System Specs for Respax Redundant server set
A typical Respax server configuration consists of a set of 2 identical servers sitting next to each other on the LAN. The servers are completely redundant to each other. When one fails, the other one takes over giving the tour operator full redundancy for operating system failures as well as hardware failures in any of the two servers.
The Respax servers are configured with 64-bit open-source Linux (Ubuntu) environments as the main underlying operating system and Postgres 8.2 as the main Respax database. The server set is also configured with Vmware running virtual server software which allows us to run several virtual servers on top of server hardware. Typical applications running on these virtual servers include Respax Taskmaster, Respax for Windows over Terminal services and Apache web servers.
The server set is using DRBD to fully raid the disk drives on each machine allowing for full replication of postgres, critical configuration files and Vmware slices. The server set is also configured with a Heartbeat system which is continuously monitoring the servers to ensure they are available at all times.
The server hardware required needs to support Virtualisation Technology which all newer hardware supports now. At time of writing this document, we recommend minimum hardware to have Quad-Core, single Intel CPU system or Dual-Quad Core Intel Xeon with 4gb ram. Dual 73GB SAS drives are recommended but SATA drives will work fine.
Ideal configuration is Dual Intel Quad-core XEON processors with 4gb of ram and dual raided sas drives. If you are running a Terminal Services environment for your users to access Respax as a virtual server on this server set, we recommend that you add a dedicated SAS drive for this virtual server.