Lock-on-active does not prevent rule evaluation. It prevents only rule activation.
The reason nodes are still evaluated is because Rete eagerly calculates all partial matches, because as soon as you move your execution to another group, the previously locked group will start activating rules again and so the partial matches are already calculated. I understand that this might not be desired in some cases, but that is how rete in general works.
Edson
2010/7/16 mmarmol
<marcelo.marmol@gmail.com>
He guys, I am using the meta-data lock-on-active on some rules and is working
great, but i have seen rules with lock-on-active = true are been eveluated
everytime the facts are modified even after they have been fired. Why is
that? I dont see the point of evaluating a rule with lock-on-active true
when it has already fired, it wont fire again. There is a way to resolve
this? or to implement something to avoid reevaluation? I am using drools
5.1.0M2 with a StatefulKnowledgeSession.
--
View this message in context: http://drools-java-rules-engine.46999.n3.nabble.com/lock-on-active-why-keeps-evaluating-tp973374p973374.html
Sent from the Drools - User mailing list archive at Nabble.com.
--
Edson Tirelli
JBoss Drools Core Development
JBoss by Red Hat @ www.jboss.com