{% include JB/setup %}
HBase Shell is a JRuby IRB client for Apache HBase. This interpreter provides all capabilities of Apache HBase shell within Apache Zeppelin. The interpreter assumes that Apache HBase client software has been installed and it can connect to the Apache HBase cluster from the machine on where Apache Zeppelin is installed.
To get start with HBase, please see HBase Quickstart.
By default, Zeppelin is built against HBase 1.0.x releases. To work with HBase 1.1.x releases, use the following build command:
# HBase 1.1.4 mvn clean package -DskipTests -Phadoop-2.6 -Dhadoop.version=2.6.0 -P build-distr -Dhbase.hbase.version=1.1.4 -Dhbase.hadoop.version=2.6.0
To work with HBase 1.2.0+, use the following build command:
# HBase 1.2.0 mvn clean package -DskipTests -Phadoop-2.6 -Dhadoop.version=2.6.0 -P build-distr -Dhbase.hbase.version=1.2.0 -Dhbase.hadoop.version=2.6.0
If you want to connect to HBase running on a cluster, you'll need to follow the next step.
In conf/zeppelin-env.sh
, export HBASE_HOME
environment variable with your HBase installation path. This ensures hbase-site.xml
can be loaded.
For example
export HBASE_HOME=/usr/lib/hbase
or, when running with CDH
export HBASE_HOME="/opt/cloudera/parcels/CDH/lib/hbase"
You can optionally export HBASE_CONF_DIR
instead of HBASE_HOME
should you have custom HBase configurations.
In a notebook, to enable the HBase Shell interpreter, click the Gear icon and select HBase Shell.
In a paragraph, use %hbase
to select the HBase Shell interpreter and then input all commands. To get the list of available commands, use help
.
%hbase help
For example, to create a table
%hbase create 'test', 'cf'
And then to put data into that table
%hbase put 'test', 'row1', 'cf:a', 'value1'
For more information on all commands available, refer to HBase shell commands.