[esb-issues] [JBoss JIRA] Commented: (JBESB-1434) Caching service invokers in EsbNotifier/EsbActionHandler can require restart of server

[ http://jira.jboss.com/jira/browse/JBESB-1434?page=comments#action_12395817 ] Mark Little commented on JBESB-1434: ------------------------------------ I'm wondering if the issue is to do with undeploying the old service and leaving it long enough before deploying a new service that the SI instance fails-over to the redelivery service. If the original message isn't using a LogicalEPR, then any redelivery attempt will also fail. I agree with you that once the new service comes on line the SI instance should pick it up and move over to using it. Then of course there's the issue that if we have a complete failure in an SI instance and we try to reuse it there'll never be a 2nd attempt to refresh the cache subsequently, because serviceClusterInfo.getEPRs().size() will already be zero when we hit post. We'll refresh the cache once per delivery attempt from that point forward. I need to double check that reasoning though. We definitely need the clarification from Jiri as well. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira