MPLS Working Group H. Zhang Internet Draft J. He Intended status: Standards Track Huawei H. Li China Mobile October 26, 2009 Expires: April 2010 SD-Triggered Protection Switching in MPLS-TP draft-zhl-mpls-tp-sd-01.txt Status of this Memo This Internet-Draft is submitted to IETF in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html This Internet-Draft will expire on April 26, 2010. Copyright Notice Copyright (c) 2009 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 in effect on the date of publication of this document (http://trustee.ietf.org/license-info). Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Zhang and He Expires April 26, 2010 [Page 1] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 Abstract In MPLS-TP survivability framework, a fault condition includes both Signal Failure (SF) and Signal Degrade (SD) that can be used to trigger protection switching. This document describes Signal Degrade (SD) detection and the consequent protection switching mechanism. Table of Contents 1. Introduction.................................................2 2. Terminology..................................................3 3. SD-Triggered Protection Switching............................4 3.1. General Operation.......................................4 3.1.1. General protection architecture....................4 3.1.2. APS Protocol.......................................4 3.2. SD-Triggered Protection Based on other Traffic (i.e. non normal traffic)..............................................5 3.2.1. Detection of SD....................................5 3.2.2. Protection Switching and APS Protocol..............6 3.3. SD-Triggered Protection Based on Normal Traffic.........6 3.3.1. Broadcast Bridge with special APS request priority.6 3.3.2. Broadcast Bridge with extra protection switching coordination..............................................7 4. Analysis.....................................................7 5. Security Considerations......................................7 6. IANA Considerations..........................................7 7. Acknowledgments..............................................8 8. References...................................................9 8.1. Normative References....................................9 8.2. Informative References..................................9 Author's Addresses..............................................9 1. Introduction In packet transport network, protection switching can be triggered by fault conditions and external manual commands. The fault conditions include Signal Failure (SF) and Signal Degrade (SD). SF on a transport entity can be detected by fault management OAM functions; SD on a transport entity can be detected by performance monitoring OAM functions. The SD condition used for protection switching mostly depends on packet loss ratio. For some services that are sensitive to time and synchronization, the SD condition may depend on packet loss ratio, packet delay and delay variation. Zhang and He Expires April 26, 2010 [Page 2] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 In MPLS-TP OAM tools, the packet loss measurement (LM) is used to measure the amount of the lost service packets and compute the loss ratio of service packet. The packet Delay Measurement (DM) is used to measure the packet delay and delay variation by sending periodic DM packets during the diagnostic interval. The LM and DM mechanisms are out of scope of this document. The detection of SD (e.g. packet loss) must be based on service packets. But in some situation, there is no normal traffic on working/protection transport entity, which makes the detection of SD based on service packets impossible and causes switch flapping. This document describes the mechanism for SD detection and consequent protection switching in 1:1 and 1:n protection architecture. 2. Terminology The reader is assumed to be familiar with the terminology in MPLS-TP. The relationship between ITU-T and IETF terminologies on MPLS-TP can be found in [Rosetta stone]. MPLS-TP: MPLS Transport Profile SF: Signal Failure SF-W: SF for working entity SF-P: SF for protection entity SD: Signal Degrade SD-W: SD for working entity SD-P: SD for protection entity APS: Automatic Protection Switching OAM: Operations, Administration and Maintenance CC/CV: Continuity Check/Connectivity Verification LM: Loss Measurement DM: Delay Measurement Zhang and He Expires April 26, 2010 [Page 3] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 3. SD-Triggered Protection Switching 3.1. General Operation 3.1.1. General protection architecture In case of MPLS-TP 1+1 protection architecture, the normal traffic is permanently sent on both working and protection entities by the permanent bridge at the source. Therefore, the detection of SD or the clearing of SD on both working and protection entities can be based on the characteristics of the original traffic. In case of MPLS-TP revertive 1:1 and 1:n protection architecture: - If the protection switching is not active, normal traffic is transported together with OAM on the working entity while on the protection entity OAM is transported alone or together with the possible extra traffic. - If the protection switching is active, only OAM is transported on the working entity. Normal traffic is transported on the protection entity together with OAM. The extra traffic originally transported on protection entity is disrupted. In case of MPLS-TP non-revertive 1:1 and 1:n protection architecture: - If the protection switching is not active, normal traffic is transported together with OAM on the working entity while on the protection entity OAM is transported alone or together with the possible extra traffic. - If the protection switching is active, OAM is transported alone or possibly together with extra traffic on the working entity. Normal traffic is transported on the protection entity together with OAM traffic. That is, the normal traffic is switched from the working entity to the protection entity and the extra traffic is switched from the protection entity to the working entity. As mentioned above, in case of 1:1 and 1:n protection architecture, there may be no normal traffic on working/protection transport entity, which makes the detection of SD impossible and consequently switch flapping may happen. 3.1.2. APS Protocol In APS request types, similar to the definition of SF-W (SF for working entity) and SF-P (SF for protection entity), a definition of Zhang and He Expires April 26, 2010 [Page 4] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 SD-W (SD for working entity) and SD-P (SD for protection entity) is required to prevent flapping. 3.2. SD-Triggered Protection Based on other Traffic (i.e. non normal traffic) 3.2.1. Detection of SD In case of 1:1 and 1:n protection architecture, for the transport entity on which there is no normal traffic, the detection of SD can be based on other service packets, e.g. - Extra traffic The extra traffic can be used instead of normal traffic in packet loss measurement. That is, the performance monitoring OAM packets (LM) will count the number of extra service packets and the loss measurement is based on extra service, which will be used as the input of SD condition of the transport entity that lacks normal service. - OAM packets for Test The Test OAM packets can be used to simulate the characteristics of the normal traffic and replace the normal service. With the performance monitoring OAM packets (LM) the SD condition of the transport entity can be detected. - OAM packets for CC/CV The CC/CV OAM packets can be used instead of normal service in packet loss measurement. That is, the performance monitoring OAM packets (LM) will count CC/CV packets and the loss measurement is based on CC/CV packets, which will be used as the input of SD condition of the transport entity where there is no normal service. The above-mentioned SD-triggered protection switching mechanism based on other traffic is general and flexible without constraint on the bridge type by using the extra traffic, OAM packets (test or CC/CV). However, the characteristics of the extra traffic are usually not the same as that of the normal traffic so that the performance measurement is not exactly reflecting the condition of real service. OAM packets for CC/CV are sent in fixed transmission period (3.3ms, 10ms, 1s, etc.) which as well doesn't reflect the condition of real services. It is applicable in places without strict requirement for SD measurement. Zhang and He Expires April 26, 2010 [Page 5] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 The performance measurement by Test OAM packets is accurate. But the usage of test packets on protection entity defeats the objective of the 1:1 and 1:n architectures, which is to have the protection entity bandwidth available for best effort traffic during the time there is no fault or degradation of the working transport entity. The test packets consume now this bandwidth. For the case of the 1:1 protection, this makes the bandwidth usage of this 1:1 architecture similar to the bandwidth usage of the 1+1 architecture. 3.2.2. Protection Switching and APS Protocol For the SD-triggered protection based on other traffic, if a broadcast bridge is used, the other traffic used to detect SD is required on the protection transport entity only. The priority of SD-P and SD-W shall be fixed as SD-P > SD-W similar to SF-P > SF-W. Therefore, a protection switching based on SD detection shall not be initiated if there exists also an SD condition on the protection transport entity. The priority of SD-P and SD-W can be flexible as well based on the actual degraded degree of signal. Therefore, traffic is sent on the transport entity with better SD condition if both SD-W and SD-P exist. 