[ https://issues.jboss.org/browse/JBTM-2017?page=com.atlassian.jira.plugin.... ] Michael Musgrove commented on JBTM-2017: ---------------------------------------- The test is in two stages (first generate a crash and then attempt recovery). The problem is happening because a socket (for the orb) opened in part 1 of the test is still closing but it is needed by the second part of the test. The socket is in the TIMED_WAIT state but will eventually close. I added a fix to perform 10 retries (with back-off) to give the socket close protocol time to complete. First some context: our test suite launches Tasks which have timeouts associated with them (to avoid hangs impacting later tests). This latest failure is because the task was timed out before my "retry fix" had time to run to completion. The logs show the task being reaped after 60 seconds but the code in org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery12#CrashRecovery12_Test03 specifies 240 seconds before it should be eligible for being reaped so but there isn't sufficient information in the logs to indicate why it was reaped early. I ran just the failing test all evening without a failures. I plan to resolve the issue by adding more information to the task reaper so that we have more to go on if it happens again. -- 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