As we transition to the new Hyland Community, some post links may be temporarily disrupted as we finalize redirects. We apologize for the inconvenience and are working to resolve this as quickly as possible. Details.
activiti.cfg.xml
JavaDelegate
Why is this a problem? Since the service tasks are configured using an asynchronous continuation, it is possible that the corresponding jobs are all acquired at the same time and delegated to different worker threads by the JobExecutor. The consequence is that the transactions in which the services are executed and in which the 3 individual executions arrive at the parallel join can overlap. And if they do so, each individual transaction will not "see", that another transaction is arriving at the same parallel join concurrently and thus assume that it has to wait for the others. However, if each transaction assumes that it has to wait for the other ones, none will continue the process after the parallel join and the process instance will remain in that state forever.