Planning the Installation and Configuration

Use the following section in conjunction with the worksheets in the Sun Cluster 3.0 Release Notes as a checklist before installing and configuring Sun Cluster HA for iPlanet Web Server.

Consider the following prior to starting your installation:

• Will you be running Sun Cluster HA for iPlanet Web Server as a failover or as a scalable service? For information on the two types of services, see the Sun Cluster 3.0 Concepts document. For scalable services, consider the following:

  • What nodes will host the scalable service? In most cases, you will want to scale across all nodes; however, you can limit the set of nodes that host the service.

  • Will your iPlanet Web Server instances require sticky IP? This is a resource property setting, Load_balancing_policy, which stores the client state in memory so return traffic from the same node always goes to the same cluster node. You can choose from several load balancing policies, as described in the table on resource properties in Appendix A, Standard Properties.

    Exercise caution when changing Load_balancing_weights for an online scalable service that has Load_balancing_policy set to LB_STICKY or LB_STICKY_WILD. Changing those properties while the service is online can cause existing client affinities to be reset, hence a different node might service a subsequent client request even if the client had been previously serviced by another node in the cluster.

    Similarly, when a new instance of the service is started on a cluster, existing client affinities might be reset.

• Where will the Web server root reside?

• Does the Web server serve data for another highly available application? If so, resource dependencies might exist between the resources so that one starts or stops before the other. For a description of the resource property Resource_dependencies that sets these dependencies, see Appendix A, Standard Properties.

• Determine the resource groups to use for network addresses and application resources and the dependencies between them. For a description of the resource group property RG_dependencies that sets these dependencies, see Appendix A, Standard Properties.

• Provide the logical host name (for failover services) or shared address (for scalable services) to be used by clients to access the data service.

• Since you can configure iPlanet Web Server to bind to INADDR_ANY, if you plan on running multiple instances of the iPlanet Web Server data service or multiple data services on the same node, each instance must bind to a unique network address and port number.

• Determine the entries for the Confdir_list and Port_list properties. For failover services, both of these properties can have only one entry. For scalable services, they can have multiple entries; however, the number of entries must be the same and must map to each other in the order specified. For details, see "How to Register and Configure Sun Cluster HA for iPlanet Web Server".

• Determine where to place logs, error files, and the PID file on the local file system.

• Determine where to place the contents on the cluster file system.