Internet Engineering Task Force Jia Jia Liao Internet Draft Ping Zhang Expires: October 2006 Zheng Bin Li An Shi Xu National Laboratory on Local Fiber-Optic Communication Network & Advanced Optical Communication System Peking University, China April 2006 Recovery in Optical Burst Switching Network draft-liao-mpls-obs-00.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. Copyright Notice Copyright (C) The Internet Society (2006). All Rights Reserved. Abstract Protection and restoration at optical layer is critical to network integrity since data is transmitted and switched at a considerably high speed in optical domain. A few second halt may cause tens to thousands gigabit loss. Optical burst switching is a promising technology, bridging optical circuit switching and optical package switching. Unlike optical circuit switching and time division multiplexing, OBS is featured with unidirectional reservation and statistical multiplexing of wavelength resources. The general idea behind protection and restoration techniques is to utilize redundant bandwidth resources as backup. The flexibility brought by OBS provides alternatives for existed protection and restoration schemes at optical layer. Liao,Zhang,Li,Xu [Page 1] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Conventions The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC2119 [RFC 2119]. Liao,Zhang,Li,Xu [Page 2] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Table of Contents 1. Introduction.....................................................4 1.1. OBS Network Architecture....................................4 1.2. Multi-layer Interoperation..................................5 2. Fault Management.................................................5 2.1. Failure Profiles............................................6 2.1.1. Link Failure..........................................6 2.1.2. Node Failure..........................................6 2.2. Fault Detection.............................................7 2.2.1. Link Fault Detection..................................7 2.2.2. Node Fault Detection..................................7 2.3. Fault Notification..........................................7 2.3.1. Link Fault Notification...............................7 2.3.2. Node Fault Notification...............................8 3. Restoration at OBS Layer.........................................9 3.1. Motivation to Restore at OBS Layer..........................9 3.2. Single Layer Restoration Schemes...........................10 3.2.1. Link Failure Restoration Schemes.....................10 3.2.2. Node Failure Restoration Schemes.....................10 3.3. Multi-layer Restoration Schemes............................11 3.3.1. IP Dynamic Routing...................................11 3.3.2. MPLS Protection Switching............................11 3.3.3. Optical Layer Resilient Schemes......................12 3.3.4. Recovery Scheme Comparison...........................12 3.3.5. Operational Coordination.............................12 4. Acknowledgements................................................13 5. References......................................................13 6. AUTHORS' ADDRESSES..............................................13 7. IPR NOTICE......................................................13 8. FULL COPYRIGHT STATEMENT........................................14 Liao,Zhang,Li,Xu [Page 3] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 1. Introduction The basic difference among Optical Circuit Switching (OCS), Optical Burst Switching (OBS) and Optical Packet Switching (OPS) is that the three work at different granularity. OCS, which has already been widely deployed, aims to switch at wavelength, waveband or even fiber level. However, in most cases, individual users can hardly afford a whole wavelength, and thus TDM is applied to provide each channel with fixed percentage of the total bandwidth by splitting wavelength into recurring time-slots. This approach has been proved to be less bandwidth efficient than OPS, which is able to switch packets like IP network in optical domain. However, some critical technologies essential to OPS, such as optical random access memory, are far away from maturity. OBS, supposed to bridge above two mechanisms, is able to switch bufferlessly at sub-wavelength level. OBS, featured with unidirectional reservation and statistical multiplexing of wavelength resources, has brought great flexibility to optical bandwidth distribution. 1.1. OBS Network Architecture OBS network is composed of two sub-planes, namely data plane and control plan, as shown in Figure 1. In data plane, traffic from OBS client layer (e.g. IP or ATM layer) is aggregated into Data Bursts (DBs) at ingress edge nodes which perform as an interface to the upperlayer and local at the edge of OBS layer [IPOWDM]. DBs will be sent through core nodes to their egress edge nodes without o-e-o conversion, in which a few optical fiber delay lines may be applied to reduce overall blocking probability. DBs usually contains tens to thousands of thousand bits including payload and frame overhead. [ core ] /[ node ]\ / | \ / |------| \ / | -\/- | \ | [ core ]/ /| -/\- |\ \[ core ] | from |------[ node ]\ / |------| \ /[ node ]------| to upperlayer| | \/ \/ | |upperlayer ------->| |------| /\ /\ |------| |--------> ------->|------| -\/- |/ \ / \| -\/- |------|--------> ------->| | -/\- |\ \[ core ] /| -/\- | |--------> traffic | |------| \ [ node ] / |------| | traffic | \ | / | Ingress edge node \ |------| / Egress edge node \| -\/- |/ | -/\- | |------| Figure 1: OBS Network Architecture Liao,Zhang,Li,Xu [Page 4] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Control plane mainly conduct routing and resource reservation by configuring optical switching fabric according to signalling. An offset time ahead of DB transmission, Control Packets (CPs) will be sent through core nodes to establish an available light path from ingress edge node to the egress one. At each stop at core nodes, CPs will experience o-e-o conversion and be processed electrically to trigger switching fabric. As long as every configuration is successful, DB is able to transmit across the network. However, once one of the core nodes along light path fails to act, the DB will have to be discarded and bandwidth that has already been reserved will be released. 1.2. Multi-layer Interoperation OBS layer, viewed as an data-link layer, aims to provide reliable end-to-end path for data transmission. As shown in Figure 2, its server layer is optical layer with huge physical bandwidth and its client layer can be network layer (e.g. IP) or others like ATM. With the increase in volume and importance of IP traffic, applications based on IP has become dominant. Thus in this draft, we only consider IP as the client layer, for which OBS acts to provide available bitpipe. |+++++++++++++++++++| |+++++++++++++++++++| | Application Layer |<-------->| Application Layer | |+++++++++++++++++++| |+++++++++++++++++++| | IP Layer |<-------->| Network Layer | |+++++++++++++++++++| |+++++++++++++++++++| | OBS Layer |<-------->| Data-link Layer | |+++++++++++++++++++| |+++++++++++++++++++| | Optical Layer |<-------->| Physical Layer | |+++++++++++++++++++| |+++++++++++++++++++| Figure 2: Layered Network Optical layer, lying under OBS layer, focuses on optical signal transmission, amplifying, multiplexing and demultiplexing. From the view of OBS, optical layer offers Optical Channel-Path (OCh-P), connecting distributed OBS nodes. OCh-P represents the end-to-end transport of a lightpath across multiple regenerators in the path [Optical]. OBS nodes are classifies as edge nodes and core nodes. Edge nodes consist of aggregating queues, CPs' generator and CPs' and DBs' transmitter or receiver. Core nodes comprise CPs' processor, switch driver, and switching fabric. 2. Fault Management Liao,Zhang,Li,Xu [Page 5] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Fault management involves detecting problems in the network and alerting the management systems appropriately through alarms. If a certain parameter is being monitored and its value falls outside its present range, the network equipment generates an alarm [Optical]. In other cases, alarms could also be triggered by outright failures, such as the failure of switch driver or other components in the system. Fault management also includes restoring service in the event of failure, but we organize the latter as a separate section. 2.1. Failure Profiles Various failures may occur in a multi-layer network. Some are caused by OBS network elements; Some are not but may be restored by OBS layer mechanisms. Providing an exhaustive list of all the possible failure types is aimless, but it is worth listing the main categories of failures, namely link failures and node failures, from OBS perspective. 2.1.1. Link Failures Several types of failures may result in OBS link failure. Fiber cut may cause all the OCh-Ps in a physical link to fail. If optical layer refuses to provide protection, it can be passed on to OBS layer. Optical equipment failures such as amplifier failure also belong to optical layer, but it may affect several OCh-Ps and decrease the transmission capacity to some degree. Such failure can be protected at optical layer too. OBS node interface failure may occur in signalling channel, as unlike DBs, CPs have to experience o-e-o conversion and electrical process at core nodes. However, such type of failure can not be protected by optical layer. (TBC) 2.1.2. Node Failures There are multiple possible causes of node failures whose nature has very different implications. Power supply outage provokes both a control and switching plane failure. But in most cases, backup power supplier will take over to work. Switching fabric failure at core nodes would cause traffic loss or mis-forwarding. So measures should be taken to monitor switching fabric status and report to the control model or even to other nodes. Software failures make impact on some specific features or software crash of the node operating system, for example, packets' aggregation failure at edge nodes or CPs' process failure at core nodes. Liao,Zhang,Li,Xu [Page 6] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Hence, besides software bug eradication by testing, a modular architecture may limit software component failures to a controllable scope and the failed software component may be restarted independently of the other modules. (TBC) 2.2. Fault Detection Fault detection is the first phase in recovery cycle and its time is part of total recovery time. This time may depend, for instance, on the speed of fault detection in a lower network layer and notification toward upper layers, on the time it takes for the node to gather all abnormal information from various signals and derive the exact fault state from diagnosis and so on. 2.2.1. Link Fault Detection According to link failure profiles, most link failures can be detected at optical layer from the perspective of OBS, for example by optical channel-path trace. This trace can be inserted at the end of the CPs and monitored at various locations at control plane along the lightpath. Moreover, Optical receivers at each node can perform as detectors at data plane. Once the optical signal-to-noise ratio falls below a threshold, alarms would be triggered. (TBC) 2.2.2. Node Fault Detection When comes to node failure, in case of switching fabric failure, core node may provide the function of self supervising at optical layer. But, unlike link failures, most other node failures can hardly be detected at optical layer and thus lower layer failure notification is not suitable here. However, the mechanism based on hello protocol could be feasible by sending a periodic hello message between two neighbors. When one of the node stops receiving hello messages for a configurable period, it concludes that a failure of the link between them or the objective node itself has failed. (TBC) 2.3. Fault Notification Once failures have been detected and located, other network elements in the same domain or network should be informed by fault alarming and the propagation time is also part of recovery time. OBS network can be distributedly or centrally controlled. In the former case, routing function is performed by each edge node, while in the latter one, a central controller is responsible for scheduling. Here, we only consider distributed situation. 2.3.1. Link Fault Notification OCh-Ps are usually unidirectional. Thus any failure at OCh-P, such as fiber cut or amplifier outage, will be detected by Recovery Tail-End (RTE) instead of Recovery Head-End (RHE)[Recovery]. On detecting, RTE immediately starts to notify other nodes by Link State Advertisement (LSA). LSA, can be broadcasted or transmitted from end to end. Liao,Zhang,Li,Xu [Page 7] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 In OBS network, edge nodes keep route's table, perform routing and set offset time between CPs and DBs. Thus LSA must be firstly reported to each edge node and every node in the network should keep the routes to each edge node. As LSA is critical to DBs' routing, LSA transmission must be reliable. So each edge node having received LSA must return an acknowledgement to RTE. Sometimes, existed link failure may cut off LSA transmission and thus alternative route should be pre-computed. In case of core node acting deflection, these core nodes need to be alarmed too, otherwise deflected DBs would probably blocked by link failures. (TBC) 2.3.2. Node Fault Notification According to node failure profiles, switching fabric at core node and outright crash will be discussed in this paragraph. Single switch failure may reduce core node switching throughput and may be resolved by substituting the outage switch with a new one. However, when most part of switching fabric fails to work, the total switching node must alarm other nodes about node failure by broadcasting or transmitting Node State Advertisement (NSA) from end to end. Similar to link fault propagation, NSA must be firstly sent to edge node and then to deflective core node, if necessary. upperlayer traffic | 2. NSA |------| 1. Hello |------| <------->| <-------- | core | --------> | core | <------->|------------| node |---------------| node | <------->| |------| no response |------| edge node (a) node failure upperlayer traffic | 2. NSA |------|1. Hello |------| <------->| <-------- | core |-------->X | core | <------->|------------| node |---------------| node | <------->| |------| no response |------| edge node (b) link failure upperlayer traffic | 3. NSA |------| 1. Hello |------| <------->| <-------- | core | --------> | core | <------->|------------| node |---------------| node | <------->| |------| X<-------- |------| 2. Hello edge node (c) link failure Figure 3: Hello Message Loss Liao,Zhang,Li,Xu [Page 8] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 Power supply outage or software OS crash may lead to traffic black-holded and the control model of core node is unable to report failure by it self. Thus, such kind of failure must be discovered and notified of by neighbor nodes sending NSA. Once a neighbor node does not receive hello message from the objective node for a given time, this node should report the failure to others. However, in such case as shown in Figure 3, the neighbor node can hardly distinguish node failure from OCh-P failure. Therefore, such a NSA need to cooperate with a LSA to identify the real problem. (TBC) 3. Restoration at OBS Layer OBS layer performs as a data-link layer located between optical layer and network layer (IP layer). The major duties of this layer is to provide reliable and quick bitpipes for its client layer (network layer) and to make effective utilization of huge bandwidth of its server layer. 3.1. Motivation to Restore at OBS Layer OBS is a promising technology to explore huge optical bandwidth for upperlayer applications. Besides OCh-P failure, network elements outage at OBS layer, such as edge node or signalling process model failure at core node, can be hardly protected by optical layer, though upperlayer traffic can be partly restored by IP layer itself. Thus, to restore at OBS layer could at least provides survivability for OBS network elements. Restoration at IP layer is so versatile to deal with failures at lower layer, but the problem is that total recovery process is rather time-consuming and even can not meet the QoS of real-time traffic. Recovery time of IP restoration mechanisms usually ranges from tens of seconds to minutes. However, restoration at OBS layer is more responsive and faster than its IP's counterpart, as OBS is able to allocate optical bandwidth directly. Protection schemes at optical layer is not mature in mesh topology as compared