Internet Engineering Task Force T. Tsou Internet-Draft Huawei Technologies (USA) Intended status: Informational December 11, 2011 Expires: June 4, 2012 Address Acquisition For Multicast Content When Source and Receiver Support Differing IP Versions draft-tsou-multrans-addr-acquisition-00 Abstract In a typical IP television (IPTV) system, the receiver acquires information about available program content, the subscriber selects a program to watch, and the receiver signals to the network to begin receiving the program in the form of multicast content. The program content information is typically XML-encoded, can be transmitted in multiple segments, possibly over multiple channels, but includes media stream descriptions for the individual program descriptions that may also be XML-encoded or may use the Session Description Protocol (SDP, RFC 4566). The media stream descriptions provide multicast group and unicast source addresses that are used in the subsequent signalling to the network. During the transition from IPv4 to IPv6, scenarios can occur where the IP version supported by the receiver differs from that supported by the source. This memo examines and evaluates alternative strategies for allowing the receiver to acquire addresses in such scenarios in the version it supports. Status of this Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. 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." This Internet-Draft will expire on June 4, 2012. Copyright Notice Tsou Expires June 4, 2012 [Page 1] Internet-Draft Multicast Address Acquisition December 2011 Copyright (c) 2011 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 3 2. IPTV System Background . . . . . . . . . . . . . . . . . . . . 3 3. Which Problem Are We Solving? . . . . . . . . . . . . . . . . . 5 4. Possible Solutions . . . . . . . . . . . . . . . . . . . . . . 5 4.1. The Reactive Strategy . . . . . . . . . . . . . . . . . . . 6 4.2. Dynamic Modification . . . . . . . . . . . . . . . . . . . 6 4.3. Administrative Preparation . . . . . . . . . . . . . . . . 7 5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 8 6. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 8 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 8 8. Security Considerations . . . . . . . . . . . . . . . . . . . . 8 9. Informative References . . . . . . . . . . . . . . . . . . . . 8 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 9 Tsou Expires June 4, 2012 [Page 2] Internet-Draft Multicast Address Acquisition December 2011 1. Introduction Discussion of the multicast transition problem has focussed on the IPTV scenario. Within this scenario, the operation of viewing a program follows a well-defined sequence: first the receiver acquires program information that includes the detailed information needed to request delivery of specific content. At some subsequent time the user chooses to view a program, possibly by selecting it from a displayed program guide, or simply by selecting a channel. The receiver uses its pre-acquired information to signal to the network to receive the desired content. In particular, the receiver initiates reception of multicast content using the source and multicast group addresses supplied within the program information. With an all-IPv4 system, it is evident that the program information will include IPv4 source and group addresses only. Suppose now, as can occur in some transition scenarios, that IPv6 receivers appear within the system. Then there will be a mismatch: the IPv6 receivers will be unable to use the addresses that are provided in the program information. This memo examines the possible strategies for remedying this mismatch, evaluating them in terms of their impact on receiver implementation and network operation. 1.1. Terminology This memo uses no mandatory language. The term "receiver" in this document refers to the functionality at the user location that communicates with the network and receives multicast content on behalf of the user. 2. IPTV System Background Numerous organizations have been involved in the development of specifications for IPTV. Those specifications and the requirements of individual providers have influenced the development of existing receivers. Any solution to the multicast transition problem described in Section 1 has to take account of the effort involved not only in the direct development of a new generation of receivers, but also in evolving the specifications on which those receivers are based. It is thus worthwhile to review the current situation as it affects multicast transition. The TV-Anytime forum (http://www.tv-anytime.org/) did early work in the area, formally terminating in 2005. Their work focussed on the description of program content, to facilitate the creation of such descriptions and their navigation by the user. The results are Tsou Expires June 4, 2012 [Page 3] Internet-Draft Multicast Address Acquisition December 2011 documented in the ETSI TS 102 822 series of technical specifications. The content reference identifier (CRID) is a fundamental concept in the TV-Anytime data model. It refers to a specific piece of content or to other CRIDs, the latter thereby providing a method for grouping related pieces of content. TV-Anytime registered the CRID: URL schema in [RFC4078]. Quoting from the abstract of that document: