Monitoring scientific computing infrastructure using Nagios

Abstract

This document will describe the Nagios configuration and use by the Scientific Computing Technology (SCT) group at the eScience department of STFC. The Nagios monitoring system has been extensively utilised for the purpose of increasing quality of service of the many services provided by the group. Clients of our services include software developers to scientists across Europe, so it is imperative that the high level of service that our clients expect is provided, and this is provided by a highly scalable, configurable and powerful monitoring system such as Nagios. The Nagios monitoring system is so widespread such that , in addition to standard host/service monitoring, grid monitoring tools such as NCG(Nagios Configuration Generator) are being developed and used and are phasing out existing monitoring such as Inca[3]. The fact that Nagios is an open source project makes it an even more attractive solution as users can modify the code to suit their needs and an example of this will be given. Due to the extensive use within academia and the open source community, knowledge of its use is widespread and support readily available on the World Wide Web. At SCT, Nagios version 3.0.6 runs on RHEL 4.8 with Apache as the web server. This configuration works well, and is reported to run on many flavours of Linux [2]. This report will outline the experience and usage of Nagios within SCT and does not give an exhaustive overview of the technology; if readers require this, they are refered to [1].