Today’s business expectations are high for the IT solutions. A lot must be taken into account while implementing, testing or deploying a new service. If an IT company would like to deliver a service, which is recognised as stable by the customers, an acceptable level of availability must be provided. This level can be lower or higher. Both must be verified with authentic information by the company, in order to earn the end users' trust.
Monitoring systems were designed for the sake of checking the availability and of the insurance of quick reaction on the provider's side. The internet is more widely spread in the continuously developing world, the users can reach a service in more and more places, hence it is essential for these services to be reachable anytime and for the users to not have to wait for them. It can be achieved with the monitoring systems, because they were designed especially for this. Informations can easily be get with their help from the given system, or just the part of it, such as the utilization of memory, hard disk, processor etc. Nevertheless these systems can give informations not just about phisical devices, but about one or another service on the server. A monitoring system can be so parameterized to act as a user (e.g. log on). This way the administrators of the system can easily react to a problematic or peak case.
I have made literature research, during which I have came to know the Nagios monitoring and the Splunk loganalyzer system, as well as standards and technologies, that were necessary for the design of the infrastructure during the thesis.
I have made a monitoring system, which is capable of showing the condition of multiple systems and their services from the informations gathered from a log file. Besides I have developed a phone application, which is able to notify the administrator in case of emergency immediately.