . . . . . . . . . . . . . . . . . . . . 315.85.8. R1bis Message Format . . . . . . . . . . . . . . . . . . 335.95.9. I2bis Message Format . . . . . . . . . . . . . . . . . . 345.105.10. Update Request Message Format . . . . . . . . . . . . . . 365.115.11. Update Acknowledgement Message Format . . . . . . . . . . 385.125.12. Keepalive Message Format . . . . . . . . . . . . . . . . 395.135.13. Probe Message Format . . . . . . . . . . . . . . . . . . 395.145.14. Option Formats . . . . . . . . . . . . . . . . . . . . .39 5.14.140 5.14.1. Responder Validator Option Format . . . . . . . . .41 5.14.242 5.14.2. Locator List Option Format . . . . . . . . . . . . . 425.14.35.14.3. Locator Preferences Option Format . . . . . . . . .43 5.14.444 5.14.4. CGA Parameter Data Structure Option Format . . . . .45 5.14.546 5.14.5. CGA Signature Option Format . . . . . . . . . . . . 465.14.65.14.6. ULID Pair Option Format . . . . . . . . . . . . . . 475.14.75.14.7. Forked Instance Identifier Option Format . . . . . . 485.14.8Nordmark & Bagnulo Expires November 16, 2006 [Page 2] Internet-Draft Shim6 Protocol May 2006 5.14.8. Probe Option Format . . . . . . . . . . . . . . . . 48Nordmark & Bagnulo Expires September 5, 2006 [Page 2] Internet-Draft Shim6 Protocol March 2006 5.14.95.14.9. Reachability Option Format . . . . . . . . . . . . .48 5.14.1049 5.14.10. Payload Reception Report Option Format . . . . . .48. 49 6. Conceptual Model of a Host . . . . . . . . . . . . . . . . .49 6.150 6.1. Conceptual Data Structures . . . . . . . . . . . . . . .49 6.250 6.2. Context States . . . . . . . . . . . . . . . . . . . . .5051 7. Establishing ULID-Pair Contexts . . . . . . . . . . . . . .52 7.1 Uniqness. 53 7.1. Uniqueness of Context Tags . . . . . . . . . . . . . . .. 52 7.253 7.2. Locator Verification . . . . . . . . . . . . . . . . . .52 7.353 7.3. Normal context establishment . . . . . . . . . . . . . .53 7.454 7.4. Concurrent context establishment . . . . . . . . . . . .53 7.554 7.5. Context recovery . . . . . . . . . . . . . . . . . . . .55 7.656 7.6. Context confusion . . . . . . . . . . . . . . . . . . .57 7.7. 58 7.7. Sending I1 messages . . . . . . . . . . . . . . . . . .58 7.8. 59 7.8. Retransmitting I1 messages . . . . . . . . . . . . . . .58 7.959 7.9. Receiving I1 messages . . . . . . . . . . . . . . . . .59 7.9.1. 60 7.10. Sending R1 messages . . . . . . . . . . . . . . . . . . . 61 7.10.1. Generating the R1 Validator . . . . . . . . . . . .60 7.1061 7.11. Receiving R1 messages and sending I2 messages . . . . .61 7.11. 62 7.12. Retransmitting I2 messages . . . . . . . . . . . . . . .62 7.1263 7.13. Receiving I2 messages . . . . . . . . . . . . . . . . .62 7.13. 63 7.14. Sending R2 messages . . . . . . . . . . . . . . . . . .64 7.14. 65 7.15. Match for Context Confusion . . . . . . . . . . . . . .64 7.15. 65 7.16. Receiving R2 messages . . . . . . . . . . . . . . . . .65 7.16. 66 7.17. Sending R1bis messages . . . . . . . . . . . . . . . . .66 7.16.167 7.17.1. Generating the R1bis Validator . . . . . . . . . . .66 7.1767 7.18. Receiving R1bis messages and sending I2bis messages . .67 7.18. 68 7.19. Retransmitting I2bis messages . . . . . . . . . . . . .68 7.19. 69 7.20. Receiving I2bis messages and sending R2 messages . . . .6869 8. Handling ICMP Error Messages . . . . . . . . . . . . . . . .7071 9. Teardown of the ULID-Pair Context . . . . . . . . . . . . .72. 73 10. Updating the Peer . . . . . . . . . . . . . . . . . . . .73 10.1. . 74 10.1. Sending Update Request messages . . . . . . . . . . . .73 10.2. 74 10.2. Retransmitting Update Request messages . . . . . . . . .73 10.374 10.3. Newer Information While Retransmitting . . . . . . . . .74 10.475 10.4. Receiving Update Request messages . . . . . . . . . . .74 10.5. 75 10.5. Receiving Update Acknowledgement messages . . . . . . .76. 77 11. Sending ULP Payloads . . . . . . . . . . . . . . . . . . .77 11.1. 78 11.1. Sending ULP Payload after a Switch . . . . . . . . . . .7778 12. Receiving Packets . . . . . . . . . . . . . . . . . . . .79 12.1. . 80 12.1. Receiving Payload Extension Headers . . . . . . . . . .79 12.2. 80 12.2. Receiving Shim Control messages . . . . . . . . . . . .79 12.3. 80 12.3. Context Lookup . . . . . . . . . . . . . . . . . . . . .8081 13. Initial Contact . . . . . . . . . . . . . . . . . . . . .82. . 83 14. Protocol constants . . . . . . . . . . . . . . . . . . . .83. 84 15. Implications Elsewhere . . . . . . . . . . . . . . . . . .84. 85 16. Security Considerations . . . . . . . . . . . . . . . . .86. . 87 17. IANA Considerations . . . . . . . . . . . . . . . . . . .88 18. Acknowledgements . . . . . . . . . . . . . . . . . . . . . 90 A. Open Issues . . . . . . . . .. .. . . . . . . . . . . . . 9189 Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 3] Internet-Draft Shim6 ProtocolMarchMay 2006B. Possible Protocol Extensions . . . . . . . . .18. Acknowledgements . . . . . . .92 C. Change Log. . . . . . . . . . . . . . . 91 Appendix A. Possible Protocol Extensions . . . . . . . . . .94 D.92 Appendix B. Simplified State Machine . . . . . . . . . . . .. . . . . . 97 D.194 Appendix B.1. Simplified State Machine diagram . . . . . . . .. . . . 102 E.100 Appendix C. Context Tag Reuse . . . . . . . . . . . . . . . .. . . . . 103 E.1101 Appendix C.1. Context Recovery . . . . . . . . . . . . . . . .. . . . 103 E.2101 Appendix C.2. Context Confusion . . . . . . . . . . . . . . . .. . . 103 E.3101 Appendix C.3. Three Party Context Confusion . . . . . . . . . .. . . 104 F.102 Appendix D. Design Alternatives . . . . . . . . . . . . . . .. . . . . 105 F.1103 Appendix D.1. Context granularity . . . . . . . . . . . . . . .. . . 105 F.2103 Appendix D.2. Demultiplexing of data packets in shim6 communications .105 F.2.1 Flow-label. . . . . . . . . . . . . . . . 103 Appendix D.2.1. Flow-label . . . . .106 F.2.2 Extension Header. . . . . . . . . . . . . . 104 Appendix D.2.2. Extension Header . . . .108 F.3 Context Loss Detection. . . . . . . . . . . . 106 Appendix D.3. Context Loss Detection . . . . . . . . . .109 F.4 Securing locator sets. . . 107 Appendix D.4. Securing locator sets . . . . . . . . . . . . . .111 F.5109 Appendix D.5. ULID-pair context establishment exchange . . . . 112 Appendix D.6. Updating locator sets . . . . . . .114 F.6 Updating locator sets. . . . . . . 113 Appendix D.7. State Cleanup . . . . . . . . . . . . .115 F.7 State Cleanup. . . . . 113 Appendix E. Change Log . . . . . . . . . . . . . . . .115. . . 116 19. References . . . . . . . . . . . . . . . . . . . . . . . .118 19.1. 120 19.1. Normative References . . . . . . . . . . . . . . . . . .118 19.2120 19.2. Informative References . . . . . . . . . . . . . . . . .118120 Authors' Addresses . . . . . . . . . . . . . . . . . . . . .120. . 122 Intellectual Property and Copyright Statements . . . . . . .121. . 123 Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 4] Internet-Draft Shim6 ProtocolMarchMay 2006 1. Introduction This document describes a layer 3 shim approach and protocol for providing locator agility below the transport protocols, so that multihoming can be provided for IPv6 with failover and load sharing properties [15], without assuming that a multihomed site will have a provider independent IPv6 address which is announced in the global IPv6 routing table. The hosts in a site which has multiple provider allocated IPv6 address prefixes, will use the shim6 protocol specified in this document to setup state with peer hosts, so that the state can later be used to failover to a different locator pair, should the original one stop working. We assume that redirection attacks are prevented using the mechanism specified in HBA [7]. The reachabilitydetectionand failure detection, including how a new working locator pair is discovered after a failure, is specified in a separate documents [8] This document allocates message types and option types for that sub-protocol, and leaves the specification of the message and option formats as well as the protocol behavior to that document.1.11.1. Goals The goals for this approach is to: o Preserve established communications through certain classes of failures, for example, TCP connections and application communications using UDP. o Have minimal impact on upper layer protocols in general and on transport protocols in particular. o Address the security threats in [19] through the combination of the HBA/CGA approach specified in a separate document [7], and techniques described in this document. oDo notNot requireanextra roundtrip up front to setup shim specific state. Instead allow the upper layer traffic (e.g., TCP) to flow as normal and defer the setup of the shim state until some number of packets have been exchanged. o Take advantage of multiple locators/addresses for load spreading so that different sets of communication to a host (e.g., different connections) might use different locators of the host. Note that this might cause load to be spread unevenly, thus we use the term "load spreading" instead of "load balancing". This capability Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 5] Internet-Draft Shim6 ProtocolMarchMay 2006 might enable some forms of traffic engineering, but the details for traffic engineering, including what requirements can be satisfied, are not specified in this document, and form part of a potential extensions to this protocol.1.21.2. Non-Goals The assumption is that the problem we are trying to solve is site multihoming, with the ability to have the set of site locator prefixes change over time due to site renumbering. Further, we assume that such changes to the set of locator prefixes can be relatively slow and managed; slow enough to allow updates to the DNS to propagate. But it is not a goal to try to make communication survive a renumbering event (which causes all the locators of a host to change to a new set of locators). This proposal does not attempt to solve the, perhaps related, problem of host mobility. However, it might turn out that the shim6 protocol can be a useful component for future host mobility solutions, e.g., for route optimization. This proposal also does not try to provide a new network level or transport level identifier namespace separated from the current IP address namespace. Even though such a concept would be useful to ULPs and applications, especially if the management burden for such a name space was negligible and there was an efficient yet secure mechanism to map from identifiers to locators, such a name space isn't necessary (and furthermore doesn't seem to help) to solve the multihoming problem.1.31.3. Locators as Upper-layer Identifiers This approach does not introduce a new identifier name space but instead uses the locator that is selected in the initial contact with the remote peer as the preserved upper-level identifier. While there may be subsequent changes in the selected network level locators over time in response to failures in using the original locator, the upper level protocol stack elements will continue to use this upper level identifier without change. This implies that the ULID selection is performed as today's default address selection as specified in RFC 3484 [12]. Some extensions are needed to RFC 3484 to try different source addresses, whether or not the shim6 protocol is used, as outlined in [13]. Underneath, and transparently, the multihoming shim selects working locator pairs with the initial locator pair being the ULID pair. When communication fails the shim can test and select alternate locators. A subsequent section discusses the issues when the selected ULID is not initially working hence there is a need to switch locators up front. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 6] Internet-Draft Shim6 ProtocolMarchMay 2006front.Using one of the locators as the ULID has certain benefits for applications which have long-lived session state, or performs callbacks or referrals, because both the FQDN and the 128-bit ULID work as handles for the applications. However, using a single 128- bit ULID doesn't provide seamless communication when that locator is unreachable. See [22] for further discussion of the application implications. There has been some discussion of using non-routable locators, such as unique-local addresses [18], as ULIDs in a multihoming solution. While this document doesn't specify all aspects of this, it is believed that the approach can be extended to handle such a case. For example, the protocol already needs to handle ULIDs that are not initially reachable. Thus the same mechanism can handle ULIDs that are permanently unreachable from outside their site. The issue becomes how to make the protocol perform well when the ULID is known a priori to be not reachable (e.g., the ULID is a ULA), for instance, avoiding any timeout and retries in this case. In addition one would need to understand how the ULAs would be entered in the DNS to avoid a performance impact on existing, non-shim6 aware, IPv6 hosts potentially trying to communicate to the (unreachable) ULA.1.41.4. IP Multicast IP Multicast requires that the IP source address field contain a topologically correct locator for interface that is used to send the packet, since IP multicast routing uses both the source address and the destination group to determine where to forward the packet. (This isn't much different than the situation with widely implemented ingress filtering [10] for unicast.) While in theory it would be possible to apply the shim re-mapping of the IP address fields between ULIDs and locators, the fact that all the multicast receivers would need to know the mapping to perform, makes such an approach difficult in practice. Thus it makes sense to have multicast ULPs operate directly on locators and not use the shim. This is quite a natural fit for protocols which use RTP [14], since RTP already has an explicit identifier in the form of the SSRC field in the RTP headers. Thus the actual IP address fields are not important to the application. In summary, IP multicast will not need the shim to remap the IP addresses. This doesn't prevent the receiver of multicast to change its locators, since the receiver is not explicitly identified; theNordmark & Bagnulo Expires September 5, 2006 [Page 7] Internet-Draft Shim6 Protocol March 2006destination address is a multicast address and not the unicast locator of the receiver.1.5Nordmark & Bagnulo Expires November 16, 2006 [Page 7] Internet-Draft Shim6 Protocol May 2006 1.5. Renumbering Implications As stated above, this approach does not try to make communication survive renumbering in the general case. When a host is renumbered, the effect is that one or more locators become invalid, and zero or more locators are added to the host's network interface. This means that the set of locators that is used in the shim will change, which the shim can handle as long as not all the original locators become invalid at the same time. But IP addresses are also used as ULID, and making the communication survive locators becoming invalid can potentially cause some confusion at the upper layers. The fact that a ULID might be used with a different locator over time open up the possibility that communication between two ULIDs might continue to work after one or both of those ULIDs are no longer reachable as locators, for example due to a renumbering event. This opens up the possibility that the ULID (or at least the prefix on which it is based) is reassigned to another site while it is still being used (with another locator) for existing communication.WorstIn the worst case we could end up with two separate hosts using the same ULID while both of them are communicating with the same host. This potential source for confusion can be avoided if we require that any communication using a ULID must be terminated when the ULID becomes invalid (due to the underlying prefix becoming invalid). If that behavior is desired, it can be accomplished by explicitly discarding the shim state when the ULID becomes invalid. The context recovery mechanism will then make the peer aware that the context is gone, and that the ULID is no longer present at the same locator(s). However, terminating the communication might be overkill. Even when an IPv6 prefix is retired and reassigned to some other site, there is a very small probability that another host in that site picks the same 128 bit address (whether using DHCPv6, stateless address autoconfiguration, or picking a random interface ID [11]). Should the identical address be used by another host, then there still wouldn't be a problem until that host attempts to communicate with the same peer host with which the initial user of the IPv6 address was communicating. The protocol as specified in this document does not perform any action when an address becomes invalid. As we gain furtherNordmark & Bagnulo Expires September 5, 2006 [Page 8] Internet-Draft Shim6 Protocol March 2006understanding of the practical impact of renumbering this might change in a future version of the protocol.1.6Nordmark & Bagnulo Expires November 16, 2006 [Page 8] Internet-Draft Shim6 Protocol May 2006 1.6. Placement of the shim ----------------------- | Transport Protocols | ----------------------- ------ ------- -------------- ------------- IP endpoint | AH | | ESP | | Frag/reass | | Dest opts | sub-layer ------ ------- -------------- ------------- --------------------- | shim6 shim layer | --------------------- ------ IP routing | IP | sub-layer ------ Figure 1: Protocol stack The proposal usesana multihoming shim layer within the IP layer, i.e., below the ULPs, as shown in Figure 1, in order to provide ULP independence. The multihoming shim layer behaves as if it is associated with an extension header, which would be placed after any routing-related headers in the packet (such as any hop-by-hop options, or routing header). However, when the locator pair is the ULID pair there is no data that needs to be carried in an extension header, thus none is needed in that case. Layering AH and ESP above the multihoming shim means that IPsec can be made to be unaware of locator changes the same way that transport protocols can be unaware. Thus the IPsec security associations remain stable even though the locators are changing. This means that the IP addresses specified in the selectors should be the ULIDs. Layering the fragmentation header above the multihoming shim makes reassembly robust in the case that there is broken multi-path routing which results in using different paths, hence potentially different source locators, for different fragments. Thus, effectively the multihoming shim layer is placed between the IP endpoint sublayer, which handles fragmentation, reassembly, and IPsec, and the IP routing sublayer, which selects which next hop and interface to use for sending out packets. Applications and upper layer protocols use ULIDs which the shim6Nordmark & Bagnulo Expires September 5, 2006 [Page 9] Internet-Draft Shim6 Protocol March 2006layer will map to/from different locators. The shim6 layer maintains state, called ULID-pair context, per ULID pairs (that is, applies to all ULP connections between the ULID pair) in order to perform this Nordmark & Bagnulo Expires November 16, 2006 [Page 9] Internet-Draft Shim6 Protocol May 2006 mapping. The mapping is performed consistently at the sender and the receiver, thus from the perspective of the upper layer protocols, packets appear to be sent using ULIDs from end to end, even though the packets travel through the network containing locators in the IP address fields, and even though those locators might be changed by the transmitting shim6 layer. The context state in this approach is maintained per remote ULID i.e. approximately per peer host, and not at any finer granularity. In particular, it is independent of the ULPs and any ULP connections. However, the forking capability enables shim-aware ULPs to use more than one locator pair at a time for an single ULID pair. ---------------------------- ---------------------------- | Sender A | | Receiver B | | | | | | ULP | | ULP | | | src ULID(A)=L1(A) | | ^ | | | dst ULID(B)=L1(B) | | | src ULID(A)=L1(A) | | v | | | dst ULID(B)=L1(B) | | multihoming shim | | multihoming shim | | | src L2(A) | | ^ | | | dst L3(B) | | | src L2(A) | | v | | | dst L3(B) | | IP | | IP | ---------------------------- ---------------------------- | ^ ------- cloud with some routers ------- Figure 2: Mapping with changed locators The result of this consistent mapping is that there is no impact on the ULPs. In particular, there is no impact on pseudo-header checksums and connection identification. Conceptually one could view this approach as if both ULIDs and locators are being present in every packet, and with a header compression mechanism applied that removes the need for the ULIDs to be carried in the packets once the compression state has been established. In order for the receiver to recreate a packet with the correct ULIDs there is a need to include some "compression tag" in the data packets. This serves to indicate the correct context to use for decompression when the locator pair in the packet is insufficient to uniquely identify the context.Nordmark & Bagnulo Expires September 5, 2006 [Page 10] Internet-Draft Shim6 Protocol March 2006 1.71.7. Traffic Engineering At the time of this writing it is not clear what requirements for Nordmark & Bagnulo Expires November 16, 2006 [Page 10] Internet-Draft Shim6 Protocol May 2006 traffic engineering make sense for the shim6 protocol, since the requirements must both result in some useful behavior as well as be implementable using a host-to-host locator agility mechanism like shim6. Inherent in a scalable multihoming mechanism that separates locators from identifiers is that each host ends up with multiple locators. This means that at least for initial contact, it is the remote peer that needs to select which peer locator to try first. In the case of shim6 this is performed by applying RFC 3484 address selection. This is quite different than the common case of IPv4 multihoming where the site has a single IP address prefix, since in that case the peer performs no destination address selection. Thus in "single prefix multihoming" the site, and in many cases its upstream ISPs, can use BGP to exert some control of the ingress used to reach the site. This capability can't easily be recreated in "multiple prefix multihoming" such as shim6. The protocol provides a placeholder, in the form of the Locator Preferences option, which can be used by hosts to express priority and weight values for each locator. This is intentionally made identical to the DNS SRV [9] specification of priority and weight, so that DNS SRV records can be used for initial contact and the shim for failover, and they can use the same way to describe the preferences. The format allows adding additional notions of "metrics" over time. But the Locator Preference option is merely a place holder when it comes to providing traffic engineering; in order to use this in a large site there would have to be a mechanism by which the host can find out what preference values to use, either statically (e.g., some new DHCPv6 option) or dynamically. Thus traffic engineering is listed as a possible extension in AppendixB.A. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 11] Internet-Draft Shim6 ProtocolMarchMay 2006 2. Terminology This document uses the terms MUST, SHOULD, RECOMMENDED, MAY, SHOULD NOT and MUST NOT defined in RFC 2119 [1]. The terms defined in RFC 2460 [2] are also used.2.12.1. Definitions This document introduces the following terms: upper layer protocol (ULP) A protocol layer immediately above IP. Examples are transport protocols such as TCP and UDP, control protocols such as ICMP, routing protocols such as OSPF, and internet or lower-layer protocols being "tunneled" over (i.e., encapsulated in) IP such as IPX, AppleTalk, or IP itself. interface A node's attachment to a link. address An IP layer name that contains both topological significance and acts as a unique identifier for an interface. 128 bits. This document only uses the "address" term in the case where it isn't specific whether it is a locator or an identifier. locator An IP layer topological name for an interface or a set of interfaces. 128 bits. The locators are carried in the IP address fields as the packets traverse the network. identifier An IP layer name for an IP layer endpoint. The transport endpoint name is a function of the transport protocol and would typically include the IP identifier plus a port number. NOTE: This proposal does not specify any new form of IP layer identifier, but still separates the identifying and locating properties of the IP addresses. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 12] Internet-Draft Shim6 ProtocolMarchMay 2006 upper-layer identifier (ULID) An IP address which has been selected for communication with a peer to be used by the upper layer protocol. 128 bits. This is used for pseudo-header checksum computation and connection identification in the ULP. Different sets of communication to a host (e.g., different connections) might use different ULIDs in order to enable load spreading. Since the ULID is just one of the IP locators/ addresses of the node, there is no need for a separate name space and allocation mechanisms. address field The source and destination address fields in the IPv6 header. As IPv6 is currently specified this fields carry "addresses". If identifiers and locators are separated these fields will contain locators for packets on the wire. FQDN Fully Qualified Domain Name ULID-pair context The state that the multihoming shim maintains between a pair of Upper-layer identifiers. The context is identified by a context tag for each direction of the communication, and also identified by the pair of ULID and a Forked Instance Identifier (see below). Context tag Each end of the context allocates a context tag for the context. This is used to uniquely associate both received control packets and payload extension headers as belonging to the context. Current locator pair Each end of the context has a current locator pair which is used to send packets to the peer. The two ends might use different current locator pairs though. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 13] Internet-Draft Shim6 ProtocolMarchMay 2006 Default context At the sending end, the shim uses the ULID pair (passed down from the ULP) to find the context for that pair. Thus, normally, a host can have at most one context for a ULID pair. We call this the "default context". Context forking A mechanism which allows ULPs that are aware of multiple locators to use separate contexts for the same ULID pair, in order to be able use different locator pairs for different communication to the same ULID. Context forking causes more than just the default context to be created for a ULID pair. Forked Instance Identifier (FII) In order to handle context forking, a context is identified by a ULID-pair and a forked context identifier. The default context has a FII of zero. Initial contact We use this term to refer to the pre-shim communication when some ULP decides to start communicating with a peer by sending and receiving ULP packets. Typically this would not invoke any operations in the shim, since the shim can defer the context establishment until some arbitrary later point in time. Hash Based Addresses (HBA) A form of IPv6 address where the interface ID is derived from a cryptographic hash of all the prefixes assigned to the host. See [7]. Cryptographically Generated Addresses (CGA) A form of IPv6 address where the interface ID is derived from a cryptographic hash of the public key. See [6]. CGA Parameter Data Structure (PDS) The information that CGA and HBA exchanges in order to inform the peer of how the interface ID was computed. See [6]., [7]. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 14] Internet-Draft Shim6 ProtocolMarchMay 20062.22.2. Notational Conventions A, B, and C are hosts. X is a potentially malicious host. FQDN(A) is the domain name for A. Ls(A) is the locator set for A, which consists of the locators L1(A), L2(A), ... Ln(A). ULID(A) is an upper-layer ID for A. In this proposal, ULID(A) is always one member of A's locator set. CT(X) is a context tag assigned by X. This document also makes use of internal conceptual variables to describe protocol behavior and external variables that an implementation must allow system administrators to change. The specific variable names, how their values change, and how their settings influence protocol behavior are provided to demonstrate protocol behavior. An implementation is not required to have them in the exact form described here, so long as its external behavior is consistent with that described in this document. See Section 6 for a description of the conceptual data structures. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 15] Internet-Draft Shim6 ProtocolMarchMay 2006 3. Assumptions The design intent is to ensure that the shim6 protocol is capable of handling path failures independently of the number of IP addresses (locators) available to the two communicating hosts, and independently of which host detects the failure condition. In the case when host A and host B have an active shim6 state, with host A having only one locator and host B having multiple locators, it might be that host B is trying to send a packet to host A, and has detected a failure condition with the current locator pair. As host B has multiple locators it presumably has multiple ISPs. In this case there are probably alternate egress paths for host B to be able to try to reach A, but B can not vary the destination address (host A locator) to select such alternate paths, since A has only one locator. This leads to the assumption that a host should be able to cause different egress paths from its site to be used. The most reasonable approach to accomplish this is to have the host use different source addresses and have the source address affect the selection of the site egress. The details of how this can be accomplished is beyond the scope of this document, but without this capability the ability of the shim to try different "paths" by trying different locator pairs will have limited utility. The above assumption applies whether or not the ISPs perform ingress filtering. In addition, when the site's ISPs perform ingress filtering based on packet source addresses, shim6 assumes that packets sent with different source and destination combinations have a reasonable chance of making it through the relevant ISP's ingress filters. This can be accomplished in several ways (all outside the scope of this document), such as having the ISPs relax there ingress filters, or selecting the egress such that it matches the IP source address prefix. Further discussion of this issue is captured in [20]. The shim6 approach assumes that there are no IPv6-to-IPv6 NATs on the paths, i.e., that the two ends can exchange their own notion of their IPv6 addresses and that those addresses will also make sense to their peer. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 16] Internet-Draft Shim6 ProtocolMarchMay 2006 4. Protocol Overview The shim6 protocol operates in several phases over time. The following sequence illustrates the concepts: o An application on host A decides to contact B using some upper- layer protocol. This results in the ULP on A sending packets to B. We call this the initial contact. Assuming the IP addresses selected by Default Address Selection [12] and its extensions [13] work, then there is no action by the shim at this point in time. Any shim context establishment can be deferred until later. o Some heuristic on A or B (or both) determine that it is appropriate to pay the shim6 overhead to make this host-to-host communication robust against locator failures. For instance, this heuristic might be that more than 50 packets have been sent or received, or a timer expiration while active packet exchange is in place. This makes the shim initiate the 4-way context establishment exchange. As a result of this exchange, both A and B will know a list of locators for each other. If the context establishment exchange fails, the initiator will then know that the other end does not support shim6, and will continue with standard unicast behavior for the session. o Communication continues without any change for the ULP packets. In particular, there are no shim extension headers added to the ULP packets, since the ULID pair is the same as the locator pair. In addition, there might be some messages exchanged between the shim sub-layers for (un)reachability detection. o At some point in time something fails. Depending on the approach to reachability detection, there might be some advice from the ULP, or the shim (un)reachability detection might discover that there is a problem. At this point in time one or both ends of the communication need to probe the different alternate locator pairs until a working pair is found, and switch to using that locator pair. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 17] Internet-Draft Shim6 ProtocolMarchMay 2006 o Once a working alternative locator pair has been found, the shim will rewrite the packets on transmit, and tag the packets with shim6 Payload extension header, which contains the receiver's context tag. The receiver will use the context tag to find the context state which will indicate which addresses to place in the IPv6 header before passing the packet up to the ULP. The result is that from the perspective of the ULP the packet passes unmodified end-to-end, even though the IP routing infrastructure sends the packet to a different locator. o The shim (un)reachability detection will monitor the new locator pair as it monitored the original locator pair, so that subsequent failures can be detected. o In addition to failures detected based on end-to-end observations, one endpoint might know for certain that one or more of its locators is not working. For instance, the network interface might have failed or gone down (at layer 2), or an IPv6 address might have become deprecated or invalid. In such cases the host can signal its peer that this address is no longer recommended to try. Thus this triggers something similar to a failure handling in that a new, working locator pair must be found. The protocol also has the ability to express other forms of locator preferences. A change in any preferences can be signaled to the peer, which will made the peer record the new preferences. A change in the preferences might optionally make the peer want to use a different locator pair. If it makes this decision, it follows the same locator switching procedure as after a failure (by verifying that its peer is indeed present at the alternate locator, etc). o When the shim thinks that the context state is no longer used, it can garbage collect the state; there is no coordination necessary with the peer host before the state is removed. There is a recovery message defined to be able to signal when there is no context state, which can be used to detect and recover from both premature garbage collection, as well as complete state loss (crash and reboot) of a peer. The exact mechanism to determine when the context state is no longer used is implementation dependent. An implementation might use the existence of ULP state (where known to the implementation) as an indication that the state is still used, combined with a timer (to handle ULP state that might not be known to the shim Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 18] Internet-Draft Shim6 ProtocolMarchMay 2006 sub-layer) to determine when the state is likely to no longer be used. NOTE: The ULP packets in shim6 can be carried completely unmodified as long as the ULID pair is used as the locator pair. After a switch to a different locator pair the packets are "tagged" with a shim6 extension header, so that the receiver can always determine the context to which they belong. This is accomplished by including an 8-octet shim6 Payload Extension header before the (extension) headers that are processed by the IP endpoint sublayer and ULPs. If subsequently the original ULIDs are selected as the active locator pair then the tagging of packets with the shim6 extension header can also be stopped.4.14.1. Context Tags A context between two hosts is actually a context between two ULIDs. The context is identified by a pair of context tags. Each end gets to allocate a context tag, and once the context is established, most shim6 control messages contain the context tag that the receiver of the message allocated. Thus at a minimum the combination of <peer ULID, local ULID, local context tag> have to uniquely identify one context. But since the Payload extension headers are demultiplexed without looking at the locators in the packet, the receiver will need to allocate context tags that are unique for all its contexts. The context tag is a 47-bit number (the largest which can fit in an 8-octet extension header). The mechanism for detecting a loss of context state at the peer that is currently proposed in this document assumes that the receiver can tell the packets that need locator rewriting, even after it has lost all state (e.g., due to a crash followed by a reboot). This is achieved because after a rehoming event the packets that need receive-side rewriting, carry the Payload extension header.4.24.2. Context Forking It has been asserted that it will be important for future ULPs, in particular, future transport protocols, to be able to control which locator pairs are used for different communication. For instance, host A and host B might communicate using both VoIP traffic and ftp traffic, and those communications might benefit from using different locator pairs. However, the fundamental shim6 mechanism uses a single current locator pair for each context, thus a single context can not accomplish this. For this reason, the shim6 protocol supports the notion of context Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 19] Internet-Draft Shim6 ProtocolMarchMay 2006 forking. This is a mechanism by which a ULP can specify (using some API not yet defined) that a context for e.g., the ULID pair <A1, B2> should be forked into two contexts. In this case the forked-off context will be assigned a non-zero Forked Instance Identifier, while the default context has FII zero. The Forked Instance Identifier is a 32-bit identifier which has no semantics in the protocol other then being part of the tuple which identifies the context. The hosts can allocate FIIs e.g., as sequential numbers for any given ULID pair. No other special considerations are needed in the shim6 protocol to handle forked contexts. Note that forking as specified does NOT allow A to be able to tell B that certain traffic (a 5-tuple?) should be forked for the reverse direction. The shim6 forking mechanism as specified applies only to the sending of ULP packets. If some ULP wants to fork for both directions, it is up to the ULP to set this up, and then instruct the shim at each end to transmit using the forked context.4.34.3. API Extensions Several API extensions have been discussed for shim6, but their actual specification is out of scope for this document. The simplest one would be to add a socket option to be able to have traffic bypass the shim (not create any state, and not use any state created by other traffic). This could be an IPV6_DONTSHIM socket option. Such an option would be useful for protocols, such as DNS, where the application has its own failover mechanism (multiple NS records in the case of DNS) and using the shim could potentially add extra latency with no added benefits. Some other API extensions are discussed in AppendixB 4.4A 4.4. Securing shim6 The mechanisms are secured using a combination of techniques: o The HBA technique [7] for verifying the locators to prevent an attacker from redirecting the packet stream to somewhere else. o Requiring a Reachability Probe+Reply before a new locator is used as the destination, in order to prevent 3rd party flooding attacks. o The first message does not create any state on the responder. Essentially a 3-way exchange is required before the responder Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 20] Internet-Draft Shim6 ProtocolMarchMay 2006 creates any state. This means that a state-based DoS attack (trying to use up all of memory on the responder) at least provides an IPv6 address that the attacker was using. o The context establishment messages use nonces to prevent replay attacks, and to prevent off-path attackers from interfering with the establishment. o Every control message of the shim6 protocol, past the context establishment, carry the context tag assigned to the particular context. This implies that an attacker needs to discover that context tag before being able to spoof any shim6 control message. Such discovery probably requires to be along the path in order to be sniff the context tag value. The result is that through this technique, the shim6 protocol is protected against off-path attackers.4.54.5. Overview of Shim Control Messages The shim6 context establishment is accomplished using four messages; I1, R1, I2, R2. Normally they are sent in that order from initiator and responder, respectively. Should both ends attempt to set up context state at the same time (for the same ULID pair), then their I1 messages might cross in flight, and result in an immediate R2 message. [The names of these messages are borrowed from HIP [25].] R1bis and I2bis messages are defined, which are used to recover a context after it has been lost. A R1bis message is sent when a shim6 control or Payload extension header arrives and there is no matching context state at the receiver. When such a message is received, it will result in the re-creation of the shim6 context using the I2bis and R2 messages. The peers' lists of locators are normally exchanged as part of the context establishment exchange. But the set of locators might be dynamic. For this reason there is a Update Request and Update Acknowledgement messages, and a Locator List option. Even when the list of locators is fixed, a host might determine that some preferences might have changed. For instance, it might determine that there is a locally visible failure that implies that some locator(s) are no longer usable. This uses a Locator Preferences option in the Update Request message. The mechanism for (un)reachability detection is called Forced Bidirectional Communication (FBD). The FBD approach uses a Keepalive message, which is sent when a host has received packets from theNordmark & Bagnulo Expires September 5, 2006 [Page 21] Internet-Draft Shim6 Protocol March 2006peer, but the ULP has not given the host an opportunity to send any Nordmark & Bagnulo Expires November 16, 2006 [Page 21] Internet-Draft Shim6 Protocol May 2006 packet to the peer. The message type is reserved in this document, but the message format and processing rules are specified in [8]. In addition, when the context is established and there is a failure there needs to be a way to probe the set of locator pairs to efficiently find a working pair. This document reserves an Probe message type, with the packet format and processing rules specified in [8]. The above probe and keepalive messages assume we have an established ULID-pair context. However, communication might fail during the initial contact (that is, when the application or transport protocol is trying to setup some communication). This is handled using the mechanisms in the ULP to try different address pairs as specified in [12] [13]. In the future versions of the protocol, and with a richer API between the ULP and the shim, the shim might be help optimize discovering a working locator pair during initial contact. This is for further study.4.64.6. Extension Header Order Since the shim is placed between the IP endpoint sub-layer and the IP routing sub-layer in the host, the shim header will be placed before any endpoint extension headers (fragmentation headers, destination options header, AH, ESP), but after any routing related headers (hop- by-hop extensions header, routing header, a destinations options header which precedes a routing header). When tunneling is used, whether IP-in-IP tunneling or the special form of tunneling that Mobile IPv6 uses (with Home Address Options and Routing header type 2), there is a choice whether the shim applies inside the tunnel or outside the tunnel, whicheffectsaffects the location of the shim6 header. In most cases IP-in-IP tunnels are used as a routing technique, thus it makes sense to apply them on the locators which means that the sender would insert the shim6 header after any IP-in-IP encapsulation; this is what occurs naturally when routers apply IP- in-IP encapsulation. Thus the packets would have: o Outer IP header o Inner IP header o Shim6 extension header (if needed> o ULP But the shim can also be used to create "shimmed tunnels" i.e., whereNordmark & Bagnulo Expires September 5, 2006 [Page 22] Internet-Draft Shim6 Protocol March 2006an IP-in-IP tunnel uses the shim to be able to switch the tunnel Nordmark & Bagnulo Expires November 16, 2006 [Page 22] Internet-Draft Shim6 Protocol May 2006 endpoint addresses between different locators. In such a case the packets would have: o Outer IP header o Shim6 extension header (if needed> o Inner IP header o ULP In any case, the receiver behavior is well-defined; a receiver processes the extension headers in order. However, the precise interaction between Mobile IPv6 and shim6 is for further study, but it might make sense to have Mobile IPv6 operate on locators as well, meaning that the shim would be layered on top of the MIPv6 mechanism. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 23] Internet-Draft Shim6 ProtocolMarchMay 2006 5. Message Formats The shim6 messages are all carried using a new IP protocol number [to be assigned by IANA]. The shim6 messages have a common header, defined below, with some fixed fields, followed by type specific fields. The shim6 messages are structured as an IPv6 extension header since the Payload extension header is used to carry the ULP packets after a locator switch. The shim6 control messages use the same extension header formats so that a single "protocol number" needs to be allowed through firewalls in order for shim6 to function across the firewall.5.15.1. Common shim6 Message Format The first 17 bits of the shim6 header is common for the Payload extension header and the control messages and looks as follows: 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Next Header | Hdr Ext Len |P| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: The payload which follows this header. Hdr Ext Len: 8-bit unsigned integer. Length of the shim6 header in 8-octet units, not including the first 8 octets. P: A single bit to distinguish Payload extension headers from control messages.5.25.2. Payload Extension Header Format The payload extension headers is used to carry ULP packets where the receiver must replace the content of the source and/or destination fields in the IPv6 header before passing the packet to the ULP. Thus this extension header is required when the locators pair that is used is not the same as the ULID pair.Nordmark & Bagnulo Expires September 5, 2006 [Page 24] Internet-Draft Shim6 Protocol March 20060 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Next Header | 0 |1| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Receiver Context Tag | Nordmark & Bagnulo Expires November 16, 2006 [Page 24] Internet-Draft Shim6 Protocol May 2006 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: The payload which follows this header. Hdr Ext Len: 0 (since the header is 8 octets). P: Set to one. A single bit to distinguish this from the shim6 control messages. Receiver Context Tag: 47-bit unsigned integer. Allocated by the receiver for use to identify the context.5.35.3. Common Shim6 Control header The common part of the header has a next header and header extension length field which is consistent with the other IPv6 extension headers, even if the next header value is always "NO NEXT HEADER" for the control messages; only the payload extension header use the Next Header field. The shim6 headers must be a multiple of 8 octets, hence the minimum size is 8 octets. The common shim control message header is as follows: 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Next Header | Hdr Ext Len |0| Type |Type-specific|0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Type-specific format | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields:Nordmark & Bagnulo Expires September 5, 2006 [Page 25] Internet-Draft Shim6 Protocol March 2006Next Header: 8-bit selector. Normally set to NO_NXT_HDR (59). Hdr Ext Len: 8-bit unsigned integer. Length of the shim6 header in 8-octet units, not including the first 8 octets. P: Set to zero. A single bit to distinguish this from the shim6 payload extension header. Nordmark & Bagnulo Expires November 16, 2006 [Page 25] Internet-Draft Shim6 Protocol May 2006 Type: 7-bit unsigned integer. Identifies the actual message from the table below. Type codes 0-63 will not trigger R1bis messages on a missing context, while 64- 127 will trigger R1bis. 0: A single bit (set to zero) which allows shim6 and HIP to have a common header format yet telling shim6 and HIP messages apart. Checksum: 16-bit unsigned integer. The checksum is the 16-bit one's complement of the one's complement sum of the entire shim6 header message starting with the shim6 next header field, and ending as indicated by the Hdr Ext Len. Thus when there is a payload following the shim6 header, the payload is NOT included in the shim6 checksum. Note that unlike protocol like ICMPv6, there is no pseudo-header checksum part of the checksum, in order to provide locator agility without having to change the checksum. Type-specific: Part of message that is different for different message types.Nordmark & Bagnulo Expires September 5, 2006 [Page 26] Internet-Draft Shim6 Protocol March 2006+------------+-----------------------------------------------------+ | Type Value | Message | +------------+-----------------------------------------------------+ | 1 | I1 (first establishment message from the initiator) | | | | | 2 | R1 (first establishment message from the responder) | | | | | 3 | I2 (2nd establishment message from the initiator) | | | | | 4 | R2 (2nd establishment message from the responder) | | | | | 5 | R1bis (Reply to reference to non-existent context) | | | | | 6 | I2bis (Reply to a R1bis message) | | | | | 64 | Update Request | | | | | 65 | Update Acknowledgement | | | | | 66 | Keepalive | | | | | 67 | Probe Message | +------------+-----------------------------------------------------+ Table 15.4Nordmark & Bagnulo Expires November 16, 2006 [Page 26] Internet-Draft Shim6 Protocol May 2006 5.4. I1 Message Format The I1 message is the first message in the context establishment exchange. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 1 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Initiator Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Initiator Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+Nordmark & Bagnulo Expires September 5, 2006 [Page 27] Internet-Draft Shim6 Protocol March 2006Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options. Type: 1 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Initiator Context Tag: 47-bit field. The Context Tag the initiator has allocated for the context. Initiator Nonce: 32-bit unsigned integer. A random number picked by the initiator which the responder will return in the R1 message. The following options are defined for this message: Nordmark & Bagnulo Expires November 16, 2006 [Page 27] Internet-Draft Shim6 Protocol May 2006 ULID pair: When the IPv6 source and destination addresses in the IPv6 header does not match the ULID pair, this option MUST be included. An example of this is when recovering from a lost context. Forked Instance Identifier: When another instance of an existent context with the same ULID pair is being created, a Forked Instance Identifier option is included to distinguish this new instance from the existent one.5.5Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.5. R1 Message Format The R1 message is the second message in the context establishment exchange. The responder sends this in response to an I1 message, without creating any state specific to the initiator.Nordmark & Bagnulo Expires September 5, 2006 [Page 28] Internet-Draft Shim6 Protocol March 20060 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 2 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum | Reserved2 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Initiator Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Responder Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options. Type: 2 Nordmark & Bagnulo Expires November 16, 2006 [Page 28] Internet-Draft Shim6 Protocol May 2006 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Reserved2: 16-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Initiator Nonce: 32-bit unsigned integer. Copied from the I1 message. Responder Nonce: 32-bit unsigned integer. A number picked by the responder which the initiator will return in the I2 message. The following options are defined for this message: Responder Validator: Variable length option. Typically a hash generated by the responder, which the responder uses together with the Responder Nonce value to verify that an I2 message is indeed sent in response to a R1 message, and that the parameters in the I2 message are the same as those in the I1 message.Nordmark & Bagnulo Expires September 5, 2006 [Page 29] Internet-Draft Shim6 Protocol March 2006 5.6Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.6. I2 Message Format The I2 message is the third message in the context establishment exchange. The initiator sends this in response to a R1 message, after checking the Initiator Nonce, etc. Nordmark & Bagnulo Expires November 16, 2006 [Page 29] Internet-Draft Shim6 Protocol May 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 3 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Initiator Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Initiator Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Responder Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Reserved2 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 2, since the header is 24 octets when there are no options. Type: 3 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Initiator Context Tag: 47-bit field. The Context Tag the initiator has allocated for the context. Initiator Nonce: 32-bit unsigned integer. A random number picked by the initiator which the responder will return in the R2 message. Responder Nonce: 32-bit unsigned integer. Copied from the R1 message. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 30] Internet-Draft Shim6 ProtocolMarchMay 2006Responder Nonce: 32-bit unsigned integer. Copied from the R1 message.Reserved2: 32-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. (Needed to make the options start on a multiple of 8 octet boundary.) The following options are defined for this message: Responder Validator: Variable length option. Just a copy of the Responder Validator option in the R1 message. ULID pair: When the IPv6 source and destination addresses in the IPv6 header does not match the ULID pair, this option MUST be included. An example of this is when recovering from a lost context. Forked Instance Identifier: When another instance of an existent context with the same ULID pair is being created, a Forked Instance Identifier option is included to distinguish this new instance from the existent one. Locator list: Optionally sent when the initiator immediately wants to tell the responder its list of locators. When it is sent, the necessary HBA/CGA information for verifying the locator list MUST also be included. Locator Preferences: Optionally sent when the locators don't all have equal preference. CGA Parameter Data Structure: Included when the locator list is included so the receiver can verify the locator list. CGA Signature: Included when the some of the locators in the list use CGA (and not HBA) for verification.5.7Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.7. R2 Message Format The R2 message is the fourth message in the context establishment exchange. The responder sends this in response to an I2 message. The R2 message is also used when both hosts send I1 messages at the same time and the I1 messages cross in flight. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 31] Internet-Draft Shim6 ProtocolMarchMay 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 4 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Responder Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Initiator Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options. Type: 4 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Responder Context Tag: 47-bit field. The Context Tag the responder has allocated for the context. Initiator Nonce: 32-bit unsigned integer. Copied from the I2 message. The following options are defined for this message: Locator List: Optionally sent when the responder immediately wants to tell the initiator its list of locators. When it is sent, the necessary HBA/CGA information for verifying the locator list MUST also be included. Locator Preferences: Optionally sent when the locators don't all have equal preference. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 32] Internet-Draft Shim6 ProtocolMarchMay 2006 CGA Parameter Data Structure: Included when the locator list is included so the receiver can verify the locator list. CGA Signature: Included when the some of the locators in the list use CGA (and not HBA) for verification.5.8Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.8. R1bis Message Format Should a host receive a packet with a shim Payload extension header or shim6 control message with type code 64-127 (such as an Update or Probe message), and the host does not have any context state for the received context tag, then it will generate a R1bis message. This message allows the sender of the packet referring to the non- existent context to re-establish the context with a reduced context establishment exchange. Upon the reception of the R1bis message, the receiver can proceed reestablishing the lost context by directly sending an I2bis message. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 5 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Packet Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Responder Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options.Type: 5Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 33] Internet-Draft Shim6 ProtocolMarchMay 2006 Type: 5 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Packet Context Tag: 47-bit unsigned integer. The context tag contained in the received packet that triggered the generation of the R1bis message. Responder Nonce: 32-bit unsigned integer. A number picked by the responder which the initiator will return in the I2bis message. The following options are defined for this message: Responder Validator: Variable length option. Typically a hash generated by the responder, which the responder uses together with the Responder Nonce value to verify that an I2bis message is indeed sent in response to a R1bis message.5.9Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.9. I2bis Message Format The I2bis message is the third message in the context recovery exchange. This is sent in response to a R1bis message, after checking that the R1bis message refers to an existing context, etc. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 34] Internet-Draft Shim6 ProtocolMarchMay 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 6 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Initiator Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Initiator Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Responder Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Reserved2 | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Packet Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 3, since the header is 32 octets when there are no options. Type: 6 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Initiator Context Tag: 47-bit field. The Context Tag the initiator has allocated for the context. Initiator Nonce: 32-bit unsigned integer. A random number picked by the initiator which the responder will return in the R2 message. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 35] Internet-Draft Shim6 ProtocolMarchMay 2006 Responder Nonce: 32-bit unsigned integer. Copied from the R1bis message. Reserved2: 49-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. (Note that 17 bits are not sufficient since the options need start on a multiple of 8 octet boundary.) Packet Context Tag: 47-bit unsigned integer. Copied from the Packet Context Tag contained in the received R1bis. The following options are defined for this message: Responder Validator: Variable length option. Just a copy of the Responder Validator option in the R1bis message. ULID pair: When the IPv6 source and destination addresses in the IPv6 header does not match the ULID pair, this option MUST be included. Forked Instance Identifier: When another instance of an existent context with the same ULID pair is being created, a Forked Instance Identifier option is included to distinguish this new instance from the existent one. Locator list: Optionally sent when the initiator immediately wants to tell the responder its list of locators. When it is sent, the necessary HBA/CGA information for verifying the locator list MUST also be included. Locator Preferences: Optionally sent when the locators don't all have equal preference. CGA Parameter Data Structure: Included when the locator list is included so the receiver can verify the locator list. CGA Signature: Included when the some of the locators in the list use CGA (and not HBA) for verification.5.10Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.10. Update Request Message Format The Update Request Message is used to update either the list of locators, the locator preferences, and both. When the list of locators is updated, the message also contains the option(s) Nordmark & Bagnulo Expires November 16, 2006 [Page 36] Internet-Draft Shim6 Protocol May 2006 necessary for HBA/CGA to secure this. The basic sanity check that prevents off-path attackers from generating bogus updates is the context tag in the message.Nordmark & Bagnulo Expires September 5, 2006 [Page 36] Internet-Draft Shim6 Protocol March 2006The update message contains options (the Locator List and the Locator Preferences) that, when included, completely replace the previous locator list and locator preferences, respectively. Thus there is no mechanism to just send deltas to the locator list. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 64 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Receiver Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Request Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options. Type: 64 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Receiver Context Tag: 47-bit field. The Context Tag the receiver has allocated for the context. Request Nonce: 32-bit unsigned integer. A random number picked by the initiator which the peer will return in the acknowledgement message. The following options are defined for this message: Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 37] Internet-Draft Shim6 ProtocolMarchMay 2006 Locator List: The list of the sender's (new) locators. The locators might be unchanged and only the preferences have changed. Locator Preferences: Optionally sent when the locators don't all have equal preference. CGA Parameter Data Structure (PDS): Included when the locator list is included and the PDS was not included in the I2/I2bis/R2 messages, so the receiver can verify the locator list. CGA Signature: Included when the some of the locators in the list use CGA (and not HBA) for verification.5.11Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.11. Update Acknowledgement Message Format This message is sent in response to a Update Request message. It implies that the Update Request has been received, and that any new locators in the Update Request can now be used as the source locators of packets. But it does not imply that the (new) locators have been verified to be used as a destination, since the host might defer the verification of a locator until it sees a need to use a locator as the destination. