Automation of Site Configuration Management

Although there are countless tools to track and manage the configuration of large numbers of Unix systems, there seems to be a lack of tools to manage the interaction and dependencies between systems. As our site has grown, many machines provide services that are required for the operation of other machines and applications. We have been unable to maintain accurate lists of services and servers, and even routine system upgrades have resulted in unexpected service outages. To address this problem, we are developing a system to automatically detect many of these service dependencies, and generate up to date server listings. In addition, it provides a general framework for indexing and accessing troubleshooting, operational, installation and a number of other types of documentation. The system also assists in verifying the configuration of systems being installed, and assists with the real time testing of services.