[ https://issues.jboss.org/browse/DROOLS-164?page=com.atlassian.jira.plugin... ] Davide Sottara commented on DROOLS-164: --------------------------------------- Citing the Drools Expert guide: "Whenever a ruleflow-group becomes active or an agenda-group receives the focus, any rule within that group that has lock-on-active set to true will not be activated any more; irrespective of the origin of the update, the activation of a matching rule is discarded". When you insert the fact, both rules become active. Rule 1 then fires first because of the higher salience. When it modifies the Tradebooking AFTER setting the action (notice that it is a key attribute!), two things happen: a new activation (with an effectively new TradeHeader) is tentatively generated, but rejected because of the lock-on-active; the previous activation is cancelled because its supporting facts no longer hold. Finally, the String blocks a new activation of Rule 1. - Removing salience 1 lets Rule2 fire before Rule1, so the latter does not have the time to cancel the former - Removing the setAction() causes Rule2 to re-activate "in place". Since the new facts are equals, the engine does not try to replace the activation with a new one. So lock-on-active does not apply (correctly) - Removing the modify of course prevents any re-evaluation and tentative reactivation of Rule2 - Removing lock-on-active prevents the blocking effect. So, the engine behaves correctly. You should reconsider modifying an attribute when it's a key, or find alternatives to lock-on-active such as property reactivity. Please let me know if you need more explanations. The ticket should be rejected. Best Davide -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira