Deployment Guide for Advanced Monitoring of a Blue Gene Environment
An IBM Redpaper publication
Note: This is publication is now archived. For reference only.
Abstract
The larger a data center grows, the more monitoring it requires. But limited staff, mixed hardware, multiple software configurations, and complex network topologies all combine to increase the complexity of monitoring. In addition, highly expensive monitoring solutions require cost justification, additional training, higher software budgets, and expensive maintenance contracts, and they can add their own idiosyncrasies to a complex environment. Adding more people to provide 24/7 operational coverage is not cost-effective, except in the largest computing environments.
How can we make the most effective use of limited technical staff and alert the proper resource to occasional but critical system errors or developing potential problems? How can we enhance troubleshooting? In addition, every data center is faced with updated architectural requirements and unanticipated configuration changes. Constant requests for new hardware and software requirements must be weighed against limited budgets and the need for uptime.
Nagios offers a flexible solution: a widely supported and extensible monitoring application. And it is free. Teaming Nagios up with the renowned IBM® Blue Gene® system creates an attractive combination.
Table of Contents
Chapter 1. The monitoring solution overview
Chapter 2. Nagios, Fruity, and NRPE installation
Chapter 3. Apache WebServer, MYSQL, Nagios, Fruity, and NRPE configuration
Chapter 4. Nagios implementation and its plug-ins
Appendix A. Blue Gene/L plug-ins
Appendix B. Blue Gene/P plug-ins
Appendix C. DB2 plug-ins
Appendix D. Other plug-ins
Appendix E. Plug-in resources
Appendix F. NRPE scripts
Appendix G. Additional material