Hi, I have written client application in java, which writes Apache log data to HDFS Cluster. When the HDFS cluster is brought down, the client attempts to connect to the Cluster. I have a few issues regarding the same. 1) When the HDFS cluster is brought down, the Application does not get any exception, instead I see ipc reconnecting messages displayed from DFS client . Is there a way to get an exception/notice, when the cluster is down? 2) The HDFS DFS client, which is not part of the Application code, tries to connect to HDFS cluster infinitely. Is there a way to configure the number of times the client tries to connect? Do we have an application which writes data to HDFS cluster and is fault-torant? (Handle HDFS cluster shutdown/restart) Thanks Suresh