The Design And Implementation Of Server Running Condition Monitroing System

At present with the deepening of the informatization construction, The IT systems ofall enterprises become increasingly complex , various of network equipment, server,middleware, business system let network administrator hard to manage . It will lead to thebusiness interruption because of equipment failure even working overtime to deploymentand management , seriously affecting the normal operation of enterprises and institutions.From the point of view of the business , the server running condition monitoringsystem make the traditional technology equipment management integration to the businessmanagement platform, Not only can the completion of the monitoring equipment, but alsocan position root causes and performance bottlenecks, forecasting business developmenttrend and stability according to the composition of the business. At the same time, thesystem can monitor infrastructure and application, intricate IT facilities into simplebusiness view, make the enterprise provide a stable and reliable services for IT departmentfrom the point of view of the IT services. A business service is made up by network, server,and application and so on. Once a component appears problem, problems will reaction inbusiness services. The IT department can understand business service operation at any time.This paper focuses on the research content is put forward the multi-thread datainformation collection mechanism, based on the monitoring probes, solve how to improvethe efficiency and the accuracy of gathering monitoring system's information; Then putforward the automatic connection method based on the Heartbeat bag to ensure the reliableconnectivity between the server and client. Finally, design and develop a server runningcondition monitoring system. Specific include the following aspects:First,Based on the multi-thread data information collection mechanism, we designedfive probes in the way of plugin, such as cpuProbe,mailProbe,packetProbe,processProbe,to collect various kinds of data such as CPU usage, sending mail, network packets, processinformation. We use multithreading mode make five probe concurrent job to improve theefficiency and accuracy of gathering information.Second,The automatic connection method based on the heart bag. According to theprinciple of heart bag send simple communication timing bag, if in the designated periodhave not received the other party response, then judged each other have offline. Thismethod used to detection TCP abnormal disconnection, then automatically connectedcombine with the daemon technology. Third,Use these research results, design and development a server running conditionmonitoring system. The system set "data collection, state analysis, graphics show and textmessage alarm" four functions in one body. Through this platform, network managementcan Control all kinds of network equipment's running state, such as discoverying the poorconfiguration, or can improve and optimizing the network equipment performance. Whenthe equipment appears unusual, can send the standard form alarm messages to theadministrator through text messages in time, and provide the reliable operation for thenetwork equipment.