[ https://issues.jboss.org/browse/JBTM-1816?page=com.atlassian.jira.plugin.... ] Tom Jenkinson commented on JBTM-1816: ------------------------------------- The reason in the difference in the tests is because if JacORB runs first it has put its RC entry in the object store. You can replicate the issue by using the ObjectStore in the attached jacorb-rc-in-object-store.tgz. If you put it as ArjunaJTS/jts/target/test-classes/ObjectStore then even "./build.sh -f ArjunaJTS/jts/pom.xml test -Dtest=JavaIdlRecoveryUnitTest -Didlj-enabled" would fail. iirc the RC manager uses a known UID hence why running JacORB then JDKORB in the same ObjectStore will be a problem. If you notice, the specific test that fails isn't actually the one that is generated in the log you pasted, so success or fail you get those log messages. i.e. those log messages come from tests where failure is expected. That being said, I don't know if there are any tests where success is expected - we should make sure we have some of those as that log message implies that the JDK orb hasn't been configured correctly. i.e. do we have a test that correctly initializes a JDK ORB RC manager? The only tests I see in JavaIdlRecoveryUnitTest are testing failure scenarios. -- 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