{% include JB/setup %}
Build tested on OSX, CentOS 6.
Checkout source code from https://github.com/apache/incubator-zeppelin
mvn install -DskipTests
mvn install -DskipTests -Dspark.version=1.1.0 -Dhadoop.version=2.2.0
Change spark.version and hadoop.version to your cluster's one.
Note that is you uses custom build spark, you need build Zeppelin with custome built spark artifact. To do that, deploy spark artifact to local maven repository using
sbt/sbt publish-local
and then build Zeppelin with your custom built Spark
mvn install -DskipTests -Dspark.version=1.1.0-Custom -Dhadoop.version=2.2.0
Configuration can be done by both environment variable(conf/zeppelin-env.sh) and java properties(conf/zeppelin-site.xml). If both defined, environment vaiable is used.
spark.jars, spark.files property in ZEPPELIN_JAVA_OPTS adds jars, files into SparkContext. for example,
ZEPPELIN_JAVA_OPTS="-Dspark.jars=/mylib1.jar,/mylib2.jar -Dspark.files=/myfile1.dat,/myfile2.dat"
or you can do it dynamically with dependency loader
bin/zeppelin-daemon.sh start
After successful start, visit http://localhost:8080 with your web browser. Note that port 8081 also need to be accessible for websocket connection.
bin/zeppelin-daemon.sh stop