[ https://issues.jboss.org/browse/JBTM-1833?page=com.atlassian.jira.plugin.... ] Michael Musgrove commented on JBTM-1833: ---------------------------------------- This test uses an external recovery manager. Two servers are started and a client makes a transactional request to both of them. The second server crashes in prepare whilst the first server crashes in rollback. The two servers are then restarted and a second client triggers a recovery attempt. Both servers should have had rollback called on them. This scenario is very similar to spurious failures of CrashRecovery02_2 test 28 and CrashRecovery05_2 tests 28, 76 and 88 (JBTM-1685, JBTM-1510, JBTM-1842 and JBTM-1844) fixed with commit 8849ffcab38813f7e to master on Aug 2nd. The fix was to wait a bit longer before asking the servers whether recovery had replayed the pending transaction. I will apply a similar fix and also back port the fixes for JBTM-1685, JBTM-1510, JBTM-1842 and JBTM-1844. -- 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