Overview of Configuring the xvp™ Suite
Please click on the links to the left for instructions on how to configure the different components of the xvp suite. Please pay particular attention to the Security Issues. If you have problems getting anything to work, please read the Troubleshooting page before posting messages to the mailing list.
Note: If you are not familiar with configuring software on Linux, you will probably have a better time if you install the complete xvpappliance virtual machine, rather than the separate components.
To help explain how the pieces of xvp fit together, the overall architecture is presented in the diagram to the right. Click on it to enlarge (opens in new window).
Yellow boxes identify components of the xvp suite and their configuration data. The two dashed vertical lines separate the client tier (left), components which must (as far as the clients are concerned) run on the same server (middle), and backend components (right).
You can, if you wish, run the middle tier in a XenServer or Xen Cloud Platform virtual machine, but do not try to run directly on a Xen host server.
By default, and as shown in the above diagram, console connections from xvpweb are made directly to the VNC port(s) on xvp itself. As an alternative, it is possible to tunnel the console connections over the existing HTTP or HTTPS connection, so that, for example, in an Internet-facing environment it is not then necessary to open VNC ports to external access.
It is not trivial to set up tunnelling, and it poses its own security risks: for a detailed discussion, see the separate Tunnelling Guide.
Another separate document describes the XVP Extensions to the RFB Protocol used by xvp and xvpviewer.