[ https://issues.jboss.org/browse/ISPN-6795?page=com.atlassian.jira.plugin.... ] Dan Berindei commented on ISPN-6795: ------------------------------------ The originator should *not* wait until the timeout expires, it should throw an {{RpcException}} the moment it receives a response from all the targets that are still in the cluster (assuming none of the responses are valid). The fact that we ignored the suspected flag for some staggered get responses and we waited for the timeout to expire is a bug, but it's orthogonal. I agree that an exception from the remote node is very likely a bug (as we're not throwing {{OutdatedTopologyException}} for get commands at this point), so I'd make the change. It's just that it's not "fixing" {{ClusteredGetResponseValidityFilter}}, it's changing its behaviour. -- This message was sent by Atlassian JIRA (v6.4.11#64026)