to ring topology or point-to-point connection. So restoration at OBS layer may offer alternatives for optical protection at a finer granularity. For example, some link failure may be detected at optical layer, but protection mechanisms at the same layer will not be triggered with failure passed up on to OBS layer. Then OBS restoration starts immediately to work. DBs play as containers to carry upper layer packets across OBS switching nodes. At ingress edge node, packets sharing the same destination or QoS could be enclosed into DBs. CPs could help to route according to not only addresses but also service level. IP layer can also provide differentiated service for each packet, but obviously OBS is able to carry out it far more efficient as data can be routed by larger containers, DBs, instead of packets. Liao,Zhang,Li,Xu [Page 9] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 3.2. Single Layer Restoration Schemes Protection or restoration schemes should meet the need of performance evaluation criteria, such as recovery time, bandwidth efficiency and so on. Performance criteria is defined by service level of traffic. Hence a wide range of recovery mechanisms may exist to supplement with each other to serve for various traffic. 3.2.1. Link Failure Restoration Schemes When comes to link failure, traditional optical protection mechanisms aim to seek abundant wavelength resources to establish a new lightpath for data transmission. However, as OBS works at sub-wavelength granularity and is characteristic of statistical multiplexing, in case of failure, what OBS restoration schemes seek is time slot at different wavelength, as the multiplexing density of DBs is controllable, unlike OCS or TDM. One or two channel in a link failing to work means the decrease of transmission capacity of that link. Once nodes are equipped with wavelength conversion, traffic can be easily multiplexed to other channels in the same link without help of additional wavelength. When the transmission capacity of a certain link has decreased below a threshold and become intolerable, traffic that is used to pass this link will have to be deflected or rerouted. For example, cable cut can be considered as total channel failure at that link. Deflection is a local method, in which a new route will be selected by RTE. This new route may be pre-computed or computed on-the-fly. Then RTE transmits LSAs with the new route to each edge nodes by broadcasting or end-to-end transmission. In case of core node able to deflect, these core nodes need to be alarmed too, otherwise deflected DBs would probably blocked by link failures. Rerouting is a global and more radical method by edge nodes selecting a new path. Once a link failure occurs, RTE simply reports the location of the failure to each edge node, which will figure out a new path by pre-computing or computing on-the-fly. Deflecting method is usually faster than rerouting, but rerouting can be more bandwidth effective as it focuses on the global resources [Restoration]. So in practice, two methods may work sequencially or integratedly to collaborate with each other. (TBC) 3.2.2. Node Failure Restoration Schemes Node failure is much severer than link failure, which can obviously cause all the links connected to it to fail and even change the topology of network. So a backup node and backup switching fabric is indispensable. Once failures occur, all traffic could be switched to the backup one. In case of planned node failure, resulted from hardware and software of switching node updating, traffic may be gracefully rerouted to the backup node. Liao,Zhang,Li,Xu [Page 10] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 If no backup node is available, recovery schemes for node failure at OBS layer can only restore traffics forwarded by the fault core node. Traffics that begin or end with the fault node can not be restored. And as well edge nodes can not be restored either. With respect to core nodes, there are two different kinds, namely ingress/egress switching node and intermediate switching node according to JIT protocol. Ingress switching node is the outlet of traffic from edge node and egress switching node is the destination, so any failure at above two nodes is similar to that at edge nodes. Only traffic forwarded by fault intermediate switching node can be restored by deflecting or rerouting. (TBC) 3.3. Multi-layer Restoration Schemes OBS layer is an intermediate layer between optical layer and IP layer. In realistic network, each of them has its own recovery mechanisms. However, not every failure in a particular network layer can be resolved by recovery mechanism in that same layer. Upon detection of a fault, more than one layers could initiate recovery actions. If these recovery mechanisms are merely triggered by detection of a fault, an uncoordinated and inefficient action may result. 