[ https://issues.jboss.org/browse/JBTM-1685?page=com.atlassian.jira.plugin.... ] Michael Musgrove commented on JBTM-1685: ---------------------------------------- Seems to be timing related. This test starts 2 servers both of which are asked to crash during commit. On restart an external recovery manager is supposed to replay the commit phase. A second client starts and uses CrashRecoveryDelays.java to ensure enough time has elapsed (using method awaitReplayCompletion). [Note CrashRecoveryDelays has 2 approaches, either awaitReplayCompletion or explicitly tell the recovery manager to trigger recovery (method awaitRecovery)]. After the second client has waited for the replay it asks each server what it did. When the test fails the servers are still waiting for the replayCompletion request. I modified the second client to wait longer before asking the servers what they did and the test passes consistently. -- 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