[ http://jira.jboss.com/jira/browse/JBESB-1458?page=comments#action_12394258 ] Kevin Conner commented on JBESB-1458: ------------------------------------- Jirka, can you please attach a standalone example of this failure. Thanks. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBESB-1458?page=all ] Kevin Conner reassigned JBESB-1458: ----------------------------------- Assignee: Kevin Conner (was: Kurt Stam) This is not a jBPM specific issue, it is a generic issue associated with one way services. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBESB-1458?page=comments#action_12395624 ] Kevin Conner commented on JBESB-1458: ------------------------------------- The analysis above is only correct up to stage 6, the real issue is a deadly embrace between two faulting services The sequence of events is as follows 1) we have sent message form jBPM to ESB 2) The message is processed 3) Reponse is sent to jBPM callback service 4) Callback command is executed and exception is thrown in jBPM code 5) This exception is handled by pipeline processor 6) The pipline process will send message fault message *** CHANGES HERE *** 7) The fault is received by the original ESB service targetted in 1) 8) Original service is not expecting the fault message but attempts to process it 9) This processing leads to a NPE 10) This pipeline creates a fault message and sends it back to the jBPM callback service 11) Go back to 4) As the callback service is receiving a response from the ESB service it should not send a fault messages in response. The original ESB service has not specified a FaultTo address nor is it expecting to handle one as it has already processed the original jBPM request. This problem is shared with all one way services in that they should not be sending any fault message if no FaultTo target has been defined. This also reflects one of the major differences between WS-A 2004 and WS-A 2005. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBESB-1458?page=all ] Kevin Conner reopened JBESB-1458: --------------------------------- Correcting the component -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://jira.jboss.com/jira/browse/JBESB-1458?page=all ] Kevin Conner closed JBESB-1458. ------------------------------- Resolution: Done -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira