To install DRACOON, an understanding of how the infrastructure works is needed in addition to the various requirements that must be met.
DRACOON consists of multiple components that are described in a short summary here and in more detail in the subsequent chapters.
Depending on the size and required redundancy of the installation in question, all the components can run on one single server in the most simple installation; complex environments can consist of practically any number of servers. It is usually helpful to combine the installation with technologies that are already available to the customer (for example, a connection via an existing load balancer, clustered storage, etc.).
- The cornerstone of DRACOON is the API server. The API server is a Java application that requires a Tomcat application server to operate. This API provides all the external DRACOON functions. All the clients communicate directly with this API.
- The API server requires a connection to a MariaDB database in which the configuration and meta data is stored.
- The files uploaded to DRACOON themselves are stored on a corresponding file system that has no special requirements (for example, a local hard disk, iSCSI target, NFS share, FC LUN). As a basic principle, the data can be stored at any storage location that can be mounted in the back-end server.
- For load balancing, content switching and SSL offloading, reverse proxy functionality is required (e.g. HAProxy, Apache, Bluecoat, Kemp, F5).
- The Web client (DRACOON Web App) is also a standalone component and delivered as a Java application that requires a Tomcat application server to operate.
- As an option, a WebDAV proxy can be installed through which users can mount dataspaces as network drives in their end devices. Java 8 or OpenJDK 1.8 is required to do so.