论文部分内容阅读
1.引言目前的商用网管平台,如IBM的TivoliNetView/6000、HP的OpenView,其故障管理上的实现方式均是通过配置窗口将管理员的经验转变成一系列规则集,然后通过内部解析设定规则集中包含的若干MIB(管理信息库)变量的阈值,一旦监测进程或驻留在被管对象里的代理进程发现超过阈值,就以陷阱方式向网管站报警,内置的故障管理模块对陷阱信息经过过滤和关联后定位和鉴别故障原因,并启动预定制的恢复程序。尽管传统的故障管理系统能够在一定程度上发挥作用,但仍然存在以下弊端:
1. INTRODUCTION Current commercial network management platforms such as IBM Tivoli NetView / 6000, HP OpenView, and fault management are implemented through a configuration window that transforms administrators’ experiences into a series of rule sets and then sets rules through internal parsing The set of thresholds for several MIB (Management Information Base) variables contained in a trap is used to trap the trap to the NMS once the monitoring process or the agent process residing in the managed object detects that the threshold is exceeded. The built-in fault management module traps the trap information Filter and correlate to locate and identify the cause of the failure and initiate a pre-built recovery procedure. Although the traditional fault management system can play a role to some extent, it still has the following disadvantages: