If you see an error message [1] in the log, it means that backup server is not able to handle replication packets from live in defined timeout and the replication is stopped. You need to restart backup node to re-trigger the replication process. The timeout is defined on cluster-connection, attribute call-timeout. The default value is 30 seconds what is enough in most cases. However if your disk operations are slow or you configure journal-min-files to a high value, initial pre-creation of journal files can take considerable amount of time and backup is not able to respond to next replication packets until it finishes the pre-creation of files. In this case you should consider to increase call-timeout or decrease journal-min-files.

For more information see discussion in JBEAP-4742.

[1]

AMQ222207: The backup server is not responding promptly introducing latency beyond the limit. Replication server being disconnected now.