Navigation:  »No topics above this level«

Cluster Server Support

Previous pageReturn to chapter overviewNext page

If you use a clustered server for running WebOffice 10 R3 (fail over respectively load balancer cluster), you can configure WebOffice 10 R3 to centralize the relevant configuration files.

 

Note: For using WebOffice 10 R3 in a cluster every hardware id of a cluster node has to have its own license file. The additional license files have to be named license.properties2, license.properties3, ...

Additional cluster licenses are charged wit 50% of the normal price.

 

Navigate to <WebOffice application installation directory>\WEB-INF\web_user.xml

Open the configuration file using a XML or text editor.

 

WebOffice 10 R3 needs a workspace directory as well as a configuration_path directory for working correctly. You can configure a centralized directory using file paths containing logical drive names (e.g. O:\WebOffice\workspace and O:\WebOffice\configuration_path) or UNC paths (e.g. \\SANNodeName\WebOffice\workspace and \\SANNodeName\WebOffice\configuration_path).

 

Note that changes on the web_user.xml file only take effect after reloading the WebOffice 10 R3 web application.

 

You may configure centralized workspace and configuarion_path properties:

 

  <!-- Path to the WebOffice 10 R3 configuration directory -->

  <context-param>

    <param-name>configuration_path</param-name>

    <param-value>/WEB-INF/classes/configuration</param-value>

  </context-param>

  <!-- Work folder. Changing this value is only necessary if e.g. working in cluster environments.

 In the Work folder some information gets stored/retrieved:

         -User profiles ('profiles' sub directory)

         -States ('states' sub directory)

         -Reports ('reports' sub directory)

         -Print outputs ('archive' sub directory). -->

  <context-param>

    <param-name>workspace</param-name>

    <param-value>/WEB-INF/work</param-value>

  </context-param>

 

Note that for reloading the application of project configurations it is necessary to perform the reload operation on each of the cluster nodes explicitly because if you only do the reload operation on the logical cluster node, the performed action will cause a state change on one of the cluster nodes only (there is no automatism for propagating the application state change).