The Uniform Resource Locator (URL) scheme "CRID:" has been devised to allow references to current or future scheduled publications of broadcast media content over television distribution platforms and the Internet. The initial intended application is as an embedded link within scheduled programme description metadata that can be used by the home user or agent to associate a programme selection with the corresponding programme location information for subsequent automatic acquisition. The process of location resolution for the CRID: URL for an individual piece of content locates the content itself so that the user can access it. TV-Anywhere left the details of that process unspecified. The Open IPTV Forum (http://www.oipf.tv) has focussed on defining the user-to-network interface, particularly for fixed broadband access. The architecture is based on the ETSI NGN (Next Generation Networks) model. The receiver uses SIP (Session Initiation Protocol [RFC3261]) signalling to obtain authorization and resources for a session, before signalling at the multicast level to acquire the program. [Further research is needed to determine whether the source and multicast group address are provided in the SDP (Session Description Protocol [RFC4566]) in the response to the SIP request, or whether the receiver already has this information from the electronic program guide before it starts.] Finally, the Open Mobile Alliance (OMA, http://www.openmobilealliance.org/) has defined a series of specifications relating to broadcast services over wireless networks. The source and multicast group addresses used to acquire a given program instance are provided in SDP fragments either directly embedded in the primary electronic program guide or pointed to by it. The OMA architecture provides functionality to adapt access information within the program guide to the requirements of the transport network to which the user is attached, but this functionality appears to be primarily directed toward overcoming differences in technology rather than a general capability for modification. Tsou Expires June 4, 2012 [Page 4] Internet-Draft Multicast Address Acquisition December 2011 In conclusion, it appears that there are at least two extant sources of specifications for the receiver interface, each providing its own data model, XML data schema, and detailed architecture. In the OMA case, the access information including the source and multicast group addresses is definitely embedded as an SDP fragment within a larger set of XML-encoded program metadata. [Further research needed for the Open IPTV Forum, as already noted.] The OMA metadata can be supplied to the receiver in multiple segments, through multiple channels. This complicates the task of intercepting that metadata and modifying it in a particular transport network. 3. Which Problem Are We Solving? The problem addressed by this memo was stated fairly clearly (the authors trust) above, but it does not hurt to further clarify the issue being addressed. In some transition scenarios, the source supports one IP version while the receiver and the provider network support the other (e.g., the 6-6-4 scenario). In this case, the problem stated above is unambiguous: how are addresses of the version supported by the receiver delivered to the receiver, possibly with the help of the provider network? In other transition scenarios, the source and provider network support one IP version while the receiver supports another. In this case there are actually two problems: how to get provided addresses to the receiver that it understands (as already stated), and how to make those addresses usable in a network supporting a different version? This second problem is the subject of a different memo and out of scope of the present one. There is also a third class of scenarios, where the source and receiver support the same IP version but the intervening network supports a different one (e.g., the 4-6-4 scenario). In those scenarios, delivering addresses of the right IP version to the receiver is notionally a non-problem. The problem still can arise, if the intervening network intercepts and modifies the access information to be consistent with the IP version it supports. In this case, the problem can be re-stated as: how can such modification by avoided when it is not needed? 4. Possible Solutions This section explores three classes of solution to the problem at hand: Tsou Expires June 4, 2012 [Page 5] Internet-Draft Multicast Address Acquisition December 2011 o reactive: the receiver recognizes that addresses it has received are in the wrong version and converts them through a request to a mapping function; o dynamic modification: the network intercepts the access information and modifies it as necessary to meet the requirements of the receiver; o administrative: the electronic program guide is modified in advance of acquisition to provide alternative address versions. Two variations on this strategy are identified. 4.1. The Reactive Strategy According to this strategy, an IPv6 receiver receiving IPv4 addresses, for example, would recognize that they were the wrong version. It would package the addresses into one or two requests to a mapping function, which would return corresponding IPv6 addresses. In the 6-4-4 scenario cited above, the mapping function could be part of the user site or located in a dual-stack element at the provider edge. In the 6-6-4 case it would have to be part of the provider network, although not necessarily at its edge. This strategy clearly involves a fair amount of work to implement. Not only does the receiver need to recognize that addresses are the wrong version; it also has to implement a new protocol to the mapping function. It also has to discover that function. The mapping function itself is probably needed to solve other aspects of the multicast transition problem, so should not be considered a cost of the reactive strategy in particular. 4.2. Dynamic Modification This strategy puts the entire burden of address adaptation on the provider network. It requires that an element in that network intercept program guide information destined to the receiver, locate the access information, and translate IP address versions as necessary to suit the receiver. If the problem identified in the last paragraph of Section 3 is to be avoided, the intercepting element has to be aware of the version supported by each receiver. As noted in the description of the OMA architecture, it is possible that such an adaptive function is present, but not clear that its scope would extend to IP version changes. The need to include IP version along with other receiver-related information might or might not prove to be administratively demanding. With the dynamic modification strategy the workload on the adaptation function might be large enough to make it a bottleneck in the process of program Tsou Expires June 4, 2012 [Page 6] Internet-Draft Multicast Address Acquisition December 2011 acquisition. The mitigating factor is that program metadata will typically be retrieved rather less often than program content. This strategy has the clear advantage that it requires no changes in the receiver. 4.3. Administrative Preparation The basic idea with this strategy is that the access information in the program metadata is set up to provide the right address version in advance of acquisition by any receiver. There are two basic approaches: o separate alternative versions of the access information are prepared. The correct version is served up to the receiver when it requests it. Like the dynamic modification strategy, this approach assumes that it is administratively feasible for the program guide server to know the IP version of the requesting receiver. That may or may not be true in a given operator's context. Also as with the dynamic modification approach, no change is required in the receiver. The big advantage over dynamic modification is that there is no need for the complications of an intercepting adapting element. o The same access information instance contains alternative IP address versions. Where SDP is used, we can think of ICE or ICE- lite [RFC5245] or the proposed altc mechanism [I_D.boucadair-altc]. This requires receiver modification to recognize the alternative syntax and potentially to take part in STUN exchanges. However, it means that the same access information can be served up to all receivers in a backward- compatible manner. The administrative strategy requires that the network provider have control over the translations used in the preparation of the alternativ e versions of the access information. The network has to be aware of the translations used, so it can reuse them at other stages of the multicast acquisition process. The case where the receiver supports a different IP version from the network to which it is attached and the adaptation between versions occurs in the customer network (e.g., 6-4-4 case with dual stack customer edge device) is not amenable to the administrative approach, unless the operator is able to control the mapping used by the customer edge device. Failing that, it appears that the reactive strategy is the only one that would work in this particular case. Tsou Expires June 4, 2012 [Page 7] Internet-Draft Multicast Address Acquisition December 2011 5. Conclusions To come. 6. Acknowledgements TBD 7. IANA Considerations This memo includes no request to IANA. 8. Security Considerations To come. 9. Informative References [I_D.boucadair-altc] Boucadair, M., Kaplan, H., Gilman, R., and S. Veikkolainen, "Session Description Protocol (SDP) Alternate Connectivity (ALTC) Attribute", November 2011. [MPEG-7_DDL] ISO/IEC, "ISO/IEC 15938-2 (2002): "Information technology - Multimedia content description interface - Part 2: Description definition language".", 2002. [RFC3261] Rosenberg, J., Schulzrinne, H., Camarillo, G., Johnston, A., Peterson, J., Sparks, R., Handley, M., and E. Schooler, "SIP: Session Initiation Protocol", RFC 3261, June 2002. [RFC4078] Earnshaw, N., Aoki, S., Ashley, A., and W. Kameyama, "The TV-Anytime Content Reference Identifier (CRID)", RFC 4078, May 2005. [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session Description Protocol", RFC 4566, July 2006. [RFC5245] Rosenberg, J., "Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols", RFC 5245, April 2010. Tsou Expires June 4, 2012 [Page 8] Internet-Draft Multicast Address Acquisition December 2011 Author's Address Tina Tsou Huawei Technologies (USA) 2330 Central Expressway Santa Clara, CA 95050 USA Phone: +1 408 330 4424 Email: tina.tsou.zouting@huawei.com Tsou Expires June 4, 2012 [Page 9]