Deal with concurrent Hot Rod topology view changes -------------------------------------------------- Key: ISPN-426 URL: https://jira.jboss.org/jira/browse/ISPN-426 Project: Infinispan Issue Type: Task Components: Cache Server Affects Versions: 4.1.0.BETA1 Reporter: Galder Zamarreno Assignee: Galder Zamarreno Fix For: 4.1.0.CR1 Deal with situations where multiple nodes modify the view at the same time and the atomic putIfAbsent or replace calls do not succeed. If this happens on startup and the node has issues adding itself, a backoff and retry would most probably make sense. If unable after backoff, stop startup process and report issue. If this happens when a node is stopping and the node is trying to remove itself from the view, this is not such a big problem because logic to deal with crashed or stalled members will clear it up. If this happens when the crashed member detector is trying to remove a crashed node, backoff/retry would make sense and if still unable to remove, note it and wait for next view change that would attempt to clear it again. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira