Based On The Application Of Distributed Nagios System Design And Implementation Of Monitoring Platform

Large enterprises often has a large-scale IT information system, it maybe have 1000 to 3000 servers. It’s a very challenging work to monitor and manage these servers in daily operation and maintenance. Different applications on these servers deploy in different regions、different network partitions, use different operating system and different service. It increase manage difficulty and can not be centralized monitoring. In recent years, the virtual machine has low cost and scalable features, it reduce the number of physical servers, but increase the number of virtual hosts and system complexity, and require more monitoring work. High Availability system use load balance to shield the failed node automatically, this design require monitoring system must detect failures automatically. Nagios is a Linux-based general-purpose computer service monitoring system. It has loosely coupled, cross-platform, standard interface, highly scalable, flexible access via HTTP protocol features.This paper describe a project of unified monitoring platform in a securities company’s IT data center, determine the system requirements, use distributed Nagios system、the NSCA external components and Obsessive Compulsive Service Processor (OCSP) logic to collect、transmit、process and display the monitoring information come from sub-region monitoring nodes. In the design and implementation we make Nagios work with our other applications, and do some custom extensions for special business needs. In this project we establish the distributed application monitoring platform, apply it to a securities company, and achieve good results. This paper provides a good reference for other large enterprises to implement distributed application monitoring.