Design And Implementation Of A High Availability Cluster Nanagement System Based On Linux Platform

As the computer network further use, people to the application service reliability,stability and sustainability have high requirements.At present, many industries havetheir own server clusters, cluster server as data storage, operation, forward’s keyequipment, once the server systems are out of order, huge losses may be caused. Tomaintain the sustainability of service, construct high availability cluster technologywill become a top priority. High availability cluster is a group of computers linkedtogether through a network，any single system or computer failure will not cause thesystem to provide the service is not available, and that as a whole, to provide userswith maximum not uninterrupted service.This thesis aims at the high availability of cluster management system, treats theopen source cluster provision and management software KUSU as the basis, and seesthe dual hot backup theory of the cluster system’s management node as the support.Also, the thesis regards the cluster system fault detection and automation of businessswitch as the core. The thesis researches and realizes the high availability clustermanagement system based on Linux platform.The system, based on KUSU, has been redeveloped with adding mechanisms ofheartbeat detection, service switching, data synchronization, floating IP address and theswitching between Installer Node and Failover node, to achieve the high availabilitycluster management system with fault detecting and handling。This system employs the open-source database of PostgreSQL, which exploitsdual-machine backup mechanism. Therefore, it simplifies the cluster management andreleasing the cluster system from the problems of data synchronization between themaster and slave node. It uses Heartbeat to detect the node states and determine theactive node between the primary node and backup node in the cluster. If the primarynode fails, in order to achieve uninterrupted services provide by the cluster, the backupnode will take over the task of cluster management, and start the primary service byusing Pacemaker. According to the needs of the practical application scenarios, thesystem provides the two switching way which is automatic switching and manualswitching. Thereby ensuring that the cluster system provides uninterrupted service.The testing results show that the system can switch between the nodes and takes over services efficiently, and achieves high-availability of the cluster.