Hello, I am working on proof-of-concept with the Process API. A small cutout of code shows an example process: startnode->actionnode->rulesetnode-splitnode(gateway). The process flow starts and runs normally. However, the splitnode constraint does not seem to use the working memory of the rulesetnode and never finds a rule hit. Does the splitnode constraint rule generated by drools, use the same working memory as the rulesetnodes? If a rulesetnode has one ruleflow-group, is there a way to put the splitnode constraint rule into the same ruleflow-group? // Nodes .startNode(1) .name("Start") .done() .actionNode(2) .name("Action") .action("java", "System.out.println(\"Starting process\");") .done() .ruleSetNode(3) .name("City is in list") .ruleFlowGroup("CityRules") .done() // Gateway .splitNode(11) .name("Split 11") .type(Split.TYPE_XOR) .constraint(4, "yes", "rule", "mvel", "City(name == \"Copenhagen\")") .constraint(92, "no", "rule", "mvel", "not City(name == \"Copenhagen\")") .done() -- View this message in context: http://drools.46999.n3.nabble.com/Process-RulesetNode-and-SplitNode-worki... Sent from the Drools: User forum mailing list archive at Nabble.com.