使用beeline/jdbc 连接hive,impala,sparksql
hive和sparksql的端口号都是10000,所以在连接时我们要区分连接的是sparksql还是hive。连接hive:beeline -u jdbc:hive2://cdh1:10000/default连接sparksql:在cdh2上启动spark thirft-server服务:/opt/cloudera/parcels/CDH/lib/spark/sbin/st