Apache Hive is a data warehouse infrastructure built on top of Hadoop that supports data summarization, query, and analysis. Hive provides an SQL-like language called HiveQL that transparently converts queries to MapReduce for execution on large datasets stored in Hadoop's HDFS. Learn more at hive.apache.org.
This charm provides the Hive command line interface and the HiveServer2 service.
This charm is uses the Hadoob base layer and the HDFS interface to pull its dependencies and act as a client to a Hadoop namenode:
You may manually deploy the recommended environment as follows:
juju deploy apache-hadoop-datanode datanode
juju deploy apache-hadoop-namenode namenode
juju deploy apache-hadoop-nodemanager nodemgr
juju deploy apache-hadoop-resourcemanager resourcemgr
juju deploy apache-hadoop-plugin plugin
juju add-relation namenode datanode
juju add-relation resourcemgr nodemgr
juju add-relation resourcemgr namenode
juju add-relation plugin resourcemgr
juju add-relation plugin namenode
juju deploy mysql
juju set mysql binlog-format=ROW
Deploy Hive charm:
juju deploy apache-hive hive
juju add-relation hive plugin
juju add-relation hive mysql
Please note the special configuration for the mysql charm above; MySQL must be using row-based logging for information to be replicated to Hadoop.
From the Hive unit, start the Hive console as the hive user:
juju ssh hive/0
sudo su hive -c hive
From the Hive console, verify sample commands execute successfully:
show databases;
create table test(col1 int, col2 string);
show tables;
quit;
Exit from the Hive unit:
exit
From the Hive unit, start the Beeline console as the hive user:
juju ssh hive/0
sudo su hive -c beeline
From the Beeline console, connect to HiveServer2 and verify sample commands execute successfully:
!connect jdbc:hive2://localhost:10000 hive password org.apache.hive.jdbc.HiveDriver
show databases;
create table test2(a int, b string);
show tables;
!quit
Exit from the Hive unit:
exit