A fail-over manager application for HA-cluster

OData support
Dr. Horváth Tamás Imre
Department of Control Engineering and Information Technology

My dissertation is about developing and implementing a manager application for high-availability cluster systems, which are using standby nodes. The design phase focused on the object-oriented software development. For the development I used appropriate design patterns: factory method, observer, decorator and singleton design pattern.

The first chapter briefly introduces the features of high availability systems, which are using standby nodes, and briefly characterized the DRBD (Distributed Replicate Block Device) and Linux-HA software (Heartbeat).

The second section shows the application design and implementation, using UML class and sequential diagrams. The first part describes the implemented utility classes, the others introduces the realized main components (configuration management, system status information and state-image components), and finally the implemented application itself is presented.

The third section lists the most important test cases which were performed to check the application.

The last section summarizes the experiences I got during the development and implementation, and makes a proposal how to improve the application.


Please sign in to download the files of this thesis.