When many pieces of an infrastructure depend on a single point of failure, figuring out how to build redundancy into the system can be very difficult. A series of open source tools can come to your rescue and help you to develop a fault resistant and fault tolerant system. These tools can help to make sure that even if something can only run on one server, it has a backup waiting for something to happen so that it can kick in. When you also bring Linux into the mix you can create high-availability load balanced systems that can take a beating. When a system is working most people don't realize what's behind the scenes making it all work. When something goes wrong and it becomes noticeable many customers don't realize how infrequently this service is unavailable. With monitoring tools and statistical tools working together it becomes easy to monitor the status of servers and keep trends going to have a graphical representation of how servers and services are doing. Combining these tools over time can even work to help notify someone before a service becomes critical, allowing time to fix a problem before it's even apparent.