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 59 | Hdr Ext Len |0| Type = 65 | Reserved1 |0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Checksum |R| | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | Receiver Context Tag | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Request Nonce | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Options + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields:Next Header: NO_NXT_HDR (59).Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 38] Internet-Draft Shim6 ProtocolMarchMay 2006 Next Header: NO_NXT_HDR (59). Hdr Ext Len: At least 1, since the header is 16 octets when there are no options. Type: 65 Reserved1: 7-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. R: 1-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. Receiver Context Tag: 47-bit field. The Context Tag the receiver has allocated for the context. Request Nonce: 32-bit unsigned integer. Copied from the Update Request message. No options are currently defined for this message.5.12Future protocol extensions might define additional options for this message. The C-bit in the option format defines how such a new option will be handled by an implementation. See Section 5.14. 5.12. Keepalive Message Format This message format is defined in [8]. The message is used to ensure that when a peer is sending ULP packets on a context, it always receives some packets in the reverse direction. When the ULP is sending bidirectional traffic, no extra packets need to be inserted. But for a unidirectional ULP traffic pattern, the shim will send back some Keepalive messages when it is receiving ULP packets.5.135.13. Probe Message Format This message and its semantics are defined in [8]. The idea behind that mechanism is to be able to handle the case when one locator pair works in from A to B, and another locator pair works from B to A, but there is no locator pair which works in both directions. The protocol mechanism is that as A is sending probe messages to B, B will observe which locator pairs it has received from and report that back in probe messages it is sending to A.5.14Nordmark & Bagnulo Expires November 16, 2006 [Page 39] Internet-Draft Shim6 Protocol May 2006 5.14. Option Formats The format of the options is a snapshot of the current HIP option format [25]. However, there is no intention to track any changes to the HIP option format, nor is there an intent to use the same name space for the option type values. But using the same format will hopefully make it easier to import HIP capabilities into shim6 asNordmark & Bagnulo Expires September 5, 2006 [Page 39] Internet-Draft Shim6 Protocol March 2006extensions to shim6, should this turn out to be useful. All of the TLV parameters have a length (including Type and Length fields) which is a multiple of 8 bytes. When needed, padding MUST be added to the end of the parameter so that the total length becomes a multiple of 8 bytes. This rule ensures proper alignment of data. If padding is added, the Length field MUST NOT include the padding. Any added padding bytes MUST be zeroed by the sender, and their values SHOULD NOT be checked by the receiver. Consequently, the Length field indicates the length of the Contents field (in bytes). The total length of the TLV parameter (including Type, Length, Contents, and Padding) is related to the Length field according to the following formula: Total Length = 11 + Length - (Length + 3) % 8; 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type |C| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ ~ ~ Contents ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Type: 15-bit identifier of the type of option. The options defined in this document are below. C: Critical. One if this parameter is critical, and MUST be recognized by the recipient, zero otherwise. An implementation might view the C bit as part of the Type field, by multiplying the type values in this specification by two. Nordmark & Bagnulo Expires November 16, 2006 [Page 40] Internet-Draft Shim6 Protocol May 2006 Length: Length of the Contents, in bytes. Contents: Parameter specific, defined by Type. Padding: Padding, 0-7 bytes, added if needed.Nordmark & Bagnulo Expires September 5, 2006 [Page 40] Internet-Draft Shim6 Protocol March 2006+------+---------------------------------+ | Type | Option Name | +------+---------------------------------+ | 1 | Responder Validator | | | | | 2 | Locator List | | | | | 3 | Locator Preferences | | | | | 4 | CGA Parameter Data Structure | | | | | 5 | CGA Signature | | | | | 6 | ULID Pair | | | | | 7 | Forked Instance Identifier | | | | | 10 | Probe Option | | | | | 11 | Reachability Option | | | | | 12 | Payload Reception Report Option | +------+---------------------------------+ Table 25.14.1Future protocol extensions might define additional options for the SHIM6 messages. The C-bit in the option format defines how such a new option will be handled by an implementation. If a host receives an option that it does not understand (an option that was defined in some future extension to this protocol) or is not listed as a valid option for the different message types above, then the Critical bit in the option determines the outcome. o If C=0 then the option is silently ignored, and the rest of the message is processed. o If C=1 then the host SHOULD send back an ICMP parameter problem (type 4, code 1), with the Pointer referencing the first octet in the option Type field. When C=1 the message MUST NOT be processed. Nordmark & Bagnulo Expires November 16, 2006 [Page 41] Internet-Draft Shim6 Protocol May 2006 5.14.1. Responder Validator Option Format The responder can choose exactly what inputusesis used to compute the validator, and what one-way function (MD5, SHA1) it uses, as long as the responder can check that the validator it receives back in the I2 or I2bis message is indeed one that: 1)- it computed, 2)- it computed for the particular context, and 3)- that it isn't a replayed I2/I2bis message. Some suggestions on how to generate the validators are captured in Section7.9.17.10.1 and Section7.16.1. Nordmark & Bagnulo Expires September 5, 2006 [Page 41] Internet-Draft Shim6 Protocol March 20067.17.1. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 1 |0| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ Validator ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Validator: Variable length content whose interpretation is local to the responder. Padding: Padding, 0-7 bytes, added if needed. See Section 5.14.5.14.25.14.2. Locator List Option Format The Locator List Option is used to carry all the locators of the sender. Note that the order of the locators is important, since the Locator Preferences refers to the locators by using the index in the list. Note that we carry all the locators in this option even though some of them can be created automatically from the CGA Parameter Data Structure. Nordmark & Bagnulo Expires November 16, 2006 [Page 42] Internet-Draft Shim6 Protocol May 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 2 |0| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Locator List Generation | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Num Locators | N Octets of Verification Method | +-+-+-+-+-+-+-+-+ | ~ ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ Locators 1 through N ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields:Nordmark & Bagnulo Expires September 5, 2006 [Page 42] Internet-Draft Shim6 Protocol March 2006Locator List Generation: 32-bit unsigned integer. Indicates a generation number which is increased by one for each new locator list. This is used to ensure that the index in the Locator Preferences refer to the right version of the locator list. Num Locators: 8-bit unsigned integer. The number of locators that are included in the option. We call this number "N" below. Verification Method: N octets. The i'th octet specifies the verification method for the i'th locator. Padding: Padding, 0-7 bytes, added if needed so that the Locators start on a multiple of 8 octet boundary. NOTE that for this option there is never a need to pad at the end, since the locators are a multiple of 8 octets in length. This internal padding is included in the length field. Locators: N 128-bit locators. The defined verification methods are: Nordmark & Bagnulo Expires November 16, 2006 [Page 43] Internet-Draft Shim6 Protocol May 2006 +-------+----------+ | Value | Method | +-------+----------+ | 0 | Reserved | | | | | 1 | HBA | | | | | 2 | CGA | | | | | 3-255 | Reserved | +-------+----------+ Table 35.14.35.14.3. Locator Preferences Option Format The Locator Preferences option can have some flags to indicate whether or not a locator is known to work. In addition, the sender can include a notion of preferences. It might make sense to define "preferences" as a combination of priority and weight the same way that DNS SRV records has such information. The priority would provide a way to rank the locators, and within a given priority, the weight would provide a way to do some load sharing. See [9] for howNordmark & Bagnulo Expires September 5, 2006 [Page 43] Internet-Draft Shim6 Protocol March 2006SRV defines the interaction of priority and weight. The minimum notion of preferences we need is to be able to indicate that a locator is "dead". We can handle this using a single octet flag for each locator. We can extend that by carrying a larger "element" for each locator. This document presently also defines 2-octet and 3-octet elements, and we can add more information by having even larger elements if need be. The locators are not included in the preference list. Instead, the first element refers to locator that was in the first element in the Locator List option. The generation number carried in this option and the Locator List option is used to verify that they refer to the same version of the locator list. Nordmark & Bagnulo Expires November 16, 2006 [Page 44] Internet-Draft Shim6 Protocol May 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 3 |0| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Locator List Generation | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Element Len | Element[1] | Element[2] | Element[3] | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ ... ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Case of Element Len = 1 is depicted. Fields: Locator List Generation: 32-bit unsigned integer. Indicates a generation number for the locator list to which the elements should apply. Element Len: 8-bit unsigned integer. The length in octets of each element. This specification defines the cases when the length is 1, 2, or 3. Element[i]: A field with a number of octets defined by the Element Len field. Provides preferences for the i'th locator in the Locator List option that is in use.Nordmark & Bagnulo Expires September 5, 2006 [Page 44] Internet-Draft Shim6 Protocol March 2006Padding: Padding, 0-7 bytes, added if needed. See Section 5.14. When the Element length equals one, then the element consists of only a one octet flags field. The currently defined set of flags are: BROKEN: 0x01 TEMPORARY: 0x02 The intent of TEMPORARY is to allow the distinction between more stable addresses and less stable addresses when shim6 is combined with IP mobility, when we might have more stable home locators, and less stable care-of-locators. When the Element length equals two, then the element consists of a 1 octet flags field followed by a 1 octet priority field. The priority has the same semantics as the priority in DNS SRV records. Nordmark & Bagnulo Expires November 16, 2006 [Page 45] Internet-Draft Shim6 Protocol May 2006 When the Element length equals three, then the element consists of a 1 octet flags field followed by a 1 octet priority field, and a 1 octet weight field. The weight has the same semantics as the weight in DNS SRV records. This document doesn't specify the format when the Element length is more than three, except that any such formats MUST be defined so that the first three octets are the same as in the above case, that is, a of a 1 octet flags field followed by a 1 octet priority field, and a 1 octet weight field.5.14.45.14.4. CGA Parameter Data Structure Option Format This option contains the CGA Parameter Data Structure (PDS). When HBA is used to verify the locators, the PDS contains the HBA multiprefix extension. When CGA is used to verify the locators, in addition to the PDS option, the host also needs to include the signature in the form of a CGA Signature option. 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 4 |0| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ CGA Parameter Data Structure ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+Nordmark & Bagnulo Expires September 5, 2006 [Page 45] Internet-Draft Shim6 Protocol March 2006Fields: CGA Parameter Data Structure: Variable length content. Content defined in [6] and [7]. Padding: Padding, 0-7 bytes, added if needed. See Section 5.14.5.14.55.14.5. CGA Signature Option Format When CGA is used for verification of one or more of the locators in the Locator List option, then the message in question will need to contain this option. Nordmark & Bagnulo Expires November 16, 2006 [Page 46] Internet-Draft Shim6 Protocol May 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 5 |0| Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ CGA Signature ~ ~ +-+-+-+-+-+-+-+-+ ~ | Padding | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: CGA Signature: A variable-length field containing a PKCS#1 v1.5 signature, constructed by using the sender's private key over the following sequence of octets: 1. The 128-bit CGA Message Type tag [CGA] value for SHIM6, 0x4A 30 5662 4858 574B 3655 416F 506A 6D48. (The tag value has been generated randomly by the editor of this specification.). 2. The Locator List Generation value of the correspondent Locator List Option. 3. The subset of locators included in the correspondent Locator List Option which verification method is set to CGA. The locators MUST be included in the order they are listed in the Locator List Option.Nordmark & Bagnulo Expires September 5, 2006 [Page 46] Internet-Draft Shim6 Protocol March 2006Padding: Padding, 0-7 bytes, added if needed. See Section 5.14.5.14.65.14.6. ULID Pair Option Format I1, I2, and I2bis messages MUST contain the ULID pair; normally this is in the IPv6 source and destination fields. In case that the ULID for the context differ from the address pair included in the source and destination address fields of the IPv6 packet used to carry the I1/I2/I2bis message, the ULID pair option MUST be included in the I1/ I2/I2bis message. Nordmark & Bagnulo Expires November 16, 2006 [Page 47] Internet-Draft Shim6 Protocol May 2006 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 6 |0| Length = 36 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Reserved2 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Sender ULID + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + Receiver ULID + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Reserved2: 32-bit field. Reserved for future use. Zero on transmit. MUST be ignored on receipt. (Needed to make the ULIDs start on a multiple of 8 octet boundary.) Sender ULID: A 128-bit IPv6 address. Receiver ULID: A 128-bit IPv6 address.Nordmark & Bagnulo Expires September 5, 2006 [Page 47] Internet-Draft Shim6 Protocol March 2006 5.14.75.14.7. Forked Instance Identifier Option Format 0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Type = 7 |0| Length = 4 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Forked Instance Identifier | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Fields: Forked Instance Identifier: 32-bit field containing the identifier of the particular forked instance.5.14.85.14.8. Probe Option Format This option is defined in [8].5.14.9Nordmark & Bagnulo Expires November 16, 2006 [Page 48] Internet-Draft Shim6 Protocol May 2006 5.14.9. Reachability Option Format This option is defined in [8].5.14.105.14.10. Payload Reception Report Option Format This option is defined in [8]. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page48]49] Internet-Draft Shim6 ProtocolMarchMay 2006 6. Conceptual Model of a Host This section describes a conceptual model of one possible data structure organization that hosts will maintain for the purposes of shim6. The described organization is provided to facilitate the explanation of how the shim6 protocol should behave. This document does not mandate that implementations adhere to this model as long as their external behavior is consistent with that described in this document.6.16.1. Conceptual Data Structures The key conceptual data structure for the shim6 protocol is the ULID pair context. This is a data structure which contains the following information: o The state of the context. See Section 6.2. o The peer ULID; ULID(peer) o The local ULID; ULID(local) o The Forked Instance Identifier; FII. This is zero for the default context i.e., when there is no forking. o The list of peer locators, with their preferences; Ls(peer) o The generation number for the most recently received, verified peer locator list. o For each peer locator, the verification method to use (from the Locator List option). o For each peer locator, a bit whether it has been verified using HBA or CGA, and a bit whether the locator has been probed to verify that the ULID is present at that location. o The preferred peer locator - used as destination; Lp(peer) o The set of local locators and the preferences; Ls(local) o The generation number for the most recently sent Locator List option. o The preferred local locator - used as source; Lp(local) o The context tag used to transmit control messages and payload extension headers - allocated by the peer; CT(peer) Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page49]50] Internet-Draft Shim6 ProtocolMarchMay 2006 o The context to expect in received control messages and payload extension headers - allocated by the local host; CT(local) o Timers for retransmission of the messages during context establishment and update messages. o Depending how an implementation determines whether a context is still in use, there might be a need to track the last time a packet was sent/received using the context. o Reachability state for the locator pairs as specified in [8]. o During pair exploration, information about the probe messages that have been sent and received as specified in [8].6.26.2. Context States The states that are used to describe the shim6 protocol are as follows: +---------------------+---------------------------------------------+ | State | Explanation | +---------------------+---------------------------------------------+ | IDLE | State machine start | | | | | I1-SENT | Initiating context establishment exchange | | | | | I2-SENT | Waiting to complete context establishment | | | exchange | | | | | I2BIS-SENT | Potential context loss detected | | | | | | | | ESTABLISHED | SHIM context established | | | | | E-FAILED | Context establishment exchange failed | | | | | NO-SUPPORT | ICMPpayloadUnrecognized Next Header typeunknown| | | (type 4, code 1) received indicating | | |received indicatingthat shim6 is not| | |supported | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page50]51] Internet-Draft Shim6 ProtocolMarchMay 2006 In addition, in each of the aforementioned states, the following state information is stored: +---------------------+---------------------------------------------+ | State | Information | +---------------------+---------------------------------------------+ | IDLE | None | | | | | I1-SENT | ULID(peer), ULID(local), [FII], CT(local), | | | INIT nonce, Lp(local), Lp(peer), Ls(local) | | | | | I2-SENT | ULID(peer), ULID(local), [FII], CT(local), | | | INIT nonce, RESP nonce, Lp(local), Lp(peer),| | | Ls(local) | | | | | ESTABLISHED | ULID(peer), ULID(local), [FII], CT(local), | | | CT(peer), Lp(local), Lp(peer), Ls(local) | | | Ls(peer), INIT nonce?(to receive late R2) | | | | | I2BIS-SENT | ULID(peer), ULID(local), [FII], CT(local), | | | CT(peer), Lp(local), Lp(peer), Ls(local) | | | Ls(peer), CT(R1bis) | | | | | E-FAILED | ULID(peer), ULID(local) | | | | | NO-SUPPORT | ULID(peer), ULID(local) | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page51]52] Internet-Draft Shim6 ProtocolMarchMay 2006 7. Establishing ULID-Pair Contexts ULID-pair contexts are established using a 4-way exchange, which allows the responder to avoid creating state on the first packet. As part of this exchange each end allocates a context tag, and it shares this context tag and its set of locators with the peer. In some cases the 4-way exchange is not necessary, for instance when both ends try to setup the context at the same time, or when recovering from a context that has been garbage collected or lost at one of the hosts.7.1 Uniqness7.1. Uniqueness of Context Tags As part of establishing a new context, each host has to assign a unique context tag. Since the Payload Extension headers are demultiplexed based solely on the context tag value (without using the locators), the context tag MUST be unique for each context. In addition, in order to minimize the reuse of context tags, the host SHOULD randomly cycle through the 2^47 context tag values,(e.g. following the guidelines described in [17]).7.27.2. Locator Verification The peer's locators might need to be verified during context establishment as well as when handling locator updates in Section 10. There are two separate aspects of locator verification. One is to verify that the locator is tied to the ULID, i.e., that the host which "owns" the ULID is also the one that is claiming the locator "ownership". The shim6 protocol uses the HBA or CGA techniques for doing this verification. The other is to verify that the host is indeed reachable at the claimed locator. Such verification is needed both to make sure communication can proceed, but also to prevent 3rd party flooding attacks [19]. These different verifications happen at different times, since the first might need to be performed before packets can be received by the peer with the source locator in question, but the latter verification is only needed before packets are sent to the locator. Before a host can use a locator (different than the ULID) as the source locator, it must know that the peer will accept packets with that source locator as being part of this context. Thus the HBA/CGA verification SHOULD be performed by the host before the host acknowledges the new locator, by sending an Update Acknowledgement message, or an R2 message. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page52]53] Internet-Draft Shim6 ProtocolMarchMay 2006 Before a host can use a locator (different than the ULID) as the destination locator it MUST perform the HBA/CGA verification if this was not performed before upon the reception of the locator set. In addition, it MUST verify that the ULID is indeed present at that locator. This verification is performed by doing a return- routability test as part of the Probe sub-protocol [8]. If the verification method in the Locator List option is not supported by the host, or if the verification method is not consistent with the CGA Parameter Data Structure (e.g., the Parameter Data Structure doesn't contain the multiprefix extension, and the verification method says to use HBA), then the host MUST ignore the Locator List and the message in which it is contained, and the host SHOULD generates an ICMP parameter problem (type 4, code 0), with the Pointer referencing the octet in the Verification method that was found inconsistent.7.37.3. Normal context establishment The normal context establishment consists of a 4 message exchange in the order of I1, R1, I2, R2 as can be seen in Figure 24. Initiator Responder IDLE IDLE ------------- I1 --------------> I1-SENT <------------ R1 --------------- IDLE ------------- I2 --------------> I2-SENT <------------ R2 --------------- ESTABLISHED ESTABLISHED Figure 24: Normal context establishment7.47.4. Concurrent context establishment When both ends try to initiate a context for the same ULID pair, then we might end up with crossing I1 messages. Alternatively, since no state is created when receiving the I1, a host might send a I1 after having sent a R1 message. Since a host remembers that it has sent an I1, it can respond to an I1 from the peer (for the same ULID-pair), with a R2, resulting in the message exchange shown in Figure 25. Such behavior is needed for other reasons such as to correctly respond to retransmitted I1Nordmark & Bagnulo Expires September 5, 2006 [Page 53] Internet-Draft Shim6 Protocol March 2006messages, which occur when the R2 message has been lost. Nordmark & Bagnulo Expires November 16, 2006 [Page 54] Internet-Draft Shim6 Protocol May 2006 Host A Host B IDLE IDLE -\ I1-SENT---\ ---\ /--- --- I1 ---\ /--- I1-SENT ---\ /--- I1 ---/ ---\ /--- --> <--- -\ I1-SENT---\ ---\ /--- --- R2 ---\ /--- I1-SENT ---\ /--- R2 ---/ ---\ /--- --> <--- ESTABLISHED ESTABLISHED Figure 25: Crossing I1 messages If a host has received an I1 and sent an R1, it has no state to remember this. Thus if the ULP on the host sends down packets, this might trigger the host to send an I1 message itself. Thus while one end is sending an I1 the other is sending an I2 as can be seen in Figure 26. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page54]55] Internet-Draft Shim6 ProtocolMarchMay 2006 Host A Host B IDLE IDLE -\ ---\ I1-SENT ---\ --- I1 ---\ ---\ ---\ --> /--- /--- IDLE --- /--- R1--/ /--- <--- -\ I2-SENT---\ ---\ /--- --- I2---\ /--- I1-SENT ---\ /--- I1 ---/ ---\ /--- --> <--- ESTABLISHED -\ I2-SENT---\ ---\ /--- --- R2 ---\ /--- ---\ /--- R2 ---/ ---\ /--- --> <--- ESTABLISHED ESTABLISHED Figure 26: Crossing I2 and I17.57.5. Context recovery Due to garbage collection, we can end up with one end having and using the context state, and the other end not having any state. We need to be able to recover this state at the end that has lost it, before we can use it. This need can arise in the following cases: Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page55]56] Internet-Draft Shim6 ProtocolMarchMay 2006 o The communication is working using the ULID pair as the locator pair, but a problem arises, and the end that has retained the context state decides to probe alternate locator pairs. o The communication is working using a locator pair that is not the ULID pair, hence the ULP packets sent from a peer that has retained the context state use the shim6 Payload extension header. o The host that retained the state sends a control message (e.g. an Update Request message). In all the cases the result is that the peer without state receives a shim message for which it has to context for the context tag. In all of those cases we can recover the context by having the node which doesn't have a context state, send back an R1bis message, and have then complete the recovery with a I2bis and R2 message as can be seen in Figure 27. Host A Host B Context for CT(peer)=X Discards context for CT(local)=X ESTABLISHED IDLE ---- payload, probe, etc. -----> No context state for CT(local)=X <------------ R1bis ------------ IDLE ------------- I2bis -----------> I2BIS_SENT <------------ R2 --------------- ESTABLISHED ESTABLISHED Figure 27: Context loss at receiver If one end has garbage collected or lost the context state, it might try to create a new context state (for the same ULID pair), by sending an I1 message. The peer (that still has the context state) will reply with an R1 message and the full 4-way exchange will be performed again in this case as can be seen in Figure 28. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page56]57] Internet-Draft Shim6 ProtocolMarchMay 2006 Host A Host B Context for CT(peer)=X Discards context for ULIDs A1, B1 CT(local)=X ESTABLISHED IDLE Finds <------------ I1 --------------- Tries to setup existing for ULIDs A1, B1 context, but CT(peer) I1-SENT doesn't match ------------- R1 ---------------> Left old context in ESTABLISHED <------------ I2 --------------- Recreate context with new CT(peer) I2-SENT and Ls(peer). ESTABLISHED ------------- R2 --------------> ESTABLISHED ESTABLISHED Figure 28: Context loss at sender7.67.6. Context confusion Since each end might garbage collect the context state we can have the case when one end has retained the context state and tries to use it, while the other end has lost the state. We discussed this in the previous section on recovery. But for the same reasons, when one host retains context tag X as CT(peer) for ULID pair <A1, B1>, the other end might end up allocating that context tag as CT(local) for another ULID pair, e.g., <A3, B1> between the same hosts. In this case we can not use the recovery mechanisms since there needs to be separate context tags for the two ULID pairs. This type of "confusion" can be observed in two cases (assuming it is A that has retained the state and B has dropped it): o B decides to create a context for ULID pair <A3, B1>, and allocates X as its context tag for this, and sends an I1 to A. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page57]58] Internet-Draft Shim6 ProtocolMarchMay 2006 o A decides to create a context for ULID pair <A3, B1>, and starts the exchange by sending I1 to B. When B receives the I2 message, it allocates X as the context tag for this context. In both cases, A can detect that B has allocated X for ULID pair <A3, B1> even though that A still X as CT(peer) for ULID pair <A1, B1>. Thus A can detect that B must have lost the context for <A1, B1>. The confusion can be detected when I2/I2bis/R2 is received since we require that those messages MUST include a sufficiently large set of locators in a Locator List option that the peer can determine whether or not two contexts have the same host as the peer by comparing if there is any common locators in Ls(peer). The requirement is that the old context which used the context tag MUST be removed; it can no longer be used to send packets. Thus A would forcibly remove the context state for <A1, B1, X>, so that it can accept the new context for <A3, B1, X>. An implementation MAY re-create a context to replace the one that was removed; in this case for <A1, B1>. The normal I1, R1, I2, R2 establishment exchange would then pick unique context tags for that replacement context. This re- creation is OPTIONAL, but might be useful when there is ULP communication which is using the ULID pair whose context was removed. Note that an I1 message with a duplicate context tag should not cause the removal of the old context state; this operation needs to be deferred until the reception of the I2 message.7.77.7. Sending I1 messages When the shim layer decides to setup a context for a ULID pair, it starts by allocating and initializing the context state for its end. As part of this it assigns a random context tag to the context that is not being used as CT(local) by any other context . In the case that a new API is used and the ULP requests a forked context, the Forked Instance Identifier value will be set to a non-zero value. Otherwise, the FII value is zero. Then the initiator can send an I1 message and set the context state to I1-SENT. The I1 message MUST include the ULID pair; normally in the IPv6 source and destination fields. But if the ULID pair for the context is not used as locator pair for the I1 message, then a ULID option MUST be included in the I1 message. In addition, if a Forked Instance Identifier value is non-zero, the I1 message MUST include a Context Instance Identifier option containing the correspondent value.7.87.8. Retransmitting I1 messages If the host does not receive an I2 or R2 message in response to the Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page58]59] Internet-Draft Shim6 ProtocolMarchMay 2006 I1 message after I1_TIMEOUT time, then it needs to retransmit the I1 message. The retransmissions should use a retransmission timer with binary exponential backoff to avoid creating congestion issues for the network when lots of hosts perform I1 retransmissions. Also, the actual timeout value should be randomized between 0.5 and 1.5 of the nominal value to avoid self-synchronization. If, after I1_RETRIES_MAX retransmissions, there is no response, then most likely the peer does not implement the shim6 protocol, or there could be a firewall that blocks the protocol. In this case it makes sense for the host to remember to not try again to establish a context with that ULID. However, any such negative caching should retained for at most NO_R1_HOLDDOWN_TIME, to be able to later setup a context should the problem have been that the host was not reachable at all when the shim tried to establish the context. If the host receives an ICMP error with"payload"Unrecognized Next Header" typeunknown"(type 4, code 1) and the included packet is the I1 message it just sent, then this is a more reliable indication that the peer ULID does not implement shim6. Again, in this case, the host should remember to not try again to establish a context with that ULID. Such negative caching should retained for at most ICMP_HOLDDOWN_TIME, which should be significantly longer than the previous case.7.97.9. Receiving I1 messages A host MUST silently discard any received I1 messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an I1 message, the host extracts the ULID pair and the Forked Instance Identifier from the message. If there is no ULID-pair option, then the ULID pair is taken from the source and destination fields in the IPv6 header. If there is no FII option in the message, then the FII value is taken to be zero. Next the host looks for an existing context which matches the ULID pair and the FII. If no state is found (i.e., the state is IDLE), then the host replies with a R1 message as specified below. If such a context exists in ESTABLISHED state, the host verifies that the locator of the Initiator is included in Ls(peer) (This check is unnecessary if there is no ULID-pair option in the I1 message). Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page59]60] Internet-Draft Shim6 ProtocolMarchMay 2006 If the state exists in ESTABLISHED state and the locators do not fall in the locator sets, then the host replies with a R1 message as specified below. This completes the I1 processing, with the context state being unchanged. If the state exists in ESTABLISHED state and the locators do fall in the sets, then the host compares CT(peer) for the context with the CT contained in the I1 message. o If the context tags match, then this probably means that the R2 message was lost and this I1 is a retransmission. In this case, the host replies with a R2 message containing the information available for the existent context. o If the context tags do not match, then it probably means that the Initiator has lost the context information for this context and it is trying to establish a new one for the same ULID-pair. In this case, the host replies with a R1 message as specified below. This completes the I1 processing, with the context state being unchanged. If the state exists in other state (I1-SENT, I2-SENT, I2BIS-SENT), we are in the situation of Concurrent context establishment described in Section 7.4. In this case, the host leaves CT(peer) unchanged, and replies with a R2 message. This completes the I1 processing, with the context state being unchanged. 7.10. Sending R1 messages When the host needs to send a R1 message in response to the I1 message, it copies the Initiator Nonce from the I1 message to the R1 message, generates a Responder Nonce and calculates a Responder Validator option as suggested in the following section. No state is created on the host in thiscase. Whencase.(Note that thehost needsinformation used to generate the R1 reply message is either contained in the received I1 message or it is global information that is not associated with the particular requested context (the S and the Responder nonce values)). When the host needs to send a R2 message in response to the I1 message, it copies the Initiator Nonce from the I1 message to the R2 message, and otherwise follows the normal rules for forming an R2 message (see Section7.13). 7.9.17.14). 7.10.1. Generating the R1 Validator One way for the responder to properly generate validators is to maintain a single secret (S) and a running counter for the Responder Nonce. Nordmark & Bagnulo Expires November 16, 2006 [Page 61] Internet-Draft Shim6 Protocol May 2006 In the case the validator is generated to be included in a R1 message, for each I1 message. The responder can increase the counter, use the counter value as the responder nonce, and use the following information as input to the one-way function:Nordmark & Bagnulo Expires September 5, 2006 [Page 60] Internet-Draft Shim6 Protocol March 2006o The the secret S o That Responder Nonce o The Initiator Context Tag from the I1 message o The ULIDs from the I1 message o The locators from the I1 message (strictly only needed if they are different from the ULIDs) o The forked instance identifier if such option was included in the I1 message and then the output of the hash function is used as the validator octet string.7.107.11. Receiving R1 messages and sending I2 messages A host MUST silently discard any received R1 messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an R1 message, the host extracts the Initiator Nonce and the Locator Pair from the message (the latter from the source and destination fields in the IPv6 header). Next the host looks for an existing context which matches the Initiator Nonce and where the locators are contained in Ls(peer) and Ls(local), respectively. If no such context is found, then the R1 message is silently discarded. If such a context is found, then the host looks at the state: o If the state is I1-SENT, then it sends an I2 message as specified below. o In any other state (I2-SENT, I2BIS-SENT, ESTABLISHED) then the host has already sent an I2 message then this is probably a reply to a retransmitted I1 message, so this R1 message MUST be silently discarded. Nordmark & Bagnulo Expires November 16, 2006 [Page 62] Internet-Draft Shim6 Protocol May 2006 When the host sends an I2 message, then it includes the Responder Validator option that was in the R1 message. The I2 message MUST include the ULID pair; normally in the IPv6 source and destination fields. If a ULID-pair option was included in the I1 message then itNordmark & Bagnulo Expires September 5, 2006 [Page 61] Internet-Draft Shim6 Protocol March 2006MUST be included in the I2 message as well. In addition, if the Forked Instance Identifier value for this context is non-zero, the I2 message MUST contain a Forked Instance Identifier Option carrying this value. Besides, the I2 message contains an Initiator Nonce. This is not required to be the same than the one included in the previous I1 message. The I2 message also includes the Initiator's locator list and the CGA parameter data structure. If CGA (and not HBA) is used to verify the locator list, then Initiator also signs the key parts of the message and includes a CGA signature option containing the signature. When the I2 message has been sent, the state is set to I2-SENT.7.117.12. Retransmitting I2 messages If the initiator does not receive an R2 message after I2_TIMEOUT time after sending an I2 message it MAY retransmit the I2 message, using binary exponential backoff and randomized timers. The Responder Validator option might have a limited lifetime, that is, the peer might reject Responder Validator options that are older than VALIDATOR_MIN_LIFETIME to avoid replay attacks. Thus the initiator SHOULD fall back to retransmitting the I1 message when there is no R2 received after retransmitting the I2 message I2_RETRIES_MAX times.7.127.13. Receiving I2 messages A host MUST silently discard any received I2 messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 2, i.e., the length is at least 24 octets. Upon the reception of an I2 message, the host extracts the ULID pair and the Forked Instance identifier from the message. If there is no ULID-pair option, then the ULID pair is taken from the source and destination fields in the IPv6 header. If there is no FII option in the message, then the FII value is taken to be zero. Next the host verifies that the Responder Nonce is a recentone,one (Nonces that are no older than VALIDATOR_MIN_LIFETIME SHOULD be considered recent), and that the Responder Validator option matches the validator the host would have computed for the ULID, locators, Nordmark & Bagnulo Expires November 16, 2006 [Page 63] Internet-Draft Shim6 Protocol May 2006 responder nonce, and FII. If a CGA Parameter Data Structure (PDS) is included in the message, then the host MUST verify if the actual PDS contained in the message corresponds to the ULID(peer).Nordmark & Bagnulo Expires September 5, 2006 [Page 62] Internet-Draft Shim6 Protocol March 2006If any of the above verifications fails, then the host silentlydiscarddiscards the message and it has completed the I2 processing. If all the above verifications are successful, then the host proceeds to look for a context state for the Initiator. The host looks for a context with the extracted ULID pair and FII. If none exist then state of the (non-existing) context is viewed as being IDLE, thus the actions depend on the state as follows: o If the state is IDLE (i.e., the context does not exist) the host allocates a context tag (CT(local)), creates the context state for the context, and sets its state to ESTABLISHED. It records CT(peer), and the peer's locator set as well as its own locator set in the context. It SHOULD perform the HBA/CGA verification of the peer's locator set at this point in time, as specified in Section 7.2. Then the host sends an R2 message back as specified below. o If the state is I1-SENT, then the host verifies if the source locator is included in Ls(peer) or, it is included in the Locator List contained in the the I2 message and the HBA/CGA verification for this specific locator is successful * If this is not the case, then the message is silently discarded and the context state remains unchanged. * If this is the case, then the host updates the context information (CT(peer), Ls(peer)) with the data contained in the I2 message and the host MUST send a R2 message back as specified below. Note that before updating Ls(peer) information, the host SHOULD perform the HBA/CGA validation of the peer's locator set at this point in time as specified in Section 7.2. The host moves to ESTABLISHED state. o If the state is ESTABLISHED, I2-SENT, or I2BIS-SENT, then the host verifies if the source locator is included in Ls(peer) or, it is included in the Locator List contained in the the I2 message and the HBA/CGA verification for this specific locator is successful * If this is not the case, then the message is silently discarded and the context state remains unchanged. Nordmark & Bagnulo Expires November 16, 2006 [Page 64] Internet-Draft Shim6 Protocol May 2006 * If this is the case, then the host updates the context information (CT(peer), Ls(peer)) with the data contained in the I2 message and the host MUST send a R2 message back as specified in Section7.13.7.14. Note that before updating Ls(peer) information, the host SHOULD perform the HBA/CGA validation of the peer's locator set at this point in time as specified inNordmark & Bagnulo Expires September 5, 2006 [Page 63] Internet-Draft Shim6 Protocol March 2006Section 7.2. The context state remains unchanged.7.137.14. Sending R2 messages Before the host sends the R2 message it MUST look for a possible context confusion i.e. where it would end up with multiple contexts using the same CT(peer) for the same peer host. See Section7.14.7.15. When the host needs to send an R2 message, the host forms the message using its locators and its context tag, copies the Initiator Nonce from the triggering message (I2, I2bis, or I1), and includes the necessary options so that the peer can verify the locators. In particular, the R2 message includes the Responder's locator list and the PDS option. If CGA (and not HBA) is used to verify the locator list, then the Responder also signs the key parts of the message and includes a CGA Signature option containing the signature. R2 messages are never retransmitted. If the R2 message is lost, then the initiator will retransmit either the I2/I2bis or I1 message. Either retransmission will cause the responder to find the context state and respond with an R2 message.7.147.15. Match for Context Confusion When the host receives an I2, I2bis, or R2 it MUST look for a possible context confusion i.e. where it would end up with multiple contexts using the same CT(peer) for the same peer host. This can happen when it has received the above messages since they create a new context with a new CT(peer). Same issue applies when CT(peer) is updated for an existing context. The host takes CT(peer) for the newly created or updated context, and looks for other contexts which: o Are in state ESTABLISHED or I2BIS-SENT. o Have the same CT(peer). o Where Ls(peer) has at least one locator in common with the newly created or updated context. If such a context is found, then the host checks if the ULID pair or Nordmark & Bagnulo Expires November 16, 2006 [Page 65] Internet-Draft Shim6 Protocol May 2006 the Forked Instance Identifier different than the ones in the newly created or updated context: o If either or both are different, then the peer is reusing the context tag for the creation of a context with different ULID pairNordmark & Bagnulo Expires September 5, 2006 [Page 64] Internet-Draft Shim6 Protocol March 2006or FII, which is an indication that the peer has lost the original context. In this case, we are in the Context confusion situation, and the host MUST NOT use the old context to send any packets. It MAY just discard the old context (after all, the peer has discarded it), or it MAY attempt to re-establish the old context by sending a new I1 message and moving its state to I1-SENT. In any case, once that this situation is detected, the host MUST NOT keep two contexts with overlapping Ls(peer) locator sets and the same context tag in ESTABLISHED state, since this would result in demultiplexing problems on the peer. o If both are the same, then this context is actually the context that is created or updated, hence there is no confusion.7.157.16. Receiving R2 messages A host MUST silently discard any received R2 messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an R2 message, the host extracts the Initiator Nonce and the Locator Pair from the message (the latter from the source and destination fields in the IPv6 header). Next the host looks for an existing context which matches the Initiator Nonce and where the locators are Lp(peer) and Lp(local), respectively. Based on the state: o If no such context is found, i.e., the state is IDLE, then the message is silently dropped. o If state is I1-SENT, I2-SENT, or I2BIS-SENT then the host performs the following actions: If a CGA Parameter Data Structure (PDS) is included in the message, then the host MUST verify that the actual PDS contained in the message corresponds to the ULID(peer) as specified in Section 7.2. If the verification fails, then the message is silently dropped. If the verification succeeds, then the host records the information from the R2 message in the context state; it records the peer's locator set and CT(peer). The host SHOULD perform the HBA/CGA verification of the peer's locator set at this point in time, as specified in Section 7.2. Nordmark & Bagnulo Expires November 16, 2006 [Page 66] Internet-Draft Shim6 Protocol May 2006 The host sets its state to ESTABLISHED. o If the state is ESTABLISHED, the R2 message is silently ignored, (since this is likely to be a reply to a retransmitted I2Nordmark & Bagnulo Expires September 5, 2006 [Page 65] Internet-Draft Shim6 Protocol March 2006message). Before the host completes the R2 processing it MUST look for a possible context confusion i.e. where it would end up with multiple contexts using the same CT(peer) for the same peer host. See Section7.14. 7.167.15. 7.17. Sending R1bis messages Upon the receipt of a shim6 payload extension header where there is no current SHIM6 context at the receiver, the receiver is to respond with an R1bis message in order to enable a fast re-establishment of the lost SHIM6 context. Also a host is to respond with a R1bis upon receipt of any control messages that has a message type in the range 64-127 (i.e., excluding the context setup messages such as I1, R1, R1bis, I2, I2bis, R2 and future extensions), where the control message refers to a non existent context. We assume that all the incoming packets that trigger the generation of an R1bis message contain a locator pair (in the address fields of the IPv6 header) and a Context Tag. Upon reception of any of the packets described above, the host will reply with an R1bis including the following information: o The Responder Nonce is a number picked by the responder which the initiator will return in the I2bis message. o Packet Context Tag is the context tag contained in the received packet that triggered the generation of the R1bis message. o The Responder Validator option is included, with a validator that is computed as suggested in the next section.7.16.17.17.1. Generating the R1bis Validator One way for the responder to properly generate validators is to maintain a single secret (S) and a running counter for the Responder Nonce. In the case the validator is generated to be included in a R1bis message, for each received payload extension header or control Nordmark & Bagnulo Expires November 16, 2006 [Page 67] Internet-Draft Shim6 Protocol May 2006 message, the responder can increase the counter, use the counter value as the responder nonce, and use the following information as input to the one-way function:Nordmark & Bagnulo Expires September 5, 2006 [Page 66] Internet-Draft Shim6 Protocol March 2006o The the secret S o That Responder Nonce o The Receiver Context tag included in the received packet o The locators from the received packet and then the output of the hash function is used as the validator octet string.7.177.18. Receiving R1bis messages and sending I2bis messages A host MUST silently discard any received R1bis messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an R1bis message, the host extracts the Packet Context Tag and the Locator Pair from the message (the latter from the source and destination fields in the IPv6 header). Next the host looks for an existing context where the Packet Context Tag matches CT(peer) and where the locators match Lp(peer) and Lp(local), respectively. o If no such context is not found, i.e., the state is IDLE, then the R1bis message is silently discarded. o If the state is I1-SENT, I2-SENT, or I2BIS-SENT, then the R1bis message is silently discarded. o If the state is ESTABLISHED, then we are in the case where the peer has lost the context and the goal is to try to re-establish it. For that, the host leaves CT(peer) unchanged in the context state, transitions to I2BIS-SENT state, and sends a I2bis message, including the computed Responder Validator option, the Packet Context Tag, and the Responder Nonce received in the R1bis message. This I2bis message is sent using the locator pair included in the R1bis message. In the case that this locator pair differs from the ULID pair defined for this context, then an ULID option MUST be included in the I2bis message. In addition, if the Forked Instance Identifier for this context is non-zero, then a Nordmark & Bagnulo Expires November 16, 2006 [Page 68] Internet-Draft Shim6 Protocol May 2006 Forked Instance Identifier option carrying the instance identifier value for this context MUST be included in the I2bis message.Nordmark & Bagnulo Expires September 5, 2006 [Page 67] Internet-Draft Shim6 Protocol March 2006 7.187.19. Retransmitting I2bis messages If the initiator does not receive an R2 message after I2bis_TIMEOUT time after sending an I2bis message it MAY retransmit the I2bis message, using binary exponential backoff and randomized timers. The Responder Validator option might have a limited lifetime, that is, the peer might reject Responder Validator options that are older than VALIDATOR_MIN_LIFETIME to avoid replay attacks. Thus the initiator SHOULD fall back to retransmitting the I1 message when there is no R2 received after retransmitting the I2bis message I2bis_RETRIES_MAX times.7.197.20. Receiving I2bis messages and sending R2 messages A host MUST silently discard any received I2bis messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 3, i.e., the length is at least 32 octets. Upon the reception of an I2bis message, the host extracts the ULID pair and the Forked Instance identifier from the message. If there is no ULID-pair option, then the ULID pair is taken from the source and destination fields in the IPv6 header. If there is no FII option in the message, then the FII value is taken to be zero. Next the host verifies that the Responder Nonce is a recentone,one (Nonces that are no older than VALIDATOR_MIN_LIFETIME SHOULD be considered recent), and that the Responder Validator option matches the validator the host would have computed for the ULID, locators, responder nonce, and FII as part of sending an R1bis message. If a CGA Parameter Data Structure (PDS) is included in the message, then the host MUST verify if the actual PDS contained in the message corresponds to the ULID(peer). If any of the above verifications fails, then the host silently discard the message and it has completed the I2bis processing. If both verifications are successful, then the host proceeds to look for a context state for the Initiator. The host looks for a context with the extracted ULID pair and FII. If none exist then state of the (non-existing) context is viewed as being IDLE, thus the actions depend on the state as follows: Nordmark & Bagnulo Expires November 16, 2006 [Page 69] Internet-Draft Shim6 Protocol May 2006 o If the state is IDLE (i.e., the context does not exist) the host allocates a context tag (CT(local)), creates the context state for the context, and sets its state to ESTABLISHED. The host SHOULDNordmark & Bagnulo Expires September 5, 2006 [Page 68] Internet-Draft Shim6 Protocol March 2006NOT use the Packet Context Tag in the I2bis message for CT(local); instead it should pick a new random context tag just as when it processes an I2 message. It records CT(peer), and the peer's locator set as well as its own locator set in the context. It SHOULD perform the HBA/CGA verification of the peer's locator set at this point in time as specified in Section 7.2. Then the host sends an R2 message back as specified in Section7.13.7.14. o If the state is I1-SENT, then the host verifies if the source locator is included in Ls(peer) or, it is included in the Locator List contained in the the I2 message and the HBA/CGA verification for this specific locator is successful * If this is not the case, then the message is silently discarded. The the context state remains unchanged. * If this is the case, then the host updates the context information (CT(peer), Ls(peer)) with the data contained in the I2 message and the host MUST send a R2 message back as specified below. Note that before updating Ls(peer) information, the host SHOULD perform the HBA/CGA validation of the peer's locator set at this point in time as specified in Section 7.2. The host moves to ESTABLISHED state. o If the state is ESTABLISHED, I2-SENT, or I2BIS-SENT, then the host verifies if the source locator is included in Ls(peer) or, it is included in the Locator List contained in the the I2 message and the HBA/CGA verification for this specific locator is successful * If this is not the case, then the message is silently discarded. The the context state remains unchanged. * If this is the case, then the host updates the context information (CT(peer), Ls(peer)) with the data contained in the I2 message and the host MUST send a R2 message back as specified in Section7.13.7.14. Note that before updating Ls(peer) information, the host SHOULD perform the HBA/CGA validation of the peer's locator set at this point in time as specified in Section 7.2. The context state remains unchanged. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page69]70] Internet-Draft Shim6 ProtocolMarchMay 2006 8. Handling ICMP Error Messages The routers in the path as well as the destination might generate various ICMP error messages, such as host unreachable, packet too big, andpayload type unknown.Unrecognized Next Header type. It is critical that these packets make it back up to the ULPs so that they can take appropriate action. This is an implementation issue in the sense that the mechanism is completely local to the host itself. But the issue of how ICMP errors are correctly dispatched to the ULP on the host are important, hence this section specifies the issue. +--------------+ | IPv6 Header | | | +--------------+ | ICMPv6 | | Header | - - +--------------+ - - | IPv6 Header | | src, dst as | Can be dispatched IPv6 | sent by ULP | unmodified to ULP | on host | ICMP error handler Packet +--------------+ | ULP | in | Header | +--------------+ Error | | ~ Data ~ | | - - +--------------+ - - Figure 29: ICMP error handling without payload extension header When the ULP packets are sent without the payload extension header, that is, while the initial locators=ULIDs are working, this introduces no new concerns; an implementation's existing mechanism for delivering these errors to the ULP will work. See Figure 29. But when the shim on the transmitting side inserts the payload extension header and replaces the ULIDs in the IP address fields with some other locators, then an ICMP error coming back will have a "packet in error" which is not a packet that the ULP sent. Thus the implementation will have to apply the reverse mapping to the "packet in error" before passing the ICMP error up to the ULP. See Figure 30. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page70]71] Internet-Draft Shim6 ProtocolMarchMay 2006 +--------------+ | IPv6 Header | | | +--------------+ | ICMPv6 | | Header | - - +--------------+ - - | IPv6 Header | | src, dst as | Needs to be IPv6 | modified by | transformed to | shim on host | have ULIDs +--------------+ in src, dst fields, Packet | SHIM6 ext. | and SHIM6 ext. | Header | header removed in +--------------+ before it can be | Transport | dispatched to the ULP Error | Header | ICMP error handler. +--------------+ | | ~ Data ~ | | - - +--------------+ - - Figure 30: ICMP error handling with payload extension header Note that this mapping is different than when receiving packets from the peer with a payload extension headers, because in that case the packets contain CT(local). But the ICMP errors have a "packet in error" with an payload extension header containing CT(peer). This is because they were intended to be received by the peer. In any case, since the <Source Locator, Destination Locator, CT(peer)> has to be unique when received by the peer, the local host should also only be able to find one context that matches this tuple. If the ICMP error is a Packet Too Big, the reported MTU must be adjusted to be 8 octets less, since the shim will add 8 octets when sending packets. After the "packet in error" has had the original ULIDs inserted, then this payload extension header can be removed. The result is a "packet in error" that is passed to the ULP which looks as if the shim did not exist. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page71]72] Internet-Draft Shim6 ProtocolMarchMay 2006 9. Teardown of the ULID-Pair Context Each host can unilaterally decide when to tear down a ULID-pair context. It is RECOMMENDED that hosts do not tear down the context when they know that there is some upper layer protocol that might use the context. For example, an implementation might know this if there is an open socket which is connected to the ULID(peer). However, there might be cases when the knowledge is not readily available to the shim layer, for instance for UDP applications which do not connect their sockets, or any application which retains some higher level state across (TCP) connections and UDP packets. Thus it is RECOMMENDED that implementations minimize premature teardown by observing the amount of traffic that is sent and received using the context, and only after it appears quiescent, tear down the state. A reasonable approach would be not to tear down a context until at least 5 minutes have passed since the last message was sent or received using the context. Since there is no explicit, coordinated removal of the context state, there are potential issues around context tag reuse. One end might remove the state, and potentially reuse that context tag for some other communication, and the peer might later try to use the old context (which it didn't remove). The protocol has mechanisms to recover from this, which work whether the state removal was total and accidental (e.g., crash and reboot of the host), or just a garbage collection of shim state that didn't seem to be used. However, the host should try to minimize the reuse of context tags by trying to randomly cycle through the 2^47 context tag values. (See AppendixEC for a summary how the recovery works in the different cases.) Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page72]73] Internet-Draft Shim6 ProtocolMarchMay 2006 10. Updating the Peer The Update Request and Acknowledgement are used both to update the list of locators (only possible when CGA is used to verify the locator(s)), as well as updating the preferences associated with each locator.10.110.1. Sending Update Request messages When a host has a change in the locator set, then it can communicate this to the peer by sending an Update Request. When a host has a change in the preferences for its locator set, it can also communicate this to the peer. The Update Request message can include just a Locator List option, to convey the new set of locators (which requires a CGA signature option as well), just a Locator Preferences option, or both a new Locator List and new Locator Preferences. Should the host send a new Locator List, the host picks a new random local generation number, records this in the context, and puts it in the Locator List option. Any Locator Preference option, whether send in the same Update Request or in some future Update Request, will use that generation number to make sure the preferences get applied to the correct version of the locator list. The host picks a random Request Nonce for each update, and keeps the same nonce for any retransmissions of the Update Request. The nonce is used to match the acknowledgement with the request.10.210.2. Retransmitting Update Request messages If the host does not receive an Update Acknowledgement R2 message in response to the Update Request message after UPDATE_TIMEOUT time, then it needs to retransmit the Update Request message. The retransmissions should use a retransmission timer with binary exponential backoff to avoid creating congestion issues for the network when lots of hosts perform Update Request retransmissions. Also, the actual timeout value should be randomized between 0.5 and 1.5 of the nominal value to avoid self-synchronization. Should there be no response, the retransmissions continue forever. The binary exponential backoff stops at MAX_UPDATE_TIMEOUT. But the only way the retransmissions would stop when there is no acknowledgement, is when the shim, through the Probe protocol or some other mechanism, decides to discard the context state due to lack of ULP usage in combination with no responses to the Probes. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page73]74] Internet-Draft Shim6 ProtocolMarchMay 200610.310.3. Newer Information While Retransmitting There can be at most one outstanding Update Request message at any time. Thus until e.g. an update with a new Locator List has been acknowledged, any even newer Locator List or new Locator Preferences can not just be sent. However, when there is newer information and the older information has not yet been acknowledged, the host can instead of waiting for an acknowledgement, abandon the previous update and construct a new Update Request (with a new Request Nonce) which includes the new information as well as the information that hadn't yet been acknowledged. For example, if the original locator list was just (A1, A2), and if an Update Request with the Locator List (A1, A3) is outstanding, and the host determines that it should both add A4 to the locator list, and mark A1 as BROKEN, then it would need to: o Pick a new random Request Nonce for the new Update Request. o Pick a new random Generation number for the new locator list. o Form the new locator list - (A1, A3, A4) o Form a Locator Preference option which uses the new generation number and has the BROKEN flag for the first locator. o Send the Update Request and start a retransmission timer. Any Update Acknowledgement which doesn't match the current request nonce, for instance an acknowledgement for the abandoned Update Request, will be silently ignored.10.410.4. Receiving Update Request messages A host MUST silently discard any received Update Request messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an Update Request message, the host extracts the Context Tag from the message. It then looks for a context which has a CT(local) that matches the context tag. If no such context is found, it sends a R1bis message as specified in Section7.16.7.17. Since context tags can be reused, the host MUST verify that the IPv6 source address field is part of Ls(peer) and that the IPv6 Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page74]75] Internet-Draft Shim6 ProtocolMarchMay 2006 destination address field is part of Ls(local). If this is not the case, the sender of the Update Request has a stale context which happens to match the CT(local) for this context. In this case the host MUST send a R1bis message, and otherwise ignore the Update Request message. If a CGA Parameter Data Structure (PDS) is included in the message, then the host MUST verify if the actual PDS contained in the packet corresponds to the ULID(peer). If this verification fails, the message is silently discarded. Then, depending on the state of the context: o If ESTABLISHED: Proceed to process message. o If I1-SENT, discard the message and stay in I1-SENT. o If I2-SENT, then send R2 and proceed to process the message. o If I2BIS-SENT, then send R2 and proceed to process the message. The verification issues for the locators carried in the Locator Update message are specified in Section 7.2. If the locator list can not be verified, this procedure might send an ICMP Parameter Problem error. In any case, if it can not be verified, there is no further processing of the Update Request. Once any Locator List option in the Update Request has been verified, the peer generation number in the context is updated to be the one in the Locator List option. If the Update message contains a Locator Preference option, then the Generation number in the preference option is compared with the peer generation number in the context. If they do not match, then the host generates an ICMP parameter problem (type 4, code 0) with the Pointer field referring to the first octet in the Generation number in the Locator Preference option. In addition, if the number of elements in the Locator Preference option does not match the number of locators in Ls(peer), then an ICMP parameter problem is sent with the Pointer referring to the first octet of the Length field in the Locator Preference option. In both cases of failures, no further processing is performed for the Locator Update message. If the generation number matches, the locator preferences are recorded in the context. Once the Locator List option (if present) has been verified and any new locator list or locator preferences have been recorded, the host Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page75]76] Internet-Draft Shim6 ProtocolMarchMay 2006 sends an Update Acknowledgement message, copying the nonce from the request, and using the CT(peer) in as the Receiver Context Tag. Any new locators, or more likely new locator preferences, might result in the host wanting to select a different locator pair for the context. For instance, if the Locator Preferences lists the current Lp(peer) as BROKEN. The host uses the Probe message in [8] to verify that the new locator is reachable before changing Lp(peer).10.510.5. Receiving Update Acknowledgement messages A host MUST silently discard any received Update Acknowledgement messages that do not satisfy all of the following validity checks in addition to those specified in Section 12.2: o The Hdr Ext Len field is at least 1, i.e., the length is at least 16 octets. Upon the reception of an Update Acknowledgement message, the host extracts the Context Tag and the Request Nonce from the message. It then looks for a context which has a CT(local) that matches the context tag. If no such context is found, it sends a R1bis message as specified in Section7.16.7.17. Since context tags can be reused, the host MUST verify that the IPv6 source address field is part of Ls(peer) and that the IPv6 destination address field is part of Ls(local). If this is not the case, the sender of the Update Acknowledgement has a stale context which happens to match the CT(local) for this context. In this case the host MUST send a R1bis message, and otherwise ignore the Update Acknowledgement message. Then, depending on the state of the context: o If ESTABLISHED: Proceed to process message. o If I1-SENT, discard the message and stay in I1-SENT. o If I2-SENT, then send R2 and proceed to process the message. o If I2BIS-SENT, then send R2 and proceed to process the message. If the Request Nonce doesn't match the Nonce for the last sent Update Request for the context, then the Update Acknowledgement is silently ignored. If the nonce matches, then the update has been completed and the Update retransmit timer can be reset. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page76]77] Internet-Draft Shim6 ProtocolMarchMay 2006 11. Sending ULP Payloads When there is no context state for the ULID pair on the sender, there is no effect on how ULP packets are sent. If the host is using some heuristic for determining when to perform a deferred context establishment, then the host might need to do some accounting (count the number of packets sent and received) even before there is a ULID- pair context. If the context is not in ESTABLISHED or I2BIS-SENT state, then it there is also no effect on how the ULP packets are sent. Only in the ESTABLISHED and I2BIS-SENT states does the host have CT(peer) and Ls(peer) set. If there is a ULID-pair context for the ULID pair, then the sender needs to verify whether context uses the ULIDs as locators, that is, whether Lp(peer) == ULID(peer) and Lp(local) == ULID(local). If this is the case, then packets can be sent unmodified by the shim. If it is not the case, then the logic in Section 11.1 will need to be used. There will also be some maintenance activity relating to (un)reachability detection, whether packets are sent with the original locators or not. The details of this is out of scope for this document and is specified in [8].11.111.1. Sending ULP Payload after a Switch When sending packets, if there is a ULID-pair context for the ULID pair, and the ULID pair is no longer used as the locator pair, then the sender needs to transform the packet. Apart from replacing the IPv6 source and destination fields with a locator pair, an 8-octet header is added so that the receiver can find the context and inverse the transformation. If there has been a failure causing a switch, and later the context switches back to sending things using the ULID pair as the locator pair, then there is no longer a need to do any packet transformation by the sender, hence there is no need to include the 8-octet extension header. First, the IP address fields are replaced. The IPv6 source address field is set to Lp(local) and the destination address field is set to Lp(peer). NOTE that this MUST NOT cause any recalculation of the ULP checksums, since the ULP checksums are carried end-to-end and the ULP pseudo-header contains the ULIDs which are preserved end-to-end. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page77]78] Internet-Draft Shim6 ProtocolMarchMay 2006 The sender skips any "routing sub-layer extension headers" that the ULP might have included, thus it skips any hop-by-hop extension header, any routing header, and any destination options header that is followed by a routing header. After any such headers the shim6 extension header will be added. This might be before a Fragment header, a Destination Options header, an ESP or AH header, or a ULP header. The inserted shim6 Payload extension header includes the peer's context tag. It takes on the next header value from the preceding extension header, since that extension header will have a next header value of SHIM6. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page78]79] Internet-Draft Shim6 ProtocolMarchMay 2006 12. Receiving Packets As in normal IPv6 receive side packet processing the receiver parses the (extension) headers in order. Should it find a shim6 extension header it will look at the "P" field in that header. If this bit is zero, then the packet must be passed to the shim6 payload handling for rewriting. Otherwise, the packet is passed to the shim6 control handling.12.112.1. Receiving Payload Extension Headers The receiver extracts the context tag from the payload extension header, and uses this to find a ULID-pair context. If no context is found, the receiver SHOULD generate a R1bis message (see Section7.16).7.17). Then, depending on the state of the context: o If ESTABLISHED: Proceed to process message. o If I1-SENT, discard the message and stay in I1-SENT. o If I2-SENT, then send R2 and proceed to process the message. o If I2BIS-SENT, then send R2 and proceed to process the message. With the context in hand, the receiver can now replace the IP address fields with the ULIDs kept in the context. Finally, the Payload extension header is removed from the packet (so that the ULP doesn't get confused by it), and the next header value in the preceding header is set to be the actual protocol number for the payload. Then the packet can be passed to the protocol identified by the next header value (which might be some function associated with the IP endpoint sublayer, or a ULP). If the host is using some heuristic for determining when to perform a deferred context establishment, then the host might need to do some accounting (count the number of packets sent and received) for packets that does not have a shim6 extension header and for which there is no context. But the need for this depends on what heuristics the implementation has chosen.12.212.2. Receiving Shim Control messages A shim control message has the checksum field verified. The Shim header length field is also verified against the length of the IPv6 packet to make sure that the shim message doesn't claim to end past the end of the IPv6 packet. Finally, it checks that the neither the Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page79]80] Internet-Draft Shim6 ProtocolMarchMay 2006 IPv6 destination field nor the IPv6 source field is a multicast address. If any of those checks fail, the packet is silently dropped. The message is then dispatched based on the shim message type. Each message type is then processed as described elsewhere in this document. If the packet contains a shim message type which is unknown to the receiver, then an ICMPv6 Parameter Problem error is generated and sent back. The pointer field in the Parameter Problem is set to point at the first octet of the shim message type. The error is rate limited just like other ICMP errors [5]. All the control messages can contain any options with C=0. If there is any option in the message with C=1 that isn't known to the host, then the host MUST send an ICMPv6 Parameter Problem, with the Pointer field referencing the first octet of the Option Type.12.312.3. Context Lookup We assume that each shim context has its own state machine. We assume that a dispatcher delivers incoming packets to the state machine that it belongs to. Here we describe the rules used for the dispatcher to deliver packets to the correct shim context state machine. There is one state machine per context identified that is conceptually identified by ULID pair and Forked Instance Identifier (which is zero by default), or identified by CT(local). However, the detailed lookup rules are more complex, especially during context establishment. Clearly, if the required context is not established, it will be in IDLE state. During context establishment, the context is identified as follows: o I1 packets: Deliver to the context associated with the ULID pair and the Forked Instance Identifier. o I2 packets: Deliver to the context associated with the ULID pair and the Forked Instance Identifier. o R1 packets: Deliver to the context with the locator pair included in the packet and the Initiator nonce included in the packet (R1 does not contain ULID pair nor the CT(local)). If no context exist with this locator pair and Initiator nonce, then silently Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page80]81] Internet-Draft Shim6 ProtocolMarchMay 2006 discard. o R2 packets: Deliver to the context with the locator pair included in the packet and the Initiator nonce included in the packet (R2 does not contain ULID pair nor the CT(local)). If no context exists with this locator pair and INIT nonce, then silently discard. o R1bis packet: deliver to the context that has the locator pair and the CT(peer) equal to the Packet Context Tag included in the R1bis packet. o I2bis packets: Deliver to the context associated with the ULID pair and the Forked Instance Identifier. o Payload extension headers: Deliver to the context with CT(local) equal to the Receiver Context Tag included in the packet. o Other control messages (Update, Keepalive, Probe): Deliver to the context with CT(local) equal to the Receiver Context Tag included in the packet. Verify that the IPv6 source address field is part of Ls(peer) and that the IPv6 destination address field is part of Ls(local). If not, send a R1bis message. o ICMP errors which contain a shim6 payload extension header or other shim control packet in the "packet in error": Use the "packet in error" for dispatching as follows. Deliver to the context with CT(peer) equal to the Receiver Context Tag, Lp(local) being the IPv6 source address, and Lp(peer) being the IPv6 destination address. In addition, the shim on the sending side needs to be able to find the context state when a ULP packet is passed down from the ULP. In that case the lookup key is the pair of ULIDs and FII=0. If we have a ULP API that allows the ULP to do context forking, then presumably the ULP would pass down the Forked Instance Identifier. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page81]82] Internet-Draft Shim6 ProtocolMarchMay 2006 13. Initial Contact The initial contact is some non-shim communication between two ULIDs, as described in Section 2. At that point in time there is no activity in the shim. Whether the shim ends up being used or not (e.g., the peer might not support shim6) it is highly desirable that the initial contact can be established even if there is a failure for one or more IP addresses. The approach taken is to rely on the applications and the transport protocols to retry with different source and destination addresses, consistent with what is already specified in Default Address Selection [12], and some fixes to that specification [13] to make it try different source addresses and not only different destination addresses. The implementation of such an approach can potentially result in long timeouts. For instance, a naive implementation at the socket API which uses getaddrinfo() to retrieve all destination addresses and then tries to bind() and connect() to try all source and destination address combinations waiting for TCP to time out for each combination before trying the next one. However, if implementations encapsulate this in some new connect-by- name() API, and use non-blocking connect calls, it is possible to cycle through the available combinations in a more rapid manner until a working source and destination pair is found. Thus the issues in this domain are issues of implementations and the current socket API, and not issues of protocol specification. In all honesty, while providing an easy to use connect-by-name() API for TCP and other connection-oriented transports is easy; providing a similar capability at the API for UDP is hard due to the protocol itself not providing any "success" feedback. But even the UDP issue is one of APIs and implementation. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page82]83] Internet-Draft Shim6 ProtocolMarchMay 2006 14. Protocol constants The protocol uses the following constants: I1_RETRIES_MAX = 4 I1_TIMEOUT = 4 seconds NO_R1_HOLDDOWN_TIME = 1 min ICMP_HOLDDOWN_TIME = 10 min I2_TIMEOUT = 4 seconds I2_RETRIES_MAX = 2 I2bis_TIMEOUT = 4 seconds I2bis_RETRIES_MAX = 2 VALIDATOR_MIN_LIFETIME = 30 seconds UPDATE_TIMEOUT = 4 seconds The retransmit timers (I1_TIMEOUT, I2_TIMEOUT, UPDATE_TIMEOUT) are subject to binary exponential backoff, as well as randomization across a range of 0.5 and 1.5 times the nominal (backed off) value. This removes any risk of synchronization between lots of hosts performing independent shim operations at the same time. The randomization is applied after the binary exponential backoff. Thus the first retransmission would happen based on a uniformly distributed random number in the range [0.5*4, 1.5*4] seconds, the second retransmission [0.5*8, 1.5*8] seconds after the first one, etc. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page83]84] Internet-Draft Shim6 ProtocolMarchMay 2006 15. Implications Elsewhere The general shim6 approach, as well as the specifics of this proposed solution, has implications elsewhere. The key implications are: o Applications that perform referrals, or callbacks using IP addresses as the 'identifiers' can still function in limited ways, as described in [22]. But in order for such applications to be able to take advantage of the multiple locators for redundancy, the applications need to be modified to either use fully qualified domain names as the 'identifiers', or they need to pass all the locators as the 'identifiers' i.e., the 'identifier' from the applications perspective becomes a set of IP addresses instead of a single IP address. o Firewalls that today pass limited traffic, e.g., outbound TCP connections, would presumably block the shim6 protocol. This means that even when shim6 capable hosts are communicating, the I1 messages would be dropped, hence the hosts would not discover that their peer is shim6 capable. This is in fact a feature, since if the hosts managed to establish a ULID-pair context, then the firewall would probably drop the "different" packets that are sent after a failure (those using the shim6 payload extension header with a TCP packet inside it). Thus stateful firewalls that are modified to pass shim6 messages should also be modified to pass the payload extension header, so that the shim can use the alternate locators to recover from failures. This presumably implies that the firewall needs to track the set of locators in use by looking at the shim6 control exchanges. Such firewalls might even want to verify the locators using the HBA/CGA verification themselves, which they can do without modifying any of the shim6 packets they pass through. o Signaling protocols for QoS or other things that involve having devices in the network path look at IP addresses and port numbers, or IP addresses and Flow Labels, need to be invoked on the hosts when the locator pair changes due to a failure. At that point in time those protocols need to inform the devices that a new pair of IP addresses will be used for the flow. Note that this is the case even though this protocol, unlike some earlier proposals, does not overload the flow label as a context tag; the in-path devices need to know about the use of the new locators even though the flow label stays the same. o MTU implications. The path MTU mechanisms we use are robust against different packets taking different paths through the Internet, by computing a minimum over the recently observed path MTUs. When shim6 fails over from using one locator pair to Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page84]85] Internet-Draft Shim6 ProtocolMarchMay 2006 another pair, this means that packets might travel over a different path through the Internet, hence the path MTU might be quite different. Perhaps such a path change would be a good hint to the path MTU mechanism to try a larger MTU? The fact that the shim will add an 8 octet payload extension header to the ULP packets after a locator switch, can also affect the usable path MTU for the ULPs. In this case the MTU change is local to the sending host, thus conveying the change to the ULPs is an implementation matter. o The precise interaction between Mobile IPv6 and shim6 is for further study, but it might make sense to have Mobile IPv6 operate on locators, meaning that the shim would be layered on top of the MIPv6 mechanism. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page85]86] Internet-Draft Shim6 ProtocolMarchMay 2006 16. Security Considerations This document satisfies the concerns specified in [19] as follows: o The HBA technique [7] for verifying the locators to prevent an attacker from redirecting the packet stream to somewhere else. o Requiring a Reachability Probe+Reply before a new locator is used as the destination, in order to prevent 3rd party flooding attacks. o The first message does not create any state on the responder. Essentially a 3-way exchange is required before the responder creates any state. This means that a state-based DoS attack (trying to use up all of memory on the responder) at least provides an IPv6 address that the attacker was using. o The context establishment messages use nonces to prevent replay attacks, and to prevent off-path attackers from interfering with the establishment. o Every control message of the shim6 protocol, past the context establishment, carry the context tag assigned to the particular context. This implies that an attacker needs to discover that context tag before being able to spoof any shim6 control message. Such discovery probably requires to be along the path in order to be sniff the context tag value. The result is that through this technique, the shim6 protocol is protected against off-path attackers. Some of the residual threats in this proposal are: o An attacker which arrives late on the path (after the context has been established) can use the R1bis message to cause one peer to recreate the context, and at that point in time the attacker can observe all of the exchange. But this doesn't seem to open any new doors for the attacker since such an attacker can observe the context tags that are being used, and once known it can use those to send bogus messages. o An attacker which is present on the path so that it can find out the context tags, can generate a R1bis message after it has moved off the path. For this packet to be effective it needs to have a source locator which belongs to the context, thus there can not be "too much" ingress filtering between the attackers new location and the communicating peers. But this doesn't seem to be that severe, because once the R1bis causes the context to be re- established, a new pair of context tags will be used, which will Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page86]87] Internet-Draft Shim6 ProtocolMarchMay 2006 not be known to the attacker. If this is still a concern, we could require a 2-way handshake "did you really loose the state?" in response to the error message. o It might be possible for an attacker to try random 47-bit context tags and see if they can cause disruption for communication between two hosts. In particular, in the case of payload packets, the effects of such attack would be similar of those of an attacker sending packets with spoofed source address. In the case of control packets, it is not enough to find the correct context tag, but additional information is required (e.g. nonces, proper source addresses) (see previous bullet for the case of R1bis). If a 47-bit tag, which is the largest that fits in an 8-octet extension header, isn't sufficient, one could use an even larger tag in the shim6 control messages, and use the low-order 47 bits in the payload extension header. o When the payload extension header is used, an attacker that can guess the 47-bit random context tag, can inject packets into the context with any source locator. Thus if there is ingress filtering between the attacker, this could potentially allow to bypass the ingress filtering. However, in addition to guessing the 47-bit context tag, the attacker also needs to find a context where, after the receiver's replacement of the locators with the ULIDs, the the ULP checksum is correct. But even this wouldn't be sufficient with ULPs like TCP, since the TCP port numbers and sequence numbers must match an existing connection. Thus, even though the issues for off-path attackers injecting packets are different than today with ingress filtering, it is still very hard for an off-path attacker to guess. If IPsec is applied then the issue goes away completely.Nordmark & Bagnulo Expires September 5, 2006 [Page 87] Internet-Draft Shim6 Protocol March 2006 17. IANA Considerations IANA is directed to allocate a new IP Protocol Number value foro The validator included in theSHIM6 Protocol. IANA is directed to recordR1 and R1bis packets are generated as aCGA message type forhash of several input parameters. However, most of the inputs are actually determined by the sender, and only the secret value S is unknown to the sender. However, the resulting protection is deemed to be enough since it would be easier for the attacker to just obtain a new validator sending a I1 packet than performing all the computations required to determine the secret S. However, it is recommended that the host changes the secret S periodically. Nordmark & Bagnulo Expires November 16, 2006 [Page 88] Internet-Draft Shim6 Protocol May 2006 17. IANA Considerations IANA is directed to allocate a new IP Protocol Number value for the SHIM6 Protocol. IANA is directed to record a CGA message type for the SHIM6 Protocol in the [CGA] namespace registry with the value 0x4A30 5662 4858 574B 3655 416F 506A 6D48. IANA is directed to establish a SHIM6 Parameter Registry with two components: SHIM6 Type registrations and SHIM6 Options registrations. The initial contents of the SHIM6 Type registry are as follows: +------------+-----------------------------------------------------+ | Type Value | Message | +------------+-----------------------------------------------------+ | 0 | RESERVED | | | | | 1 | I1 (first establishment message from the initiator) | | | | | 2 | R1 (first establishment message from the responder) | | | | | 3 | I2 (2nd establishment message from the initiator) | | | | | 4 | R2 (2nd establishment message from the responder) | | | | | 5 | R1bis (Reply to reference to non-existent context) | | | | | 6 | I2bis (Reply to a R1bis message) | | | | | 7-59 | Can be allocated using Standards Action | | | | | 60-63 | For Experimental use | | | | | 64 | Update Request | | | | | 65 | Update Acknowledgement | | | | | 66 | Keepalive | | | | | 67 | Probe Message | | | | | 68-123 | Can be allocated using Standards Action | | | | | 124-127 | For Experimental use | +------------+-----------------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page88]89] Internet-Draft Shim6 ProtocolMarchMay 2006 The initial contents of the SHIM6 Options registry are as follows:+--------------+----------------------------------++-------------+----------------------------------+ | Type | Option Name |+--------------+----------------------------------++-------------+----------------------------------+ | 0 | RESERVED | | | | | 1 | Responder Validator | | | | | 2 | Locator List | | | | | 3 | Locator Preferences | | | | | 4 | CGA Parameter Data Structure | | | | | 5 | CGA Signature | | | | | 6 | ULID Pair | | | | | 7 | Forked Instance Identifier | | | | | 8-9 | Allocated using Standards action | | | | | 10 | Probe Option | | | | | 11 | Reachability Option | | | | | 12 | Payload Reception Report Option | | | | | 13-16383 | Allocated using Standards action | | | | | 16384-32767 | For Experimental use |+--------------+----------------------------------++-------------+----------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page89]90] Internet-Draft Shim6 ProtocolMarchMay 2006 18. Acknowledgements Over the years many people active in the multi6 and shim6 WGs have contributed ideas a suggestions that are reflected in this specification. Special thanks to the careful comments from GeoffHouston andHuston, Shinta Sugimoto and Pekka Savola on earlier versions of this draft. Nordmark & Bagnulo ExpiresSeptember 5, 2006 [Page 90] Internet-Draft Shim6 Protocol March 2006 Appendix A. Open Issues The following known open issues in this protocol specification are: o NONE. Nordmark & Bagnulo Expires September 5,November 16, 2006 [Page 91] Internet-Draft Shim6 ProtocolMarchMay 2006 AppendixB.A. Possible Protocol Extensions During the development of this protocol, several issues have been brought up as important one to address, but are ones that do not need to be in the base protocol itself but can instead be done as extensions to the protocol. The key ones are: o As stated in the assumptions in Section 3, the in order for the shim6 protocol to be able to recover from a wide range of failures, for instance when one of the communicating hosts is singly-homed, and cope with a site's ISPs that do ingress filtering based on the source IPv6 address, there is a need for the host to be able to influence the egress selection from its site. Further discussion of this issue is captured in [20]. o Is there need for keeping the list of locators private between the two communicating endpoints? We can potentially accomplish that when using CGA but not with HBA, but it comes at the cost of doing some public key encryption and decryption operations as part of the context establishment. The suggestion is to leave this for a future extension to the protocol. o Defining some form of end-to-end "compression" mechanism that removes the need for including the Shim6 Payload extension header when the locator pair is not the ULID pair. o Supporting the dynamic setting of locator preferences on a site- wide basis, and use the Locator Preference option in the shim6 protocol to convey these preferences to remote communicating hosts. This could mirror the DNS SRV record's notion of priority and weight. o Potentially recommend that more application protocols use DNS SRV records to allow a site some influence on load spreading for the initial contact (before the shim6 context establishment) as well as for traffic which does not use the shim. o Specifying APIs for the ULPs to be aware of the locators the shim is using, and be able to influence the choice of locators (controlling preferences as well as triggering a locator pair switch). This includes providing APIs the ULPs can use to fork a shim context. o Whether it is feasible to relax the suggestions for when context state is removed, so that one can end up with an asymmetric distribution of the context state and still get (most of) the shim benefits. For example, the busy server would go through the context setup but would quickly remove the context state after Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 92] Internet-Draft Shim6 ProtocolMarchMay 2006 this (in order to save memory) but the not-so-busy client would retain the context state. The context recovery mechanism presented in Section 7.5 would then be recreate the state should the client send either a shim control message (e.g., probe message because it sees a problem), or a ULP packet in an payload extension header (because it had earlier failed over to an alternative locator pair, but had been silent for a while). This seems to provide the benefits of the shim as long as the client can detect the failure. If the client doesn't send anything, and it is the server that tries to send, then it will not be able to recover because the shim on the server has no context state, hence doesn't know any alternate locator pairs. o Study whether a host explicitly fail communication when a ULID becomes invalid (based on RFC 2462 lifetimes or DHCPv6), or should we let the communication continue using the invalidated ULID (it can certainly work since other locators will be used). o Study what it would take to make the shim6 control protocol not rely at all on a stable source locator in the packets. This can probably be accomplished by having all the shim control messages include the ULID-pair option. o If each host might have lots of locators, then the currently requirement to include essentially all of them in the I2 and R2 messages might be constraining. If this is the case we can look into using the CGA Parameter Data Structure for the comparison, instead of the prefix sets, to be able to detect context confusion. This would place some constraint on a (logical) only using e.g., one CGA public key, and would require some carefully crafted rules on how two PDSs are compared for "being the same host". But if we don't expect more than a handful locators per host, then we don't need this added complexity. o ULP specified timers for the reachability detection mechanism (which can be useful particularly when there are forked contexts). o Pre-verify some "backup" locator pair, so that the failover time can be shorter. o Study how shim6 and Mobile IPv6 might interact. There existing an initial draft on this topic [21]. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page 93] Internet-Draft Shim6 ProtocolMarchMay 2006 AppendixC. Change LogB. Simplified State Machine Thefollowing changes have been made since draft-ietf-shim6-proto-03: o Editorial clarifications based on comments from Geoff, Shinta, Jari. o Added "no IPv6 NATs as an explicit assumption. o Moving some things out of the Introduction and Overview sections to remove all SHOULDs and MUSTs from there. o Added requirement that any Locator Preference options which use an element length greater than 3 octets have the alreadystates are definedfirst 3 octets of flags, priority and weight. o Fixed security hole where a single message (I1) could cause CT(peer) to be updated. Now a three-way handshake is required before CT(peer) is updated for an existing context.in Section 6.2. Thefollowing changes have been made since draft-ietf-shim6-proto-02: o Replacedintent is that theContext Error messagestylized description below be consistent with theR1bis message. o Removed the Packet In Error option, since it was only usedtextual description in theContext Error message. o Introduced a I2bis message which is sent in response to an I1bis message, sincespecification, but should they conflict, theresponders processingtextual description isquite in this case than innormative. The following table describes theregularpossible actions in state IDLE and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive I1 | Send R1case. o Moved the packet formats for the KeepaliveandProbe message typesstay in IDLE | | | | | Heuristics trigger | Send I1 andEvent optionmove to[8]. Only the message type values and option type value are specified for those in this document. o Removed the unused message types. o AddedI1-SENT | | astate machine description as an appendix. o Filled in all the TBDs - except the IANA assignment of the protocol number. o Specified hownew contextrecovery| | | establishment | | | | | | Receive I2, verify | If successful, send R2 andforked contexts work together. This required the introduction of a Forked Instance option to be able to tell which of possibly forked instances is being recovered. Nordmark & Bagnulo Expires September 5, 2006 [Page 94] Internet-Draft Shim6 Protocol March 2006 o Renamed the "host-pair context"move tobe "ULID-pair context". o Picked some initial retransmit timers for I1 and I2; 4 seconds. o Added timer values as protocol constants. The retransmit timers use binary exponential backoff| | validator andrandomization (between .5| ESTABLISHED | | RESP nonce | | | | If fail, stay in IDLE | | | | | Receive I2bis, | If successful, send R2 and1.5 of the nominal value). o Require that the R1/R1bis verifiers be usable for some minimum time so that the initiator knows for how long time it can safely retransmit I2 before it needs to go backmove tosending I1 again. Picked 30 seconds. o Split the message type codes into 0-63, which will not generate R1bis messages,| | verify validator | ESTABLISHED | | and64-127 which will generate R1bis messages. This allows extensibility of the protocol with new message types while being able to control when R1bis is generated. o Expanded the context tag from 32 to 47 bits. o Specified that enough locators need to be includedRESP nonce | | | | If fail, stay inI2 andIDLE | | | | | R1, R1bis, R2messages. Specified that the HBA/CGA verification must be performed when| N/A (This context lacks thelocator set is received. o Specified that ICMP parameter problem errors are sent in certain error cases,required info | | | forinstance when the verification method is unknown tothereceiver, or there is an unknown message type or option type. o Renamed "payload message"dispatcher tobe "payload extension header". o Many editorial clarifications suggested by Geoff Huston. o Modified the dispatching ofdeliver them) | | | | | Receive payloadextension header to only compare CT(local) i.e., not compare the source and destination IPv6 address fields. The following changes have been made since draft-ietf-shim6-proto-00: o Removed the use of the flow label| Send R1bis andthe overloading of the IP protocol numbers. Instead, when the locator pair is not the ULID pair, the ULP payloads will be carried with an 8 octetstay in IDLE | | extensionheader. The belief is that it is possible to remove these extra bytes by defining future shim6 extensions that exchange more information between the hosts, without having to overload the flow labelheader | | | orthe IP protocol numbers.other control | | | packet | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page95]94] Internet-Draft Shim6 ProtocolMarchMay 2006o Grew the context tag from 20 bits to 32 bits, with the possibility to grow it to 47 bits. This implies changes to the message formats. o Almost by accident, the new shim6 message format is very close to the HIP message format. o Adopted the HIP format for the options, since this makes it easier to describe variable length options.Theoriginal, ND-style, option format requires internal padding infollowing table describes theoptions to make them 8 octet lengthpossible actions intotal, while the HIP format handles that using the option length field. o Removed some of the control messages, and renamed the other ones. o Added a "generation" number to the Locator List option, so that the peers can ensure that the preferences refer to the right "version" of the Locator List. o In order for FBD and exploration to work when there the use of the context is forked, that is different ULP messages are sent over different locator pairs, things are a lot easier if there is only one current locator pair used for each context. Thus the forking of the context is now causing a new context to be established for the same ULID; the new context having a new context tag. The original context is referred to as the "default" context for the ULID pair. o Added more background material and textual descriptions. Nordmark & Bagnulo Expires September 5, 2006 [Page 96] Internet-Draft Shim6 Protocol March 2006 Appendix D. Simplified State Machine The states are defined in Section 6.2. The intent is that the stylized description below be consistent with the textual description in the specification, but should they conflict, the textual description is normative. The following table describes the possible actions in state IDLE and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive I1 | Send R1 and stay in IDLE | | | | | Heuristics trigger | Send I1 and move to I1-SENT | | a new context | | | establishment | | | | | | Receive I2, verify | If successful, send R2 and move to | | validator and | ESTABLISHED | | RESP nonce | | | | If fail, stay in IDLE | | | | | Receive I2bis, | If successful, send R2 and move to | | verify validator | ESTABLISHED | | and RESP nonce | | | | If fail, stay in IDLE | | | | | R1, R1bis, R2 | N/A (This context lacks the required info | | | for the dispatcher to deliver them) | | | | | Receive payload | Send R1bis and stay in IDLE | | extension header | | | or other control | | | packet | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo Expires September 5, 2006 [Page 97] Internet-Draft Shim6 Protocol March 2006 The following table describes the possible actions in state I1-SENTstate I1-SENT and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive R1, verify | If successful, send I2 and move to I2-SENT | | INIT nonce | | | | If fail, discard and stay in I1-SENT | | | | | Receive I1 | Send R2 and stay in I1-SENT | | | | | Receive R2, verify | If successful, move to ESTABLISHED | | INIT nonce | | | | If fail, discard and stay in I1-SENT | | | | | Receive I2, verify | If successful, send R2 and move to | | validator and RESP | ESTABLISHED | | nonce | | | | If fail, discard and stay in I1-SENT | | | | | Receive I2bis, | If successful, send R2 and move to | | verify validator | ESTABLISHED | | and RESP nonce | | | | If fail, discard and stay in I1-SENT | | | | | Timeout, increment | If counter =< I1_RETRIES_MAX, send I1 and | | timeout counter | stay in I1-SENT | | | | | | If counter > I1_RETRIES_MAX, go to E-FAILED | | | | | Receive ICMP payload| Move to E-FAILED | | unknown error | | | | | | R1bis | N/A (Dispatcher doesn't deliver since | | | CT(peer) is not set) | | | | | Receive Payload or | Discard and stay in I1-SENT | | extension header | | | or other control | | | packet | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page98]95] Internet-Draft Shim6 ProtocolMarchMay 2006 The following table describes the possible actions in state I2-SENT and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive R2, verify | If successful move to ESTABLISHED | | INIT nonce | | | | If fail, stay in I2-SENT | | | | | Receive I1 | Send R2 and stay in I2-SENT | | | | | Receive I2 | Send R2 and stay in I2-SENT | | verify validator | | | and RESP nonce | | | | | | Receive I2bis | Send R2 and stay in I2-SENT | | verify validator | | | and RESP nonce | | | | | | Receive R1 | Discard and stay in I2-SENT | | | | | Timeout, increment | If counter =< I2_RETRIES_MAX, send I2 and | | timeout counter | stay in I2-SENT | | | | | | If counter > I2_RETRIES_MAX, send I1 and go | | | to I1-SENT | | | | | R1bis | N/A (Dispatcher doesn't deliver since | | | CT(peer) is not set) | | | | | Receive payload or | Accept and send I2 (probably R2 was sent | | extension header | by peer and lost) | | other control | | | packet | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page99]96] Internet-Draft Shim6 ProtocolMarchMay 2006 The following table describes the possible actions in state I2BIS- SENT and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive R2, verify | If successful move to ESTABLISHED | | INIT nonce | | | | If fail, stay in I2BIS-SENT | | | | | Receive I1 | Send R2 and stay in I2BIS-SENT | | | | | Receive I2 | Send R2 and stay in I2BIS-SENT | | verify validator | | | and RESP nonce | | | | | | Receive I2bis | Send R2 and stay in I2BIS-SENT | | verify validator | | | and RESP nonce | | | | | | Receive R1 | Discard and stay in I2BIS-SENT | | | | | Timeout, increment | If counter =< I2_RETRIES_MAX, send I2bis | | timeout counter | and stay in I2BIS-SENT | | | | | | If counter > I2_RETRIES_MAX, send I1 and | | | go to I1-SENT | | | | | R1bis | N/A (Dispatcher doesn't deliver since | | | CT(peer) is not set) | | | | | Receive payload or | Accept and send I2bis (probably R2 was | | extension header | sent by peer and lost) | | other control | | | packet | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page100]97] Internet-Draft Shim6 ProtocolMarchMay 2006 The following table describes the possible actions in state ESTABLISHED and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Receive I1, compare | If no match, send R1 and stay in ESTABLISHED| | CT(peer) with | | | received CT | If match, send R2 and stay in ESTABLISHED | | | | | | | | Receive I2, verify | If successful, then send R2 and stay in | | validator and RESP | ESTABLISHED | | nonce | | | | Otherwise, discard and stay in ESTABLISHED | | | | | Receive I2bis, | If successful, then send R2 and stay in | | verify validator | ESTABLISHED | | and RESP nonce | | | | Otherwise, discard and stay in ESTABLISHED | | | | | Receive R2 | Discard and stay in ESTABLISHED | | | | | Receive R1 | Discard and stay in ESTABLISHED | | | | | Receive R1bis | Send I2bis and move to I2BIS-SENT | | | | | | | | Receive payload or | Process and stay in ESTABLISHED | | extension header | | | other control | | | packet | | | | | | Implementation | Discard state and go to IDLE | | specific heuristic | | | (E.g., No open ULP | | | sockets and idle | | | for some time ) | | +---------------------+---------------------------------------------+ Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page101]98] Internet-Draft Shim6 ProtocolMarchMay 2006 The following table describes the possible actions in state E-FAILED and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Wait for | Go to IDLE | | NO_R1_HOLDDOWN_TIME | | | | | | Any packet | Process as in IDLE | +---------------------+---------------------------------------------+ The following table describes thepossible actions in state NO- SUPPORTpossible actions in state NO- SUPPORT and their respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Wait for | Go to IDLE | | ICMP_HOLDDOWN_TIME | | | | | | Any packet | Process as in IDLE | +---------------------+---------------------------------------------+ Nordmark & Bagnulo Expires November 16, 2006 [Page 99] Internet-Draft Shim6 Protocol May 2006 Appendix B.1. Simplified State Machine diagram Timeout/Null +------------+ I1/R1 +------------------| NO SUPPORT | Payload or Control/R1bis | +------------+ +---------+ | ^ | | | ICMP Error/Null| | V V | +-----------------+ Timeout/Null +----------+ | | |<---------------| E-FAILED | | +-| IDLE | +----------+ | I2 or I2bis/R2 | | | ^ | | +-----------------+ (Tiemout#>MAX)/Null| | | ^ | | | | | +------+ | | I2 or I2bis/R2 | | Heuristic/I1| I1/R2 | | Payload/Null | | | Control/Null | | I1/R1 or R2 | +--+ | Payload/Null | | R1 or R2/Null | |Heuristic/Null | (Tiemout#<MAX)/I1 | | +----------+ | | | +--------+ | | | V V | | | V | | +-------------------+ R2/Null | +----------------+ | | I2 or I2bis/R2 +------->| | | ESTABLISHED |<----------------------------| I1-SENT | | | | | +-------------------+ +----------------+ | ^ ^ | ^ ^ | | |R2/Null +-------------+ | | | | +----------+ |R1/I2 | | | | | V | | | | +------------------+ | | | | | |-------------+ | | | | I2-SENT | (Timeout#>Max)/I1 | | | | | | | | +------------------+ | | | | ^ | | | +--------------+ | | | I1 or I2bis or I2 or Payload/R2 | | | (Timeout#<Max)/I2 | | | R1 or R1bis/Null | | +-------+ (Timeout#>Max)/I1 | | R2/Null| +------------------------------------------+ | V | | +-------------------+ | | |<-+ (Timeout#<Max)/I2bis +-------->| I2bis-SENT | | I1 or I2 or I2bis/R2 R1bis/I2bis | |--+ R1 or R1bis/Null +-------------------+ Payload/R2 Nordmark & Bagnulo Expires November 16, 2006 [Page 100] Internet-Draft Shim6 Protocol May 2006 Appendix C. Context Tag Reuse The shim6 protocol doesn't have a mechanism for coordinated state removal between the peers, because such state removal doesn't seem to help given that a host can crash and reboot at any time. A result of this is that the protocol needs to be robust against a context tag being reused for some other context. This section summarizes the different cases in which a tag can be reused, and how the recovery works. The different cases are exemplified by the following case. Assume host A and B were communicating using a context with the ULID pair <A1, B2>, and that B had assigned context tag X to this context. We assume that B uses only the context tag to demultiplex the received payload extension headers, since this is the more general case. Further we assume that B removes this context state, while A retains it. B might then at a later time assign CT(local)=X to some other context, and we have several cases: o The context tag is reassigned to a context for the same ULID pair <A1, B2>. We've called this "Context Recovery" in this document. o The context tag is reassigned to a context for a different ULID pair between the same to hosts, e.g., <A3, B3>. We've called this "Context Confusion" in this document. o The context tag is reassigned to a context between B and other host C, for instance for the ULID pair <C3, B2>. That is a form of three party context confusion. Appendix C.1. Context Recovery This case is relatively simple, and is discussed in Section 7.5. The observation is that since the ULID pair is the same, when either A or B tries to establish the new context, A can keep the old context while B re-creates the context with the same context tag CT(B) = X. Appendix C.2. Context Confusion This cases is a bit more complex, and is discussed in Section 7.6. When the new context is created, whether A or B initiates it, host A can detect when it receives B's locator set (in the I2, or R2 message), that it ends up with two contexts to the same peer host (overlapping Ls(peer) locator sets) that have the same context tag CT(peer) = X. At this point in time host A can clear up any possibility of causing confusion by not using the old context to send any more packets. It either just discards the old context (it might not be used by any ULP traffic, since B had discarded it), or it Nordmark & Bagnulo Expires November 16, 2006 [Page 101] Internet-Draft Shim6 Protocol May 2006 recreates a different context for the old ULID pair (<A1, B2>), for which B will assign a unique CT(B) as part of the normal context establishment mechanism. Appendix C.3. Three Party Context Confusion The third case does not have a place where the old state on A can be verified, since the new context is established between B and C. Thus when B receives payload extension headers with X as the context tag, it will find the context for <C3, B2>, hence rewrite the packets to have C3 in the source address field and B2 in the destination address field before passing them up to the ULP. This rewriting is correct when the packets are in fact sent by host C, but if host A ever happens to send a packet using the old context, then the ULP on A sends a packet with ULIDs <A1, B2> and the packet arrives at the ULP on B with ULIDs <C3, B2>. This is clearly an error, and the packet will most likely be rejected by the ULP on B due to a bad pseudo-header checksum. Even if the checksum is ok (probability 2^-16), the ULP isn't likely to have a connection for those ULIDs and port numbers. And if the ULP is connection-less, processing the packet is most likely harmless; such a ULP must be able to copy with random packets being sent by random peers in any case. This broken state, where packets sent from A to B using the old context on host A might persist for some time, but it will not remain for very long. The unreachability detection on host A will kick in, because it does not see any return traffic (payload or Keepalive messages) for the context. This will result in host A sending Probe messages to host B to find a working locator pair. The effect of this is that host B will notice that it does not have a context for the ULID pair <A1, B2> and CT(B) = X, which will make host B send an R1bis packet to re-establish that context. The re-established context, just like in the previous section, will get a unique CT(B) assigned by host B, thus there will no longer be any confusion. In summary, there are cases where a context tag might be reused while some peer retains the state, but the protocol can recover from it. The probability of these events is low given the 47 bit context tag size. However, it is important that these recovery mechanisms be tested. Thus during development and testing it is recommended that implementations not use the full 47 bit space, but instead keep e.g. the top 40 bits as zero, only leaving the host with 128 unique context tags. This will help test the recovery mechanisms. Nordmark & Bagnulo Expires November 16, 2006 [Page 102] Internet-Draft Shim6 Protocol May 2006 Appendix D. Design Alternatives This document has picked a certain set of design choices in order to try to work out a bunch of the details, and stimulate discussion. But as has been discussed on the mailing list, there are other choices that make sense. This appendix tries to enumerate some alternatives. Appendix D.1. Context granularity Over the years various suggestions have been made whether the shim should, even if it operates at the IP layer, be aware of ULP connections and sessions, and as a result be able to make separate shim contexts for separate ULP connections and sessions. A few different options have been discussed: o Each ULP connection maps to its own shim context. o The shim is unaware of the ULP notion of connections and just operates on a host-to-host (IP address) granularity. o Hybrids where the shim is aware of some ULPs (such as TCP) and handles other ULPs on a host-to-host basis. Having shim state for every ULP connection potentially means higher overhead since the state setup overhead might become significant; there is utility in being able to amortize this over multiple connections. But being completely unaware of the ULP connections might hamper ULPs that want different communication to use different locator pairs, for instance for quality or cost reasons. The protocol has a shim which operates with host-level granularity (strictly speaking, with ULID-pair granularity, to be able to amortize the context establishment over multiple ULP connections. This is combined with the ability for shim-aware ULPs to request context forking so that different ULP traffic can use different locator pairs. Appendix D.2. Demultiplexing of data packets in shim6 communications Once a ULID-pair context is established between two hosts, packets may carry locators that differ from the ULIDs presented to the ULPs using the established context. One of main functions of the SHIM6 layer is to perform the mapping between the locators used to forward packets through the network andtheir respective triggers: +---------------------+---------------------------------------------+ | Trigger | Action | +---------------------+---------------------------------------------+ | Wait for | Gothe ULIDs presented toIDLE | | ICMP_HOLDDOWN_TIME | | | | | | Any packet | Process as in IDLE | +---------------------+---------------------------------------------+ Appendix D.1 Simplified State Machine diagram Forthetime being, a pdf version ofULP. In order to perform that translation for incoming packets, thestate machine diagram can be found at: http://www.it.uc3m.es/marcelo/state_machine.pdfSHIM6 Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page102]103] Internet-Draft Shim6 ProtocolMarchMay 2006Appendix E. Context Tag Reuse The shim6 protocol doesn't have a mechanism for coordinated state removal betweenlayer needs to first identify which of thepeers, because such state removal doesn't seemincoming packets need tohelp given that a host can crashbe translated andreboot at any time. A result of thisthen perform the mapping between locators and ULIDs using the associated context. Such operation is called demultiplexing. It should be noted thatthe protocol needs tobecause any address can berobust againstused both as acontext tag being reused for somelocator and as a ULID, additional information othercontext. This section summarizesthan thedifferent casesaddresses carried inwhich a tag canpackets, need to bereused, and how the recovery works. The different cases are exemplified by the following case. Assumetaken into account for this operation. For example, if a hostAhas address A1 andB wereA2 and starts communicatingusingwith acontextpeer with addresses B1 and B2, then some communication (connections) might use theULIDpair <A1,B2>,B1> as ULID andthat B had assigned context tag X to this context. We assume that B uses onlyothers might use e.g., <A2, B2>. Initially there are no failures so these address pairs are used as locators i.e. in thecontext tag to demultiplexIP address fields in thereceived payload extension headers, since thispackets on the wire. But when there is a failure themore general case. Further we assumeshim6 layer on A might decide to send packets thatB removesused <A1, B1> as ULIDs using <A2, B2> as the locators. In thiscontext state, while A retains it.case Bmight then at a later time assign CT(local)=Xneeds to be able to rewrite the IP address field for someother context,packets andwenot others, but the packets all haveseveral cases: o Thethe same locator pair. In order to accomplish the demultiplexing operation successfully, data packets carry a context tagis reassignedthat allows the receiver of the packet toadetermine the shim context to be used to perform the operation. Two mechanisms for carrying thesame ULID pair <A1, B2>. We've called this "Context Recovery" in this document. o Thecontext tagis reassigned to ainformation have been considered in depth during the shim protocol design. Those carrying the contextfor a different ULID pair betweentag in thesameflow label field of the IPv6 header and the usage of a new extension header tohosts, e.g., <A3, B3>. We've called this "Context Confusion" in this document. o Thecarry the contexttagtag. In this appendix we will describe the pros and cons of each approach and justify the selected option. Appendix D.2.1. Flow-label A possible approach isreassignedtoacarry the contextbetween B and other host C, for instance fortag in theULID pair <C3, B2>. That is a formFlow Label field ofthree party context confusion. Appendix E.1 Context Recoverythe IPv6 header. Thiscasemeans that when a shim6 context isrelatively simple, andestablished, a Flow Label value isdiscussed in Section 7.5.associated with this context (and perhaps a separate flow label for each direction). Theobservation issimplest approach thatsince the ULID pairdoes this isthe same, when either A or B triestoestablishhave thenew context, A can keeptriple <Flow Label, Source Locator, Destination Locator> identify theoldcontextwhile B re-createsat thecontextreceiver. The problem withthe same context tag CT(B) = X. Appendix E.2 Context Confusion This cases is a bit more complex, andthis approach isdiscussed in Section 7.6. Whenthat because thenew context is created, whether A or B initiates it, host A can detect when it receives B'slocatorset (in the I2, or R2 message), thatsets are dynamic, itends up with two contextsis not possible at any given moment tothe same peer host (overlapping Ls(peer) locator sets)be sure thathavetwo contexts for which the same context tagCT(peer) = X. At this point in time host A can clear up any possibility of causing confusion by not usingis allocated will have disjoint locator sets during theold context to send any more packets. It either just discardslifetime of theold context (it mightcontexts. Suppose that Node A has addresses IPA1, IPA2, IPA3 and IPA4 and that Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page103]104] Internet-Draft Shim6 ProtocolMarchMay 2006not be used by any ULP traffic, sinceHost Bhad discarded it), or it recreates ahas addresses IPB1 and IPB2. Suppose that two differentcontext forcontexts are established between HostA and HostB. Context #1 is using IPA1 and IPB1 as ULIDs. The locator set associated to IPA1 is IPA1 and IPA2 while theold ULID pair (<A1, B2>), for which B will assign a unique CT(B)locator set associated to IPB1 is just IPB1. Context #2 uses IPA3 and IPB2 aspartULIDs. The locator set associated to IPA3 is IPA3 and IPA4 and the locator set associated to IPB2 is just IPB2. Because the locator sets of thenormal context establishment mechanism. Appendix E.3 Three PartyContextConfusion The third case does not have a place where#1 and Context #2 are disjoint, hosts could think that theold state on Asame context tag value can beverified, since the new contextassigned to both of them. The problem arrives when later on IPA3 isestablished between Badded as a valid locator for IPA1 andC. Thus when B receives payload extension headers with XIPB2 is added asthe context tag, it will find the contexta valid locator for<C3, B2>, hence rewrite the packets to have C3IPB1 in Context #1. In this case, thesource address fieldtriple <Flow Label, Source Locator, Destination Locator> would not identify a unique context anymore andB2 in the destination address field before passing them up to the ULP. This rewriting iscorrectwhen the packets are in fact sent by host C, but if hostdemultiplexing is no longer possible. Aever happenspossible approach tosendovercome this limitation is simply not to repeat the Flow Label values for any communication established in apackethost. This basically means that each time a new communication that is usingthe old context, then the ULP on A sendsdifferent ULIDs is established, apacket withnew Flow Label value is assigned to it. By this mean, each communication that is using different ULIDs<A1, B2> and the packet arrives at the ULP on Bcan be differentiated because it has a different Flow Label value. The problem withULIDs <C3, B2>. Thissuch approach isclearly an error, and the packet will most likely be rejected bythat it requires that theULP on B due to a bad pseudo-header checksum. Even ifreceiver of thechecksum is ok (probability 2^-16),communication allocates theULP isn't likelyFlow Label value used for incoming packets, in order tohaveassign them uniquely. For this, aconnection for those ULIDs and port numbers. And ifshim negotiation of theULP is connection-less, processingFlow Label value to use in thepacketcommunication ismost likely harmless; such a ULP mustneeded before exchanging data packets. This poses problems with non- shim capable hosts, since they would not be able tocopy with random packets being sent by random peers in any case. This broken