+++ This bug was initially created as a clone of Bug #906060 +++

Description of problem: HA tests fail when run in a loop.

Version-Release number of selected component (if applicable): 0.18-mrg

How reproducible: usually in < 10 iterations

Steps to Reproduce: while make check TESTS=ha_tests.py; do true; done
  
Actual results: A test eventually fails

Expected results: no failures

Additional info:

The following tests were observed to fail occasionally:

ha_tests.LongTests.test_failover_send_receive
ha_tests.RecoveryTests.test_expected_backup_timeout
ha_tests.ReplicationTests.test_auto_delete_timeout
ha_tests.ReplicationTests.test_queue_replica_failover
ha_tests.ReplicationTests.test_alternate_exchange
ha_tests.ReplicationTests.test_stale_response
ha_tests.ReplicationTests.test_delete_qpid_4285
ha_tests.ReplicationTests.test_failover_python
ha_tests.RecoveryTests.test_expected_backup_timeout
ha_tests.RecoveryTests.test_send_receive
ha_tests.ReplicationTests.test_delete_qpid_4285

--- Additional comment from Alan Conway on 2013-01-30 13:31:38 EST ---

Fixed by the following commits on branch http://mrg1.lab.bos.redhat.com/cgit/qpid.git/log/?h=0.18-mrg-aconway-bz906060

3dedfd0 Bug 906060: HA Fix race condition in rejecting connections.
30cee33 Bug 906060: HA fix sporadic core dump in BrokerReplicator::disconnected
d831237 Bug 906060: HA fix race condition in QueueGuard
79d034f Bug 906060: HA Test bugs causing sporadic faiulres in ha_tests.ReplicationTests.test_auto_delete_timeout