[ https://issues.jboss.org/browse/JGRP-1674?page=com.atlassian.jira.plugin.... ] Dennis Reed commented on JGRP-1674: ----------------------------------- Proposed fix details: STOP_FLUSH is sent in two cases: 1. onResume called with a list of participants. STOP_FLUSH is sent to each individual participant, so no need to verify that the STOP_FLUSH is valid in this case. Did not change this use case. 2. onResume called with "null" participants STOP_FLUSH is multicast, which may be seen by members that were not in the START_FLUSH. Added current flushMembers to the STOP_FLUSH message. up(STOP_FLUSH) modified to block the STOP_FLUSH only if the newly added flushParticipants list is not null, and flushParticipants does not include this node, and this node did not sent the STOP_FLUSH. -- 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