Becasue the hadoop cluster time date is not sync, you can use ntpdate asia.pool.ntp.org
to sync date time of each hosts.
Please confirm your IP and hostname in /etc/hosts
is correct or not.
$ stop-hbase.sh
$ zkServer.sh stop // Stop zookeeper
$ rm -rf $ZOOKEEPER_HOME/data/<version-?> // Remove the version of zk in data folder
$ zkServer.sh start // Start zookeeper
$ start-hbase.sh
Please check the client hadoop version is equal to server.
http://stackoverflow.com/questions/28666183/submitting-jobs-to-spark-ec2-cluster-remotely
http://apache-spark-user-list.1001560.n3.nabble.com/Invalid-signature-file-digest-for-Manifest-main-attributes-with-spark-job-built-using-maven-td14299.html http://stackoverflow.com/questions/28459333/how-to-build-an-uber-jar-fat-jar-using-sbt-within-intellij-idea
continue...