Problem with persistence.
I have an iApp that have many accesses and not all the capacity to serve all users, so the solution we have is the following in the pool we have 6 nodes, 5 with the page that gives the service with a high Priority Group and equal limits of connection and 1 with a waiting page that has a low Priority Group and that has no limit of connections and the Priority Group Activation is Less than 1 this works well in accordance with the logic that when all the nodes reach their capacity it passes the the remaining connections for the waiting page of the node that has the Priority Group lower. The problem is that when the number of connections goes down in the nodes that have the Priority Group high some users remain in the waiting node and we suspect that it is due to the persistence of the connections that is 180 s and that the users before the expiration of that time update the page and the connection is reinitiated.