I have just installed a Cluster based on CDH 5.2 with Map Reduce V1, YARN, HBASE, Hive, Impala, etc. I would like to make the server with HIVE, a Client so that I can submit map reduce jobs from there. How do I make this server the Client? I have heard of the Gateway Role but the manager does not selection of the Gateway. Is it because the Hive,Hue. Impala are already selected?
Hi, I selected Hive Service from the Home Page, then went to Instances. I selected the Server. There was an action button with pull down menu to deploy client configuration. I did that. It completed. I then ssh to the Hive Server and then do a simple hadoop fs -ls command. I am unable to see the hdfs file system. I just see the local file system. I would like to the use the Hive Server as a hadoop client (aka Gateway). When I can submit map reduce jobs. Am I mistaken here? Also, I added two servers to the cluster and want to use as a pure GW, no hive or impala. I can only see them as a hive server. What am I am doing wrong? Thanks, Anant