【摘要】：近年來(lái)，計(jì)算機(jī)網(wǎng)絡(luò)的發(fā)展特點(diǎn)是規(guī)模不斷擴(kuò)大，復(fù)雜性不斷增加，異構(gòu)性越來(lái)越高，從而增加了網(wǎng)絡(luò)管理的難度。面對(duì)越來(lái)越復(fù)雜和重要的網(wǎng)絡(luò)，如何確保其盡可能長(zhǎng)時(shí)間的正常運(yùn)行，或當(dāng)網(wǎng)絡(luò)出現(xiàn)故障時(shí)，盡可能快地發(fā)現(xiàn)和修復(fù)故障，使其最大限度地發(fā)揮其應(yīng)用功能和效益，就成了網(wǎng)絡(luò)管理者普遍關(guān)注的問(wèn)題。目前市場(chǎng)上的網(wǎng)絡(luò)監(jiān)控產(chǎn)品大多監(jiān)控網(wǎng)內(nèi)主機(jī)，在設(shè)計(jì)時(shí)大部分產(chǎn)品都沒(méi)有考慮不同操作系統(tǒng)因素，如果網(wǎng)內(nèi)主機(jī)采用不同的操作系統(tǒng)，則可能失去了監(jiān)控功能。所以本項(xiàng)目的研究重點(diǎn)是能夠?qū)崿F(xiàn)跨平臺(tái)監(jiān)控，以及不同廠家網(wǎng)絡(luò)設(shè)備的集中監(jiān)控。該系統(tǒng)能夠在服務(wù)器和網(wǎng)絡(luò)設(shè)備出現(xiàn)故障的情況下及時(shí)將故障點(diǎn)和故障原因以顯著的方式提醒給用戶。 本文全面的分析了服務(wù)器及網(wǎng)絡(luò)設(shè)備監(jiān)控子系統(tǒng)的需求和技術(shù)可行性，決定采用C/S架構(gòu)進(jìn)行該系統(tǒng)的設(shè)計(jì)和開(kāi)發(fā)，為了實(shí)現(xiàn)服務(wù)器監(jiān)控模塊與網(wǎng)管系統(tǒng)可靠的傳輸，采用UDP可靠性機(jī)制進(jìn)行通信，為了實(shí)現(xiàn)網(wǎng)絡(luò)設(shè)備監(jiān)控模塊采集網(wǎng)絡(luò)設(shè)備的信息，通過(guò)SNMP協(xié)議完成數(shù)據(jù)的采集，采用的技術(shù)框架為Spring+Mybatis。整體系統(tǒng)的開(kāi)發(fā)使用Visual Studio2010、MyEclipse10.0、VisualBasic6，數(shù)據(jù)庫(kù)采用Oracle10g。 本文研究的主要內(nèi)容分為兩個(gè)大模塊：服務(wù)器監(jiān)控模塊和網(wǎng)絡(luò)設(shè)備監(jiān)控模塊。服務(wù)器監(jiān)控模塊主要是對(duì)多種不同操作系統(tǒng)的服務(wù)器信息進(jìn)行監(jiān)控，通過(guò)請(qǐng)求和接收配置管理子系統(tǒng)下發(fā)的配置信息，分析配置信息并寫(xiě)入策略文件中，根據(jù)策略文件生成采集執(zhí)行任務(wù)列表，執(zhí)行采集腳本從服務(wù)器上采集信息，并將采集到的信息生成XML文件上傳給網(wǎng)管系統(tǒng)。其中操作系統(tǒng)包括：Linux、AIX和Windows操作系統(tǒng)，，采集信息包括：內(nèi)存、進(jìn)程、操作系統(tǒng)、磁盤(pán)IO、磁盤(pán)分區(qū)、網(wǎng)卡流量和用戶登錄等信息。網(wǎng)絡(luò)設(shè)備監(jiān)控模塊主要是對(duì)支持SNMP協(xié)議的網(wǎng)絡(luò)設(shè)備進(jìn)行監(jiān)控，獲取服務(wù)器數(shù)據(jù)庫(kù)里的配置信息，采集網(wǎng)絡(luò)設(shè)備信息和接口信息，根據(jù)采集的數(shù)據(jù)進(jìn)行分析處理和告警處理，并將采集的數(shù)據(jù)進(jìn)行數(shù)據(jù)庫(kù)持久化操作。其中網(wǎng)絡(luò)設(shè)備包括：路由器、防火墻和交換機(jī)。 通過(guò)對(duì)系統(tǒng)進(jìn)行功能測(cè)試和非功能測(cè)試，確定系統(tǒng)已經(jīng)可以對(duì)服務(wù)器和網(wǎng)絡(luò)設(shè)備進(jìn)行監(jiān)控，滿足系統(tǒng)預(yù)期的功能要求和非功能要求，測(cè)試結(jié)果總體符合預(yù)期結(jié)果，達(dá)到系統(tǒng)的要求。
[Abstract]:In recent years, the development of computer network is characterized by increasing scale, increasing complexity and increasing heterogeneity, thus increasing the difficulty of network management. In the face of increasingly complex and important networks, how to ensure their normal operation for as long as possible, or how to detect and repair them as quickly as possible when they fail, so as to maximize their application functions and benefits, It has become a common concern of network managers. At present, most of the network monitoring products in the market do not take different operating system factors into account when designing most of the products. If the host in the network adopts different operating systems, it may lose the monitoring function. Therefore, the research focus of this project is to achieve cross-platform monitoring, as well as centralized monitoring of different manufacturers of network