I find it strange to have N rows in Quartz tables for N cluster nodes. You should have one row per route x action on route. I imagine that if you schedule the same action for the same route on 2 different cluster nodes, sometimes one cluster node will try to execute the Job scheduler by the other node: as Camel is storing ScheduledJobState in memory, then it's null it this case. I wonder why Camel is not storing this ScheduledJobState among JobParams (it's only a route + action), they would be persisted and shared by all nodes.
-- View this message in context: http://camel.465427.n5.nabble.com/Camel-JPA-Clustered-Deployment-Quartz-Scheduling-tp5733292p5736660.html Sent from the Camel - Users mailing list archive at Nabble.com.
