e-docs > WebLogic Server > Using WebLogic Server Clusters > Configuring BIG-IPTM Hardware with Clusters |
Using WebLogic Server Clusters |
Configuring BIG-IPTM Hardware with Clusters
This topic contains the following sections:
This section describes how to configure an F5 BIG-IP controller to operate with a ProductName cluster. It assumes that you have some familiarity with BIG-IP configuration tasks.
This section provides some step-by-step instructions for configuring BIG-IP. However, if you require detailed setup and administration instructions, refer to your F5 product documentation.
For information about how WebLogic Server works with external load balancers, see Load Balancing HTTP Sessions with an External Load Balancer.
Using URL Rewriting With BIG-IP and WebLogic Server
Use of URL rewriting with BIG-IP and WebLogic Server instances requires BIG-IP version 4.5 or higher, configured for Rewrite cookie persistence. Failover may not succeed if BIG-IP is set for other persistence settings.
For instructions to configure WebLogic Server for URL rewriting, see "Using URL Rewriting" in Assembling and Configuring Web Applications.
Using Session Persistence with BIG-IP and WebLogic Server
If your cluster uses in-memory replication for client session states, you must configure BIG-IP to use the Insert mode for cookies. Insert mode insures that the original WebLogic Server cookie is not overwritten, and can be used in the event that a client fails to connect to its primary WebLogic Server.
To configure Insert mode for BIG-IP cookies:
To load balance requests on a round-robin basis, set the timeout value to zero—this ensures that multiple requests from the same client are directed to the same managed server, and that a request from a different client is routed to another managed server in the cluster, in round-robin fashion.
When the timeout value is set to a value greater than zero, the load balancer sends all requests from all clients to the same managed server in the WebLogic Server cluster for the duration of the timeout period—in other words, requests from different clients will not be load balanced for the duration of the timeout.