I would like to know how the a4j:keepAlive tag works and whether it would be able to cope in the following situation.
We are using session replication between two tomcat servers in a cluster. Our understanding is that beans with session scope will be serialized between the two tomcats, the question is what will happen to request beans that use the a4j:keepAlive tag? How does the tag keep beans alive? The documentation indicates that it saves the beans after the request â€“ where are they saved and will they be replicated between the tomcat instances?
KeepAlive beans are saved in components view state, so it can be stored in server session or on the client depending on value of javax.faces.STATE_SAVING_METHOD context parameter. This feature should work for the case of session-replication environment.