equipment. The system can remind the user of the fault point and fault reason in a significant way in the event of server and network equipment failure. This paper comprehensively analyzes the requirements and technical feasibility of server and network equipment monitoring subsystem, and decides to design and develop the system with C / S architecture, in order to realize the reliable transmission between server monitoring module and network management system. The reliability mechanism of UDP is used to communicate. In order to realize the monitoring module of network equipment to collect the information of network equipment, the data is collected through SNMP protocol. The technical framework is Spring Mybatis. The development of the whole system uses Visual Studio 2010 and Visual basic 6 with my Eclipse 10.0, and the database adopts Oracle 10g. The main content of this paper is divided into two modules: server monitoring module and network equipment monitoring module. The server monitoring module mainly monitors the server information of many different operating systems. By requesting and receiving the configuration information from the configuration management subsystem, the configuration information is analyzed and written into the policy file. The collection task list is generated according to the policy file, and the collection script is executed to collect the information from the server. The collected information is generated by the XML file and uploaded to the network management system. Among them, the operating system includes: Linux Aix and Windows operating system, collecting information includes: memory, process, operating system, disk IOs, disk partition, network card traffic and user login information. The network equipment monitoring module mainly monitors the network equipment which supports SNMP protocol, acquires the configuration information in the server database, collects the network equipment information and the interface information, carries on the analysis processing and the alarm processing according to the collected data. And the collected data for database persistence operation. Network devices include routers, firewalls and switches. Through the functional test and non-functional test of the system, it is determined that the system has been able to monitor the server and network equipment to meet the expected functional and non-functional requirements of the system, and the test results are generally in line with the expected results. Meet the requirements of the system.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TP393.06

【參考文獻(xiàn)】

中國(guó)期刊全文數(shù)據(jù)庫(kù) 前10條

1 李香云;葛華;;數(shù)據(jù)包組播技術(shù)在廣播教學(xué)中的應(yīng)用[J];湖南文理學(xué)院學(xué)報(bào)(自然科學(xué)版);2012年02期

2 劉燁;;用Socket實(shí)現(xiàn)基于TCP和UDP的原理探索[J];電腦學(xué)習(xí);2009年03期

3 王建新,楊世鳳,史永江,童官軍;遠(yuǎn)程監(jiān)控技術(shù)的發(fā)展現(xiàn)狀和趨勢(shì)[J];國(guó)外電子測(cè)量技術(shù);2005年04期

4 譚

本文編號(hào)：2181943