3.3.1. IP Dynamic Routing Restoration at IP layer is mainly accomplished by exchanging, between adjacent routers, control messages that are used to update the routers' tables, thus enabling IP packets to be dynamically rerouted around link and node failures. However, it is usually slow, from several to hundreds seconds, and its behavior is unpredictable. Some enhancements of the protocol have been proposed to overcome its drawbacks. One approach is equal cost multi-path forwarding, in which the router relies on more than one path for transmitting packets sharing a common destination by maintaining multiple next-hop entries for the same destination within each router's routing table. Another approach partitions the network into multiple areas, as defined in hierarchical link state routing protocols such as OSPF and IS-IS. 3.3.2. MPLS Protection Switching MPLS protection switching is an alternative approach to circumvent the latency drawback of dynamic routing. MPLS protection entities can be set up either dynamically or in a prenegotiated way. Protection entities, dynamically set up, restore traffic based on failure information, bandwidth allocation, and optimized reroute assignment. Prenegotiated protection consists of working LSPs that have preestablished protection paths. In general, dynamic protection increase resource utilization but requires longer restoration times than preestablished protection. Liao,Zhang,Li,Xu [Page 11] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 3.3.3. Optical Layer Resilient Schemes Both the optical channel (OCh) section and optical multiplex section (OMS) feature dynamic restoration and preplanned protection. The main difference between OCh and OMS resilient schemes is represented by the granularity at which layers operate. OCh resilient schemes protect individual lightpaths, thus allowing selective recovery of optical line terminal failures. OMS resilient schemes work at the aggregated signal level, thus recovering all lightpaths present on the failed line concurrently. Protection schemes, namely Dedicated Path Protection (DPP), Shared Path Protection (SPP), Optical Unidirectional Path Switched Ring (OUPSR), Optical Bidirectional Path Switched Ring (OBPSR) and so on, guarantee service restoration completion times of hundreds, tens and even fractions of milliseconds. However, restoration schemes at optical layer are slower and less mature than protection schemes. 3.3.4. Recovery Scheme Comparison From the view of rerouting, restoration schemes at OBS layer is similar to MPLS protection switching. But MPLS is processing and routing in electronic domain, while DBs in OBS network is switched in optical domain and CPs need o-e-o conversion. In MPLS protection schemes, labels are followed closely by payload and distributed by LDP. However, OBS CPs are set out an offset time prior to BDs and the route is computed at edge nodes according to Dijsktra algorithm. OBS enables optical network to become more flexible and intelligent by enhancing signalling at control plan. DBs with proper size is multiplexed statistically onto a wavelength, which can lead to more efficient utilization of wavelength bandwidth than TDM. So restoration schemes at OBS layer may provide alternatives for traditional optical protection mechanisms, for example, deflection according to DBs' quality of service. 3.3.5. Operational Coordination Coordination between resilient schemes, at distinct layers, is required to avoid multiple schemes concurrently activated upon a single network fault. Two kinds of coordinating strategies are sequential approach and integrated approach [Recovery]. In the former scheme, the server layer may start recovery immediately, whereas the recovery mechanism in the client layer has a build-in hold-off time before initiating the client recovery process. The latter one combines several mechanisms into one integrated multi-layer recovery schemes coordinated by management plane. However, the sequential approach is easier to apply than the integrated one as the latter may cause great complexity to control and management planes. When comes to restoration at OBS layer, as the sequential approach requested, the recovery time should be shorter than that of IP layer and near to that of optical layer, for example Liao,Zhang,Li,Xu [Page 12] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 total recovery time ranging from tens milliseconds to several seconds. 4. Acknowledgements This research is funded by the National High Technology Research and Development Program of China (863 Program). The authors are grateful to other colleagues for their work and useful suggestions. 5. References [Recovery] J. Vasseur et al., "Network Recovery". Morgan Kaufmann Publishers, 2004. [Optical] R. Ramaswami and K. N. Sivarajan, "Optical Networks". Morgan Kaufmann Publishers, 2004. [IPOWDM] S. Dixit, "IP OVER WDM: Building the Next-Generation Optical Internet". WILEY-INTERSCIENCE, 2002. [Restoration] Y. Xin et al., "Fault Management with Fast Restoration for Optical Burst Switched Networks". BROADNETS'04. 6. AUTHORS' ADDRESSES Jia Jia Liao National Laboratory on Local Fiber-Optic Communication Network & Advanced Optical Communication System, Peking University, 100871 P.R. China Email: jjliao@ele.pku.edu.cn Ping Zhang National Laboratory on Local Fiber-Optic Communication Network & Advanced Optical Communication System, Peking University, 100871 P.R. China Email: zhangping@pku.edu.cn Zheng Bin Li National Laboratory on Local Fiber-Optic Communication Network & Advanced Optical Communication System, Peking University, 100871 P.R. China Email: lizhengbin@pku.edu.cn An Shi Xu National Laboratory on Local Fiber-Optic Communication Network & Advanced Optical Communication System, Peking University, 100871 P.R. China Email: lyrxas@pku.edu.cn 7. IPR NOTICE 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 Liao,Zhang,Li,Xu [Page 13] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006 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. 8. FULL 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. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. 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. Liao,Zhang,Li,Xu [Page 14] ------------------------------------------------------------------------ Internet Draft Recovery in OBS Network April 2006