state, where packets sent from A to B using the old context on host A might persist for some time, but it will not remain for very long. The unreachability detection on host A will kick in, because it does not see any return traffic (payload or Keepalive messages)negotiate an acceptable value for thecontext.Flow Label. Thiswill result in host A sending Probe messages to host Blimitation can be lifted by marking the packets that belong tofind a working locator pair. The effect of this isshim sessions from those thathost B will noticedo not. These marking would require at least a bit in the IPv6 header thatit doesis nothave a contextcurrently available, so more creative options would be required, for instance using new Next Header values to indicate that theULID pair <A1, B2> and CT(B) = X, which will make host B send an R1bispacket belongs tore-establisha shim6 enabled communication and thatcontext. The re-established context, just like intheprevious section, will get a unique CT(B) assigned by host B, thus there will no longer be any confusion. In summary, there are cases where aFlow Label carries contexttag might be reused while some peer retainsinformation as proposed in thestate, butnow expired NOID draft. . However, even if this is done, this approach is incompatible with theprotocol can recover from it. The probabilitydeferred establishment capability ofthese events is low giventhe47 bit context tag size. However, itshim protocol, which isimportant that these recovery mechanisms be tested. Thus during development and testinga preferred function, since itis recommended that implementations not use the full 47 bit space, but instead keep e.g.suppresses thetop 40 bits as zero, only leavingdelay due to thehost with 128 uniqueshim contexttags. This will help testestablishment prior to initiation of therecovery mechanisms.communication and it also allows nodes to define at Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page104]105] Internet-Draft Shim6 ProtocolMarchMay 2006Appendix F. Design Alternatives This document has picked a certain set of design choices in order to try to work out a bunchwhich stage of thedetails, and stimulate discussion. But as has been discussedcommunication they decide, based onthe mailing list, there are other choicestheir own policies, thatmake sense. This appendix triesa given communication requires toenumerate some alternatives. Appendix F.1 Context granularity Overbe protected by theyears various suggestions have been made whethershim. In order to cope with theshim should, even if it operates atidentified limitations, an alternative approach that does not constraints theIP layer, be aware of ULP connections and sessions, and as a result be ableflow label values used by communications that are using ULIDs equal tomake separatethe locators (i.e. no shimcontexts for separate ULP connections and sessions. A fewtranslation) is to only require that differentoptions have been discussed: o Each ULP connection mapsflow label values are assigned toits owndifferent shimcontext. ocontexts. In such approach communications start with unmodified flow label usage (could be zero, or as suggested in [16]). Theshimpackets sent after a failure when a different locator pair isunaware ofused would use a completely different flow label, and this flow label could be allocated by theULP notionreceiver as part ofconnections and just operates on a host-to-host (IP address) granularity. o Hybrids wherethe shim context establishment. Since it isawareallocated during the context establishment, the receiver ofsome ULPs (such as TCP) and handlesthe "failed over" packets can pick a flow label of its choosing (that is unique in the sense that no otherULPs oncontext is using it as ahost-to-host basis. Having shim statecontext tag), without any performance impact, and respecting that forevery ULP connection potentially means higher overhead sinceeach locator pair, thestate setup overhead might become significant; there is utility in being ableflow label value used for a given locator pair doesn't change due toamortize this over multiple connections. But being completely unawarethe operation of theULP connections might hamper ULPsmultihoming shim. In this approach, the constraint is that Flow Label values being used as context identifiers cannot be used by other communications thatwant different communication tousedifferentnon-disjoint locatorpairs, for instance for quality or cost reasons. The protocolsets. This means that once that a given Flow Label value has been assigned to a shimwhich operates with host-level granularity (strictly speaking, with ULID-pair granularity, to be able to amortize thecontextestablishment over multiple ULP connections. This is combined withthat has a certain locator sets associated, theabilitysame value cannot be used forshim-aware ULPs to request context forking soother communications thatdifferent ULP traffic canusedifferentan address pair that is contained in the locatorpairs. Appendix F.2 Demultiplexingsets ofdata packets in shim6 communications Oncethe context. This is aULID-pair contextconstraint in the potential Flow Label allocation strategies. A possible workaround to this constraint isestablished between two hosts,to mark shim packetsmay carry locatorsthatdiffer from the ULIDs presentedrequire translation, in order tothe ULPsdifferentiate them from regular IPv6 packets, using theestablished context. One of main functionsartificial Next Header values described above. In this case, the Flow Label values constrained are only those of theSHIM6 layerpackets that are being translated by the shim. This last approach would be the preferred approach if the context tag is toperformbe carried in themapping betweenFlow Label field. This is not only because it imposes thelocators usedminimum constraints toforward packets throughthenetwork andFlow Label allocation strategies, limiting theULIDs presentedrestrictions only tothe ULP. In orderthose packets that need toperformbe translated by the shim, but also because Context Loss detection mechanisms greatly benefit from the fact thattranslation for incoming packets,shim data packets are identified as such, allowing theSHIM6receiving end to identify if a shim context associated to a received packet is suppose to exist, as it will be discussed in the Context Loss detection appendix below. Appendix D.2.2. Extension Header Another approach, which is the one selected for this protocol, is to Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page105]106] Internet-Draft Shim6 ProtocolMarchMay 2006layer needs to first identify which ofcarry theincomingcontext tag in a new Extension Header. These context tags are allocated by the receiving end during the shim6 protocol initial negotiation, implying that each context will have two context tags, one for each direction. Data packetsneed towill betranslated and then perform the mapping between locators and ULIDsdemultiplexed using theassociated context. Such operationcontext tag carried in the Extension Header. This seems a clean approach since it does not overload existing fields. However, it introduces additional overhead in the packet due to the additional header. The additional overhead introduced iscalled demultiplexing. It8 octets. However, it should be noted thatbecause any address can be used both asthe context tag is only required when a locatorand as a ULID, additional informationother than theaddresses carried in packets, need to be taken into account for this operation. For example, if a host has address A1 and A2 and starts communicating with a peer with addresses B1 and B2, then some communication (connections) might use the pair <A1, B1> as ULID and others might use e.g., <A2, B2>. Initially there are no failures so these address pairs areone used aslocators i.e.ULID is contained in theIPpacket. Packets where both the source and destination address fieldsin the packets on the wire. But when there is a failurecontain theshim6 layer on A might decide to send packets that used <A1, B1> asULIDsusing <A2, B2> as the locators. In this case B needs to be able to rewrite the IP address field for some packets anddo notothers, but the packets all have the same locator pair. In order to accomplish the demultiplexing operation successfully, data packets carryrequire a contexttag that allows the receiver of the packet to determine the shim context to be used to perform the operation. Two mechanisms for carryingtag, since no rewriting is necessary at thecontext tag information have been considered in depth duringreceiver. This approach would reduce theshim protocol design. Those carryingoverhead, because thecontext tagadditional header is only required after a failure. On the other hand, this approach would cause changes in theflow label field ofavailable MTU for some packets, since packets that include theIPv6 header andExtension Header will have an MTU 8 octets shorter. However, path changes through theusage ofnetwork can result in different MTU in any case, thus having anew extension header to carrylocator change, which implies a path change, affect thecontext tag.MTU doesn't introduce any new issues. Appendix D.3. Context Loss Detection In this appendix we willdescribe the pros and cons of each approach and justify the selected option. Appendix F.2.1 Flow-label A possible approach ispresent different approaches considered tocarry thedetect contexttag in the Flow Label field of the IPv6 header. This means that when a shim6loss and potential context recovery strategies. The scenario being considered isestablished,the following: Node A and Node B are communicating using IPA1 and IPB1. Sometime later, aFlow Label valueshim context isassociatedestablished between them, with IPA1 and IPB1 as ULIDs and IPA1,...,IPAn and IPB1,...,IPBm as locator set respectively. It may happen, that later on, one of the hosts, e.g. Host A looses the shim context. The reason for thiscontext (and perhapscan be that Host A has aseparate flow label for each direction). The simplest approachmore aggressive garbage collection policy than HostB or thatdoes this is to have the triple <Flow Label, Source Locator, Destination Locator> identifyan error occurred in thecontextshim layer at host A resulting in thereceiver.loss of the context state. Theproblemmechanisms considered in this appendix are aimed to deal with thisapproach isproblem. There are essentially two tasks thatbecauseneed to be performed in order to cope with this problem: first, thelocator sets are dynamic, it is not possible at any given momentcontext loss must be detected and second the context needs to besure that two contextsrecovered/ reestablished. Mechanisms forwhichdetecting context. loss These mechanisms basically consist in that each end of thesamecontexttag is allocated will have disjoint locator sets duringperiodically sends a packet containing context-specific information to thelifetimeother end. Upon reception of such packets, thecontexts. Supposereceiver verifies thatNode A has addresses IPA1, IPA2, IPA3 and IPA4 andthe required context exists. In case that the context Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page106]107] Internet-Draft Shim6 ProtocolMarchMay 2006Host B has addresses IPB1 and IPB2. Suppose that two different contexts are established between HostA and HostB. Context #1 is using IPA1 and IPB1 as ULIDs. The locator set associateddoes not exist, it sends a packet notifying the problem toIPA1 is IPA1 and IPA2 whilethelocator set associatedsender. An obvious alternative for this would be toIPB1 is just IPB1. Context #2 uses IPA3create a specific context keepalive exchange, which consists in periodically sending packets with this purpose. This option was considered andIPB2 as ULIDs. The locator set associateddiscarded because it seemed an overkill toIPA3 is IPA3 and IPA4 and the locator set associateddefine a new packet exchange toIPB2deal with this issue. An alternative isjust IPB2. Because the locator sets ofto piggyback theContext #1context loss detection function in other existent packet exchanges. In particular, both shim control andContext #2 are disjoint, hosts could thinkdata packets can be used for this. Shim control packets can be trivially used for this, because they carry context specific information, so that when a node receives one of such packets, it will verify if thesamecontexttag valueexists. However, shim control frequency may not be adequate for context loss detection since control packet exchanges can beassigned to both of them. The problem arrives when latervery limited for a session in certain scenarios. Data packets, onIPA3 is added asthe other hand, are expected to be exchanged with avalid locator for IPA1 and IPB2 is added ashigher frequency but they do not necessarily carry context specific information. In particular, packets flowing before avalidlocatorfor IPB1change (i.e. packet carrying the ULIDs inContext #1. In this case,thetriple <Flow Label, Source Locator, Destination Locator> wouldaddress fields) do notidentifyneed context information since they do not need any shim processing. Packets that carry locators that differ from the ULIDs carry context information. However, we need to make auniquedistinction here between the different approaches considered to carry the contextanymoretag, in particular between those approaches where packets are explicitly marked as shim packets andcorrect demultiplexing is no longer possible. A possible approach to overcome this limitation is simplythose approaches where packets are notto repeatmarked as such. For instance, in the case where the context tag is carried in the Flow Label and packets are not marked as shim packets (i.e. no new Next Header values are defined forany communication established in a host. This basically means that each timeshim), anew communicationreceiver that has lost the associated context isusing different ULIDs is established, a new Flow Label value is assignednot able toit. By this mean, each communicationdetect that the packet isusing different ULIDs can be differentiated because it hasassociated with adifferent Flow Label value.missing context. Theproblem with such approachresult is thatit requires that the receiver ofthecommunication allocatespacket will be passed unchanged to theFlow Label value used for incoming packets,upper layer protocol, which inorderturn will probably silently discard it due toassign them uniquely. For this,ashim negotiation of the Flow Label value to use inchecksum error. The resulting behavior is that thecommunicationcontext loss isneeded before exchanging data packets.undetected. Thisposes problems with non- shim capable hosts, since they would not be ableis one additional reason tonegotiatediscard anacceptable value forapproach that carries the context tag in the FlowLabel. This limitation can be lifted by markingLabel field and does not explicitly mark the shim packets as such. On the other hand, approaches thatbelong tomark shimsessions from those that do not. These marking would require at least a bit indata packets (like theIPv6 header that is not currently available, so more creative options would be required, for instance usingExtension Header or the Flow Label with new Next Header valuesto indicate thatapproaches) allow thepacket belongsreceiver toa shim6 enabled communication and thatdetect if theFlow Label carriescontextinformation as proposed inassociated to thenow expired NOID draft. . However, even if thisreceived packet isdone,missing. In thisapproach is incompatible withcase, data packets also perform thedeferred establishment capabilityfunction ofthe shim protocol, which isapreferred function, since it suppresses the delay due to the shimcontextestablishment prior to initiation of the communication andloss detection exchange. However, italso allows nodes to define atmust be noted that only those packets that Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page107]108] Internet-Draft Shim6 ProtocolMarchMay 2006which stage ofcarry a locator that differs form thecommunication they decide, based on their own policies,ULID are marked. This basically means thata given communication requires tocontext loss will beprotected by the shim. In order to cope with the identified limitations,detected after an outage has occurred i.e. alternativeapproach that does not constraints the flow label values used by communications that are using ULIDs equal to thelocators(i.e. no shim translation) is to only require that different flow label valuesareassigned to differentbeing used. Summarizing, the proposed context loss detection mechanisms uses shimcontexts. In such approach communications start with unmodified flow label usage (could be zero, or as suggested in [16]). Thecontrol packetssent after a failure when a different locator pair is used would use a completely different flow label,andthis flow label could be allocated by the receiver as part of the shimpayload extension headers to detect contextestablishment. Since it is allocated during theloss. Shim control packets detect contextestablishment,loss during thereceiverwhole lifetime of the"failed over" packets can pick a flow label of its choosing (that is uniquecontext, but the expected frequency in some cases is very low. On thesense that noother hand, payload extension headers have a higher expected frequency in general, but they only detect contextis usingloss after an outage. This behavior implies that itas awill be common that contexttag), without any performance impact, and respectingloss is detected after a failure i.e. once that it is actually needed. Because of that, a mechanism foreach locator pair,recovering from context loss is required if this approach is used. Overall, theflow label value usedmechanism for detecting lost context would work as follows: the end that still has the context available sends agiven locator pair doesn't change duemessage referring to theoperation ofcontext. Upon themultihoming shim. In this approach,reception of such message, theconstraint is that Flow Label values being used as context identifiers cannot be used by other communications that use non-disjoint locator sets. This means that onceend thata given Flow Label valuehasbeen assigned to a shimlost the contextthat has a certain locator sets associated,identifies thesame value cannot be used for other communications that use an address pair that is contained insituation and notifies thelocator sets ofcontext loss event to thecontext. This isother end by sending aconstraint inpacket containing thepotential Flow Label allocation strategies. A possible workaround to this constraint is to mark shim packets that require translation, in order to differentiate them from regular IPv6 packets, usinglost context information extracted from theartificial Next Header values described above. In this case,received packet. One option is to simply send an error message containing theFlow Label values constrained are only thosereceived packets (or at least as much of thepacketsreceived packet thatare being translated bytheshim. This last approach would beMTU allows to fit in). One of thepreferred approach ifgoals of this notification is to allow the other end that still retains contexttag isstate, tobe carriedreestablish the lost context. The mechanism to reestablish the loss context consists in performing theFlow Label field.4-way initial handshake. This isnot only because it imposesa time consuming exchange and at this point time may be critical since we are reestablishing a context that is currently needed (because context loss detection may occur after a failure). So, another option, which is theminimum constraintsone used in this protocol, is to replace theFlow Label allocation strategies, limiting the restrictions only to those packetserror message by a modified R1 message, so thatneedthe time required to perform the context establishment exchange can betranslated byreduced. Upon theshim, but also because Context Loss detection mechanisms greatly benefit fromreception of this modified R1 message, thefactend thatshim data packets are identified as such, allowingstill has thereceiving end to identify if a shimcontextassociated tostate can finish the context establishment exchange and restore the lost context. Appendix D.4. Securing locator sets The adoption of areceived packet is suppose to exist,protocol like SHIM that allows the binding of a given ULID with a set of locators opens the doors for different types of redirection attacks asit will be discusseddescribed in [19]. The goal in terms of security for theContext Loss detection appendix below. Appendix F.2.2 Extension Header Another approach, whichdesign of the shim protocol is not to introduce any new vulnerability in theone selected for this protocol,Internet architecture. It is a non-goal to provide additional protection than the currently available in the single-homed IPv6 Internet. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page108]109] Internet-Draft Shim6 ProtocolMarchMay 2006carryMultiple security mechanisms were considered to protect thecontext tag inshim protocol. In this appendix we will present some of them. The simplest option to protect the shim protocol was to use cookies i.e. anew Extension Header. Theserandomly generated bit string that is negotiated during the contexttags are allocated byestablishment phase and then it is included in following signaling messages. By this mean, it would be possible to verify that thereceiving end duringparty that was involved in theshim6 protocolinitialnegotiation, implyinghandshake is the same party thateach context will have two context tags, one for each direction. Data packets will be demultiplexedis introducing new locators. Moreover, before using a new locator, an exchange is performed through thecontext tag carried innew locator, verifying that theExtension Header. This seemsparty located at the new locator knows the cookie i.e. that it is the same party that performed the initial handshake. While this security mechanisms does indeed provide aclean approach sincefair amount of protection, it doesnot overload existing fields. However,leave the door open for the so-called time shifted attacks. In these attacks, an attacker that once was on the path, itintroduces additional overhead indiscovers thepacket due tocookie by sniffing any signaling message. After that, theadditional header. The additional overhead introduced is 8 octets. However, it should be noted thatattacker can leave thecontext tag is only required whenpath and still perform alocator other than the one usedredirection attack, since asULIDhe iscontainedin possession of thepacket. Packets where bothcookie, he can introduce a new locator in thesourcelocator set anddestination address fields containhe can also successfully perform theULIDs do not require a context tag, since no rewritingreachability exchange if he isnecessaryable to receive packets at thereceiver. This approach would reduce the overhead, becausenew locator. The difference with theadditional headercurrent single-homed IPv6 situation isonly required after a failure. On the other hand, this approach would cause changesthat in theavailable MTU for some packets, since packets that includecurrent situation theExtension Header will have an MTU 8 octets shorter. However, path changes throughattacker needs to be on-path during thenetwork can result in different MTUwhole lifetime of the attack, while inany case, thus having a locator change, which implies athis new situation where only cookie protection if provided, an attacker that once was on the pathchange, affectcan perform attacks after he has left theMTU doesn't introduce any new issues. Appendix F.3 Context Loss Detection In this appendix we will present different approaches considered to detect context loss and potential context recovery strategies. The scenario being considered ison-path location. Moreover, because thefollowing: Node A and Node B are communicating using IPA1 and IPB1. Sometime later, a shim contextcookie isestablished between them, with IPA1 and IPB1 as ULIDs and IPA1,...,IPAn and IPB1,...,IPBm as locator set respectively. It may happen, that later on, oneincluded in signaling messages, the attacker can discover the cookie by sniffing any of them, making thehosts, e.g. Host A loosesprotocol vulnerable during the whole lifetime of the shim context.The reason for this can be that HostAhaspossible approach to increase the security was to use amore aggressive garbage collection policy than HostB orshared secret i.e. a bit string thatan error occurred inis negotiated during theshim layer at host A resulting ininitial handshake but that is used as a key to protect following messages. With this technique, theloss ofattacker must be present on thecontext state. The mechanisms considered inpath sniffing packets during the initial handshake, since it is the only moment where the shared secret is exchanged. While thisappendix are aimedimproves the security, it is still vulnerable todeal with this problem. There are essentially two taskstime shifted attacks, even though it imposes thatneedthe attacker must be on path at a very specific moment (the establishment phase) to actually beperformed in orderable tocope withlaunch the attack. While thisproblem: first,seems to substantially improve thecontext loss mustsituation, it should bedetectednoted that, depending on protocol details, an attacker may be able to force the recreation of the initial handshake (for instance by blocking messages andsecondmaking thecontext needs to be recovered/ reestablished. Mechanisms for detecting context. loss These mechanisms basically consist inparties think thateach end ofthe contextperiodically sends a packet containing context-specific information to the other end. Upon reception of such packets,has been lost), so thereceiver verifiesresulting situation may not differ that much from therequired context exists. In casecookie based approach. Another option that was discussed during thecontextdesign of the protocol Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page109]110] Internet-Draft Shim6 ProtocolMarchMay 2006does not exist, it sends a packet notifying the problem towas thesender. An obvious alternativepossibility of using IPsec forthis would be to create a specific context keepalive exchange, which consistsprotecting the shim protocol. Now, the problem under consideration inperiodically sending packets withthispurpose. This option was considered and discarded because it seemed an overkillscenario is how todefinesecurely bind an address that is being used as ULID with anew packetlocator set that can be used to exchange packets. The mechanism provided by IPsec todealsecurely bind the address used withthis issue. An alternativethe cryptographic keys isto piggybackthecontext loss detection function in other existent packet exchanges. In particular, both shim control and data packets can be used for this. Shim control packets can be trivially used for this, because they carry context specific information, sousage of digital certificates. This implies thatwhen a node receives onean IPsec based solution would require that the generation ofsuch packets, it will verify ifdigital certificates that bind thecontext exists. However, shim control frequency may not be adequate for context loss detection since control packet exchanges can be very limited for a session in certain scenarios. Data packets, onkey and theother hand, are expected to be exchanged with a higher frequency but they do not necessarily carry context specific information. In particular, packets flowing beforeULID by alocator change (i.e. packet carrying the ULIDscommon third trusted party for both parties involved in theaddress fields) do not need context information since they do not need any shim processing. Packets that carry locatorscommunication. Considering thatdiffer fromtheULIDs carry context information. However, we need to makescope of application of the shim protocol is global, this would imply adistinction here betweenglobal public key infrastructure. The major issues with this approach are the deployment difficulties associated with a global PKI. Finally two differentapproaches consideredtechnologies were selected tocarryprotect thecontext tag, in particular between those approaches where packets are explicitly marked asshimpacketsprotocol: HBA [7] andthoseCGA [6]. These two approacheswhere packets are not markedprovide a similar level of protection but they provide different functionality with a different computational cost. The HBA mechanism relies on the capability of generating all the addresses of a multihomed host assuch. For instance, inan unalterable set of intrinsically bound IPv6 addresses, known as an HBA set. In this approach, addresses incorporate a cryptographic one-way hash of thecase whereprefix-set available into thecontext taginterface identifier part. The result iscarried inthat the binding between all the available addresses is encoded within the addresses themselves, providing hijacking protection. Any peer using theFlow Label and packets are not marked asshimpackets (i.e. no new Next Header values are definedprotocol node can efficiently verify that the alternative addresses proposed forshim),continuing the communication are bound to the initial address through areceiversimple hash calculation. A limitation of the HBA technique is thathas lostonce generated theassociated contextaddress set is fixed and cannot be changed without also changing all the addresses of the HBA set. In other words, the HBA technique does notablesupport dynamic addition of address todetecta previously generated HBA set. An advantage of this approach is that it requires only hash operations to verify a locator set, imposing very low computational cost to thepacketprotocol. In a CGA based approach the address used as ULID isassociated withamissing context.CGA that contains a hash of a public key in its interface identifier. The result isthata secure binding between thepacket will be passed unchangedULID and the associated key pair. This allows each peer to use theupper layer protocol, which in turn will probably silently discard it duecorresponding private key toa checksum error.sign the shim messages that convey locator set information. Theresulting behaviortrust chain in this case isthatthecontext loss is undetected. Thisfollowing: the ULID used for the communication isone additional reasonsecurely bound todiscard an approach that carriesthecontext tag inkey pair because it contains theFlow Label field and does not explicitly markhash of theshim packets as such. Onpublic key, and theother hand, approaches that mark shim data packets (likelocator set is bound to theExtension Header orpublic key through theFlow Label withsignature. The CGA approach then supports dynamic addition of newNext Header values approaches) allowlocators in thereceiverlocator set, since in order todetect ifdo that, thecontext associatednode only needs to sign thereceived packet is missing. In this case, data packets also performnew locator with thefunctionprivate key associated with the CGA used as ULID. A limitation ofa context loss detection exchange. However, it must be noted that only those packets thatNordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page110]111] Internet-Draft Shim6 ProtocolMarchMay 2006carrythis approach is that it imposes systematic usage of public key cryptography with its associate computational cost. Any of these two mechanisms HBA and CGA provide time-shifted attack protection, since the ULID is securely bound to a locator set thatdiffers formcan only be defined by theULIDowner of the ULID. So, the design decision adopted was that both mechanisms HBA and CGA aremarked. This basically meanssupported, so thatcontext loss willwhen only stable address sets are required, the nodes can benefit from the low computational cost offered by HBA while when dynamic locator sets are required, this can bedetected afterachieved