So you don't want to actually use clustering at all since you sent eacht request to the same node every time...
To achieve a request specific node for subsequent requests within the same application, I think you need to write your own LoadBalance policies....
Where do you need this behaviour for actually?
Thanks for your reply.
I don't want to send each request to the same node every time.
My requirement is like this:
Suppose first request from UserA is handled by NodeA, i want NodeA to handle all subsequent request from UserA. Similarly if NodeB handles first request from UserB, NodeB should handle all subsequent request from UserB.
This is the constraint in our application. We are developing a application to create a meeting request in Microsoft Outlook directly using MAPI calls. For this all request from one user should be handled by one Node because of some profile creation problem.
To achieve this, i need to write my own LoadBalancePolicy class.
I want to know how to do this. Is it enough if i replace LoadBalancePolicy attribute in cluster-service.xml with my custom LoadBalancePolicy class or something else i need to do?
Thanks in advance.
Are you referring to the LoadBalancePolicy attribute in the HAJNDI service? That's the only load balancing policy in the standard cluster-service.xml file.
If so, that policy is only used by HA-JNDI so overriding it will only affect your use of HA-JNDI. By default, HA-JNDI lookups are directed to servers based on this poilcy (i.e., they don't necessarily go to the server listed in the provider url).
If this isn't what you have in mind, you need to be more specific (e.g., are you describing session load balancing).