3.3. SD-Triggered Protection Based on Normal Traffic In case of 1:1 and 1:n protection architecture, a broadcast bridge can be applied to assist SD detection in SD-triggered protection. 3.3.1. Broadcast Bridge with special APS request priority In the normal state, the normal traffic is sent only on the working transport entity and only the SD condition of the working transport entity can be evaluated. When SD is detected on the working transport entity, sink end sends SD-W indication to the source end and the selector at the sink end switches to the protection transport entity. The broadcast bridge at the source end will then send the normal traffic on both working and protection transport entities and the performance of both working and protection transport entities can be monitored. If SD is detected on protection transport entity as well, i.e. SD-W and SD-P exist simultaneously, normal traffic is selected at the sink still from the protection transport entity to avoid flapping between protection and working states. Zhang and He Expires April 26, 2010 [Page 6] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 In this case the priority of SD-W and SD-P in the APS protocol is fixed as SD-W > SD-P to avoid flapping of between protection and working states. 3.3.2. Broadcast Bridge with extra protection switching coordination The SD-W based protection switch action described in section 3.3.1 is performed under the assumption that a SD condition on a transport entity is a rare condition, and it is thus unlikely that SD on protection will co-exist with SD on working. When such assumption is not considered to be reasonable, the operation of the selector may be modified as described hereafter. When the sink end detects an SD condition, it does not switch to the protection entity immediately. Instead, the broadcast bridge at the source end will send the normal traffic on both working and protection transport entity after receiving SD-W indication sent by the sink end. Then sink end is able to detect the SD condition of working and protection transport entity. If SD is detected on the protection transport entity as well, the normal traffic is sent to both working and protection transport entities and is selected from the working transport entity; if no SD is detected on the protection transport entity the normal traffic is selected from the protection entity. SD-triggered protection switching based normal traffic is simple and efficient by applying a specific broadcast bridge but with a minor modification to the priority order of APS request (i.e. SD-W > SD-P) or to the operation order. 4. Analysis In section 3.2 and 3.3, the different mechanisms are described. They may be suitable in differenct application scenarios with different requirements for simplicity, accuracy, flexibility. [More analysis may be needed in a future version of this draft] 5. Security Considerations To be added in a future version of the document. 6. IANA Considerations No new IANA considerations. Zhang and He Expires April 26, 2010 [Page 7] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 7. Acknowledgments To be added in a future version of the document. Zhang and He Expires April 26, 2010 [Page 8] Internet-Draft draft-zhl-mpls-tp-sd-01.txt October 2009 8. References 8.1. Normative References [MPLS-TP Framework] Bocci,M., and Bryant,S., "A Framework for MPLS in Transport Networks", draft-ietf-mpls-tp-framework-06 (Work in progress), October 2009 [RFC5654] Niven-Jenkins,B., Brungard,D., and Betts,M., "Requirements of an MPLS Transport Profile", RFC5654, September 2009 [ITU-T Recommendation G.808.1] "Generic protection switching - Linear trail and subnetwork protection", ITU-T G.808.1 (03/2006) [MPLS-TP OAM Requirements] Vigoureux,M., Ward,D., and Betts,M., "Requirements for OAM in MPLS Transport Networks", draft- ietf-mpls-tp-oam-requirements-03(work in progress), August 2009 8.2. Informative References [MPLS-TP Survive Frmk] Sprecher,N., and Farrel,a., "Multiprotocol Label Switching Transport Profile Survivability Framework", draft-ietf-mpls-tp-survive-fwk-00(work in progress), April 2009 Author's Addresses Haiyan Zhang Huawei Technologies Co., Ltd. Phone: +86-755-28972333 Email: zhanghaiyan@huawei.com Jia He Huawei Technologies Co., Ltd. Phone: +86-755-28972333 Email: hejia@huawei.com Han Li China Mobile Communications Corporation Email: lihan@chinamobile.com Zhang and He Expires April 26, 2010 [Page 9]