Ok, I've just found the following statement while browsing: After the application finishes using the session, though, it must call the dispose() method in order to free the resources and used memory. Thus it seems that there is no auto garbage collection within a drools session. From this it follows I guess that when isnerting different facts for example inserting different process instances like in the example it makes no sense to use one session for several processess since references to these processes and other facts will never be removed. Thus in a multi threaded application where several processes are spawned there should be a separate session for each process and these should be tracked in a separate list if for example there are wait nodes which may be called from different threads. Is this correct? What can I use to identify a session in a knowledge base? As I see there is a getId() function. Can this be used to identify the session. Is it unique? -- View this message in context: http://drools-java-rules-engine.46999.n3.nabble.com/garbage-collected-in-... Sent from the Drools - User mailing list archive at Nabble.com.