through CGAs with anoutage has occurred i.e. alternative locatorsadditional cost. Moreover, because HBAs arebeing used. Summarizing,defined as a CGA extension, theproposed context loss detection mechanisms uses shim control packetsaddresses available in a node can simultaneously be CGAs andpayload extension headers to detect context loss. Shim control packets detect context loss during the whole lifetime of the context, butHBAs, allowing theexpected frequency in some cases is very low. Onusage of theother hand, payload extension headers haveHBA and CGA functionality when needed without requiring ahigher expected frequencychange ingeneral, but they only detectthe addresses used. Appendix D.5. ULID-pair contextloss after an outage. This behavior implies that it will be common thatestablishment exchange Two options were considered for the ULID-pair contextloss is detected afterestablishment exchange: afailure i.e. once that it is actually needed. Because of that,2-way handshake and amechanism for recovering from context loss is required if this approach is used. Overall, the mechanism4-way handshake. A key goal fordetecting lost context would work as follows:theenddesign of this exchange was thatstill has the context available sendsprotection against DoS attacks. The attack under consideration was basically amessage referringsituation where an attacker launches a great amount of ULID-pair establishment request packets, exhausting victim's resources, similar to TCP SYN flooding attacks. A 4 way-handshake exchange protects against these attacks because thecontext. Uponreceiver does not creates any state associate to a given context until the reception ofsuch message, the end that has lostthecontext identifiessecond packet which contains a prior contact proof in thesituation and notifiesform of a token. At this point thecontext loss event toreceiver can verify that at least theother endaddress used bysending a packet containingthelost context information extracted frominitiator is at some extent valid, since thereceived packet. One optioninitiator is able tosimply send an error message containing the receivedreceive packets(oratleast as much ofthis address. In thereceived packet thatworse case, theMTU allows to fit in). One ofresponder can track down thegoalsattacker using this address. The drawback of thisnotificationapproach isto allow the other endthatstill retains context state, to reestablish the lost context. The mechanism to reestablish the loss context consists in performing the 4-way initial handshake. This isit imposes atime consuming4 packet exchangeand at this point time mayfor any context establishment. This would becritical since we are reestablishinga great deal if the shim contextthat is currentlyneeded(because context loss detection may occur after a failure). So, another option, which is the one used in this protocol, istoreplace the error message by a modified R1 message, so thatbe established up front, before thetime requiredcommunication can proceed. However, thanks toperform thedeferred context establishmentexchange can be reduced. Upon the receptioncapability ofthis modified R1 message,theend that stillshim protocol, this limitation hasthe context state can finish the context establishment exchange and restore the lost context. Appendix F.4 Securing locator sets The adoption ofaprotocol like SHIM that allowsreduced impact in thebindingperformance of the protocol. (It may however have agiven ULID with a set of locators opensgreater impact in thedoors for different typessituation ofredirection attackscontext recover asdescribed in [19]. The goaldiscussed earlier, but interms of security for the design of the shim protocolthis case, it isnotpossible tointroduce any new vulnerability inperform optimizations to reduce theInternet architecture. It isnumber of packets as described above) The other option considered was anon-goal to provide additional protection than2-way handshake with thecurrently availablepossibility to fall back to a 4-way handshake inthe single-homed IPv6 Internet.case of attack. In Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page111]112] Internet-Draft Shim6 ProtocolMarchMay 2006Multiple security mechanisms were considered to protect the shim protocol. Inthisappendix we will present some of them. The simplest option to protect the shim protocol was to use cookies i.e. a randomly generated bit string that is negotiated duringapproach, thecontextULID-pair establishmentphase and then it is includedexchange normally consists infollowing signaling messages. By this mean,a 2-packet exchange and itwould be possible todoes not verify that thepartyinitiator has performed a prior contact before creating context state. In case thatwas involved ina DoS attack is detected, theinitialresponder falls back to a 4-way handshakeissimilar to thesame party thatone described previously in order to prevent the detected attack to proceed. The main difficulty with this attack isintroducing new locators. Moreover, before usinghow to detect that anew locator, an exchangeresponder isperformed through the new locator, verifying that the party located at the new locator knows the cookie i.e.currently under attack. It should be noted, that because this is 2-way exchange, it is not possible to use thesame partynumber of half open sessions (as in TCP) to detect an ongoing attack and different heuristics need to be considered. The design decision taken was thatperformedconsidering theinitial handshake. While this security mechanisms does indeed provide a fair amountcurrent impact ofprotection, it does leaveDoS attacks and thedoor open forlow impact of theso-called time shifted attacks. In these attacks, an attacker that once was on4-way exchange in thepath, it discoversshim protocol thanks to thecookie by sniffing any signaling message. After that,deferred context establishment capability, a 4-way exchange would be adopted for theattacker can leavebase protocol. Appendix D.6. Updating locator sets There are two possible approaches to thepathaddition andstill perform a redirection attack, since as he is in possessionremoval of locators: atomic and differential approaches. The atomic approach essentially send thecookie, he can introducecomplete locators set each time that anew locatorvariation in the locator setand he can also successfully performoccurs. The differential approach send thereachability exchange if he is able to receive packets atdifferences between the existing locator set and the newlocator.one. Thedifference with the current single-homed IPv6 situation is that in the current situationatomic approach imposes additional overhead, since all theattacker needslocator set has to beon-path duringexchanged each time while thewhole lifetimedifferential approach requires re-synchronization ofthe attack, while in this new situation where only cookie protection if provided, an attackerboth ends through changes i.e. thatonce was on the path can perform attacks after he has left the on-path location. Moreover, becauseboth ends have thecookie is included in signaling messages,same idea about what theattacker can discovercurrent locator set is. Because of thecookiedifficulties imposed bysniffing any of them, makingtheprotocol vulnerable duringsynchronization requirement, thewhole lifetimeatomic approach was selected. Appendix D.7. State Cleanup There are essentially two approaches for discarding an existing state about locators, keys and identifiers ofthe shim context. A possible approach to increase the security was to use a shared secret i.e.abit string that is negotiated during the initial handshake but that is used ascorrespondent node: akey to protect following messages. With this technique, the attacker must be present oncoordinated approach and an unilateral approach. In thepath sniffing packets duringunilateral approach, each node discards theinitial handshake, since it isinformation about theonly moment whereother node without coordination with theshared secretother node based on some local timers and heuristics. No packet exchange isexchanged. Whilerequired for this. In thisimproves the security, it is still vulnerable to time shifted attacks, even thoughcase, itimposeswould be possible that one of theattacker must be on path at a very specific moment (the establishment phase) to actually be able to launchnodes has discarded theattack. While this seems to substantially improvestate while thesituation, it should be noted that, depending on protocol details, an attackerother node still hasn't. In this case, a No-Context error message may beablerequired toforceinform about therecreation ofsituation and possibly a recovery mechanism is also needed. A coordinated approach would use an explicit CLOSE mechanism, akin to theinitialone specified in HIP [25]. If an explicit CLOSE handshake(for instance by blocking messagesandmaking the parties think that the context has been lost), so the resulting situation may not differ that much from the cookie based approach. Another option that was discussed during the design of the protocolNordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page112]113] Internet-Draft Shim6 ProtocolMarchMay 2006was the possibility of using IPsec for protecting the shim protocol. Now, the problem under consideration in this scenario is how to securely bind an address that is being used as ULID with a locator set that can be used to exchange packets. The mechanism provided by IPsec to securely bind the address used with the cryptographic keysassociated timer isthe usage of digital certificates. This implies that an IPsec based solutionused, then there wouldrequire that the generation of digital certificates that bind the key and the ULID byno longer be acommon third trusted partyneed forboth parties involved in the communication. Considering thatthescope of applicationNo Context Error message due to a peer having garbage collected its end of theshim protocolcontext. However, there isglobal, this would imply a global public key infrastructure. The major issues with this approach are the deployment difficulties associated withstill potentially aglobal PKI. Finally two different technologies were selectedneed toprotect the shim protocol: HBA [7] and CGA [6]. These two approaches provide a similar level of protection but they provide different functionality withhave adifferent computational cost. The HBA mechanism relies on the capability of generating allNo Context Error message in theaddressescase of amultihomed host as an unalterable setcomplete state loss ofintrinsically bound IPv6 addresses,the peer (also known asan HBA set. In this approach, addresses incorporateacryptographic one-way hash of the prefix-set available into the interface identifier part. The result iscrash followed by a reboot). Only if we assume that thebinding between allreboot takes at least theavailable addressesCLOSE timer, or that it isencoded within the addresses themselves, providing hijacking protection. Any peer usingok to not provide complete service until CLOSE timer minutes after theshim protocol nodecrash, canefficiently verify thatwe completely do away with thealternative addresses proposedNo Context Error message. In addition, other aspect that is relevant forcontinuingthis design choice is thecommunication are boundcontext confusion issue. In particular, using an unilateral approach to discard context state clearly opens theinitial address through a simple hash calculation. A limitationpossibility ofthe HBA technique is that once generated the address set is fixed and cannot be changed without also changing all the addressescontext confusion, where one of theHBA set. In other words,ends unilaterally discards theHBA techniquecontext state, while the peer doesnot support dynamic addition of address to a previously generated HBA set. An advantage ofnot. In thisapproach iscase, the end thatit requires only hash operations to verify a locator set, imposing very low computational cost tohas discarded theprotocol. In a CGA based approachstate can re-use theaddresscontext tag value usedas ULID isfor the discarded state for aCGA that containsanother context, creating ahash ofpotential context confusion situation. In order to illustrate the cases where problems would arise consider the following scenario: o Hosts A and B establish context 1 using CTA and CTB as context tags. o Later on, A discards context 1 and the context tag value CTA becomes available for reuse. o However, B still keeps context 1. This would become apublic keycontext confusion situation inits interface identifier. The resultthe following two cases: o A new context 2 isa secure bindingestablished betweentheA and B with a different ULID pair (or Forked Instance Identifier), and A uses CTA as context tag, If theassociated key pair. This allows each peer to use the corresponding private key to sign the shim messages that conveylocatorset information. The trust chainsets used for both contexts are not disjoint, we are inthis casea context confusion situation. o A new context isthe following: the ULID usedestablished between A and C and A uses CTA as context tag value for this new context. Later on, B sends Payload extension header and/or control messages containing CTA, which could be interpreted by A as belonging to context 2 (if no proper care is taken). Again we are in a context confusion situation. One could think that using a coordinated approach would eliminate these context confusion situations, making thecommunicationprotocol much simpler. However, this issecurely bound tonot thekey paircase, becauseit containseven in thehashcase ofthe public key, and the locator seta coordinated approach using a CLOSE/CLOSE ACK exchange, there isbound to the public key throughstill thesignature. The CGA approach then supports dynamic additionpossibility ofnew locators in the locator set, since in order to do that,a host rebooting without having thenode only needstime tosign the new locator withperform theprivate key associated withCLOSE exchange. So, it is true that theCGA used as ULID. A limitation ofcoordinated Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page113]114] Internet-Draft Shim6 ProtocolMarchMay 2006thisapproachis that it imposes systematic usage of public key cryptography with its associate computational cost. Anyeliminates the possibility ofthese two mechanisms HBA and CGA provide time-shifted attack protection, sincea context confusion situation because premature garbage collection, but it does not prevents theULID is securely boundsame situations due to alocator set that can only be defined by the ownercrash and reboot of one of theULID. So, the design decision adopted wasinvolved hosts. The result is thatboth mechanisms HBAeven if we went for a coordinated approach, we would still need to deal with context confusion andCGA are supported, so that when only stable address sets are required,provide thenodes can benefitmeans to detect and recover fromthe low computational cost offered by HBA while when dynamic locator sets are required,thiscan be achieved through CGAs with an additional cost. Moreover, because HBAs are definedsituations. Nordmark & Bagnulo Expires November 16, 2006 [Page 115] Internet-Draft Shim6 Protocol May 2006 Appendix E. Change Log [RFC Editor: please remove this section] The following changes have been made since draft-ietf-shim6-proto-04: o Defined I1_RETRIES_MAX asa CGA extension, the addresses available4. o Added text ina node can simultaneously be CGAs and HBAs, allowingsection 7.9 clarifying theusage ofno per context state is stored at theHBAreceiver in order to reply an I1 message. o Added text in section 5 andCGA functionality when needed without requiring a changeinthe addresses used. Appendix F.5 ULID-pair context establishment exchange Twosection 5.14 in particular, on defining additional optionswere considered for the ULID-pair context establishment exchange: a 2-way handshake(including critical anda 4-way handshake. A key goal fornon critical options). o Added text in thedesign of this exchange was that protection against DoS attacks. The attack under consideration was basically a situation where an attacker launches a great amount of ULID-pair establishment request packets, exhausting victim's resources, similarsecurity considerations about threats related toTCP SYN flooding attacks. A 4 way-handshake exchange protects against these attacks becausesecret S for generating thereceiver does not creates any state associatevalidators and recommendation toa given context until the reception of the second packet which contains a prior contact proofchange S periodically. o Added text in theform of a token. At this point the receiver can verify that at least the address used by the initiator is at some extent valid, since the initiator is able to receive packets at this address. In the worse case, the responder can track downsecurity considerations about theattacker using this address. The drawbackeffects ofthis approach is that it imposes a 4 packet exchange for any context establishment. This would be a great deal ifattacks based on guessing theshimcontextneededtag being similar tobe established up front, beforespoofing source addresses in thecommunication can proceed. However, thanks to deferred context establishment capabilitycase ofthe shim protocol, this limitation haspayload packets. o Added clarification on what areduced impactrecent nonce is inthe performanceI2 and I2bis. o Removed (empty) open issues section. o Editorial corrections. The following changes have been made since draft-ietf-shim6-proto-03: o Editorial clarifications based on comments from Geoff, Shinta, Jari. o Added "no IPv6 NATs as an explicit assumption. o Moving some things out of theprotocol. (It may however have a greater impact in the situation of context recover as discussed earlier, but in this case, it is possible to perform optimizationsIntroduction and Overview sections toreduceremove all SHOULDs and MUSTs from there. o Added requirement that any Locator Preference options which use an element length greater than 3 octets have thenumberalready defined first 3 octets ofpackets as described above) The other option considered wasflags, priority and weight. o Fixed security hole where a2-way handshake with the possibility to fall backsingle message (I1) could cause CT(peer) to be updated. Now a4-waythree-way handshakein case of attack. Inis required before CT(peer) is updated for an existing context. The following changes have been made since draft-ietf-shim6-proto-02: Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page114]116] Internet-Draft Shim6 ProtocolMarchMay 2006this approach, the ULID-pair establishment exchange normally consists in a 2-packet exchange and it does not verify thato Replaced theinitiator has performed a prior contact before creating context state. In case that a DoS attack is detected,Context Error message with theresponder falls back to a 4-way handshake similar toR1bis message. o Removed theone described previouslyPacket In Error option, since it was only used inorder to preventthedetected attack to proceed. The main difficulty with this attack is how to detect thatContext Error message. o Introduced aresponder is currently under attack. It should be noted, that because this is 2-way exchange, itI2bis message which isnot possible to use the number of half open sessions (assent inTCP)response todetectanongoing attack and different heuristics need to be considered. The design decision taken was that considering the current impact of DoS attacks and the low impact ofI1bis message, since the4-way exchangeresponders processing is quite in this case than in theshim protocol thanks toregular R1 case. o Moved thedeferred context establishment capability, a 4-way exchange would be adoptedpacket formats for thebase protocol. Appendix F.6 Updating locator sets There are two possible approachesKeepalive and Probe message types and Event option to [8]. Only theaddition and removal of locators: atomicmessage type values anddifferential approaches. The atomic approach essentially sendoption type value are specified for those in this document. o Removed thecomplete locators set each time thatunused message types. o Added avariationstate machine description as an appendix. o Filled inthe locator set occurs. The differential approach send the differences between the existing locator set and the new one. The atomic approach imposes additional overhead, sinceall thelocator set has to be exchanged each time while the differential approach requires re-synchronization of both ends through changes i.e. that both ends have the same idea about what the current locator set is. Because of the difficulties imposed by the synchronization requirement,TBDs - except theatomic approach was selected. Appendix F.7 State Cleanup There are essentially two approaches for discarding an existing state about locators, keys and identifiersIANA assignment ofa correspondent node: a coordinated approach and an unilateral approach. In the unilateral approach, each node discardstheinformation aboutprotocol number. o Specified how context recovery and forked contexts work together. This required theother node without coordination withintroduction of a Forked Instance option to be able to tell which of possibly forked instances is being recovered. o Renamed theother node based on"host-pair context" to be "ULID-pair context". o Picked somelocalinitial retransmit timersand heuristics. No packet exchange is requiredforthis. In this case, it would be possible that oneI1 and I2; 4 seconds. o Added timer values as protocol constants. The retransmit timers use binary exponential backoff and randomization (between .5 and 1.5 of thenodes has discarded the state whilenominal value). o Require that theother node still hasn't. In this case, a No-Context error message mayR1/R1bis verifiers berequiredusable for some minimum time so that the initiator knows for how long time it can safely retransmit I2 before it needs toinform aboutgo back to sending I1 again. Picked 30 seconds. o Split thesituationmessage type codes into 0-63, which will not generate R1bis messages, andpossibly a recovery mechanism is also needed. A coordinated approach would use an explicit CLOSE mechanism, akin64-127 which will generate R1bis messages. This allows extensibility of the protocol with new message types while being able to control when R1bis is generated. o Expanded theone specifiedcontext tag from 32 to 47 bits. o Specified that enough locators need to be included inHIP [25]. If an explicit CLOSE handshakeI2 and R2 messages. Specified that the HBA/CGA verification must be performed when the locator set is received. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page115]117] Internet-Draft Shim6 ProtocolMarchMay 2006associated timer is used, then there would no longer be a needo Specified that ICMP parameter problem errors are sent in certain error cases, for instance when theNo Context Error message dueverification method is unknown toa peer having garbage collected its end ofthecontext. However,receiver, or there isstill potentially a needan unknown message type or option type. o Renamed "payload message" to be "payload extension header". o Many editorial clarifications suggested by Geoff Huston. o Modified the dispatching of payload extension header to only compare CT(local) i.e., not compare the source and destination IPv6 address fields. The following changes havea No Context Error message inbeen made since draft-ietf-shim6-proto-00: o Removed thecaseuse ofa complete state lossthe flow label and the overloading of thepeer (also known as a crash followed by a reboot). Only if we assume thatIP protocol numbers. Instead, when thereboot takes at leastlocator pair is not theCLOSE timer, orULID pair, the ULP payloads will be carried with an 8 octet extension header. The belief is that it isokpossible tonot provide complete service until CLOSE timer minutes afterremove these extra bytes by defining future shim6 extensions that exchange more information between thecrash, can we completely do awayhosts, without having to overload the flow label or the IP protocol numbers. o Grew the context tag from 20 bits to 32 bits, with theNo Context Error message. In addition, other aspect thatpossibility to grow it to 47 bits. This implies changes to the message formats. o Almost by accident, the new shim6 message format isrelevantvery close to the HIP message format. o Adopted the HIP format for the options, since thisdesign choice ismakes it easier to describe variable length options. The original, ND-style, option format requires internal padding in thecontext confusion issue. In particular, using an unilateral approachoptions todiscard context state clearly opensmake them 8 octet length in total, while thepossibility of context confusion, where one ofHIP format handles that using theends unilaterally discardsoption length field. o Removed some of thecontext state, whilecontrol messages, and renamed thepeer does not. In this case,other ones. o Added a "generation" number to theendLocator List option, so thathas discardedthestatepeers canre-use the context tag value used forensure that thediscarded state for a another context, creating a potential context confusion situation. In orderpreferences refer toillustratethecases where problems would arise considerright "version" of thefollowing scenario: o Hosts A and B establish context 1 using CTA and CTB as context tags.Locator List. oLater on, A discards context 1In order for FBD and exploration to work when there thecontext tag value CTA becomes available for reuse. o However, B still keeps context 1. This would become a context confusion situation inuse of thefollowing two cases: o A newcontext2isestablished between A and B with aforked, that is differentULID pair (or Forked Instance Identifier), and A uses CTA as context tag, If the locator sets used for both contextsULP messages arenot disjoint, wesent over different locator pairs, things areinacontext confusion situation. o A new contextlot easier if there isestablished between A and C and A uses CTA as context tag valueonly one current locator pair used forthis neweach context.Later on, B sends Payload extension header and/or control messages containing CTA, which could be interpreted by A as belonging to context 2 (if no proper care is taken). Again we are in a context confusion situation. One could think that using a coordinated approach would eliminate these context confusion situations, making the protocol much simpler. However, this is not the case, because even inThus thecaseforking ofa coordinated approach using a CLOSE/CLOSE ACK exchange, there is stillthepossibility ofcontext is now causing ahost rebooting without having the timenew context toperformbe established for theCLOSE exchange. So, it is true thatsame ULID; thecoordinatednew context having a new context tag. The Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page116]118] Internet-Draft Shim6 ProtocolMarchMay 2006approach eliminates the possibility of aoriginal contextconfusion situation because premature garbage collection, but it does not prevents the same situations due to a crash and reboot of one of the involved hosts. The resultisthat even if we went for a coordinated approach, we would still needreferred todeal withas the "default" contextconfusion and providefor themeans to detectULID pair. o Added more background material andrecover from this situations.textual descriptions. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page117]119] Internet-Draft Shim6 ProtocolMarchMay 2006 19. References19.119.1. Normative References [1] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [2] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, December 1998. [3] Narten, T., Nordmark, E., and W. Simpson, "Neighbor Discovery for IP Version 6 (IPv6)", RFC 2461, December 1998. [4] Thomson, S. and T. Narten, "IPv6 Stateless Address Autoconfiguration", RFC 2462, December 1998. [5] Conta, A. and S. Deering, "Internet Control Message Protocol (ICMPv6) for the Internet Protocol Version 6 (IPv6) Specification", RFC 2463, December 1998. [6] Aura, T., "Cryptographically Generated Addresses (CGA)", RFC 3972, March 2005. [7] Bagnulo, M., "Hash Based Addresses (HBA)", draft-ietf-shim6-hba-01 (work in progress), October 2005. [8] Arkko, J. and I. Beijnum, "Failure Detection and Locator Pair Exploration Protocol for IPv6 Multihoming", draft-ietf-shim6-failure-detection-03 (work in progress), December 2005.19.219.2. Informative References [9] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for specifying the location of services (DNS SRV)", RFC 2782, February 2000. [10] 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. [11] Narten, T. and R. Draves, "Privacy Extensions for Stateless Address Autoconfiguration in IPv6", RFC 3041, January 2001. [12] Draves, R., "Default Address Selection for Internet Protocol version 6 (IPv6)", RFC 3484, February 2003. [13] Bagnulo, M., "Updating RFC 3484 for multihoming support", Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page118]120] Internet-Draft Shim6 ProtocolMarchMay 2006 draft-bagnulo-ipv6-rfc3484-update-00 (work in progress), December 2005. [14] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson, "RTP: A Transport Protocol for Real-Time Applications", STD 64, RFC 3550, July 2003. [15] Abley, J., Black, B., and V. Gill, "Goals for IPv6 Site- Multihoming Architectures", RFC 3582, August 2003. [16] Rajahalme, J., Conta, A., Carpenter, B., and S. Deering, "IPv6 Flow Label Specification", RFC 3697, March 2004. [17] Eastlake, D., Schiller, J., and S. Crocker, "Randomness Requirements for Security", BCP 106, RFC 4086, June 2005. [18] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast Addresses", RFC 4193, October 2005. [19] Nordmark, E. and T. Li, "Threats Relating to IPv6 Multihoming Solutions", RFC 4218, October 2005. [20] Huitema, C., "Ingress filtering compatibility for IPv6 multihomed sites", draft-huitema-shim6-ingress-filtering-00 (work in progress), September 2005. [21] Bagnulo, M. and E. Nordmark, "SHIM - MIPv6 Interaction", draft-bagnulo-shim6-mip-00 (work in progress), July 2005. [22] Nordmark, E., "Shim6 Application Referral Issues", draft-ietf-shim6-app-refer-00 (work in progress), July 2005. [23] Abley, J., "Shim6 Applicability Statement", draft-ietf-shim6-applicability-00 (work in progress), July 2005. [24] Huston, G., "Architectural Commentary on Site Multi-homing using a Level 3 Shim", draft-ietf-shim6-arch-00 (work in progress), July 2005. [25] Moskowitz, R., "Host Identity Protocol", draft-ietf-hip-base-05 (work in progress), March 2006. [26] Eronen, P., "IKEv2 Mobility and Multihoming Protocol (MOBIKE)", draft-ietf-mobike-protocol-08 (work in progress), February 2006. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page119]121] Internet-Draft Shim6 ProtocolMarchMay 2006 Authors' Addresses Erik Nordmark Sun Microsystems 17 Network Circle Menlo Park, CA 94025 USA Phone: +1 650 786 2921 Email: erik.nordmark@sun.com Marcelo Bagnulo Universidad Carlos III de Madrid Av. Universidad 30 Leganes, Madrid 28911 SPAIN Phone: +34 91 6248814 Email: marcelo@it.uc3m.es URI: http://www.it.uc3m.es Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page120]122] Internet-Draft Shim6 ProtocolMarchMay 2006 Intellectual Property Statement The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79. Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr. The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at ietf-ipr@ietf.org. Disclaimer of Validity This document and the information contained herein are provided on an "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Copyright Statement Copyright (C) The Internet Society (2006). This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights. Acknowledgment Funding for the RFC Editor function is currently provided by the Internet Society. Nordmark & Bagnulo ExpiresSeptember 5,November 16, 2006 [Page121]123] ----