ApacheZeppelin0.8.md

Building & Installing Apache Zeppelin 0.8 (from GIT)

Install R

On MacOSX, follow these steps: http://www.reed.edu/data-at-reed/software/R/r_studio.html

• Install R
• Install RStudio
• Download the startup_packages.R script from that page:

# List of useful packages
pkg <- c("tidyr", "dplyr", "ggplot2", "knitr", "rmarkdown")

# Check if packages are not installed and assign the
# names of the uninstalled packages to the variable new.pkg
new.pkg <- pkg[!(pkg %in% installed.packages())]

# If there are any packages in the list that aren't installed,
# install them
if (length(new.pkg)) {
  install.packages(new.pkg, repos = "http://cran.rstudio.com")
}

• Open this file in RStudio and click on "Source" to execute.
• This will prompt to OK installing various packages

Build Zeppelin

• git clone [email protected]:apache/zeppelin.git
• cd zeppelin
• mvn clean package -Pscala-2.11 -Pspark-2.2 -Phadoop-2.7 -Pr -DskipTests -Pbuild-distr

The installation package will be in ./zeppelin-distribution/target/zeppelin-0.8.0-SNAPSHOT.tar.gz

Install Zeppelin

• Extract the above tar.gz in a directory, say, $Z

Download the latest Spark 2.2+

• http://spark.apache.org/downloads.html
• Get the package "Pre-built for Apache Hadoop 2.7 and later"
• Extract the above tgz in a directory, say, $S

Configure Zeppelin

• (cd $Z/; ln -s $S spark)

• (cd $Z/; cp conf/zeppelin-env.sh.template conf/zeppelin-env.sh)

• (cd $Z/; cp conf/zeppelin-site.xml.template conf/zeppelin-site.xml)

• In $Z/conf/zeppelin-env.sh, uncomment & set JAVA_HOME and SPARK_HOME as follows:

  • For JAVA_HOME, set it to the location of JDK 1.8; e.g., export JAVA_HOME=/Library/Java/JavaVirtualMachines/jdk1.8.0_151.jdk/Contents/Home
  • export SPARK_HOME=$ZEPPELIN_HOME/spark (this assumes you've set the symlink as described above)

Start/Stop Zeppelin

• $Z/bin/zeppelin-daemon.sh start
• $Z/bin/zeppelin-daemon.sh stop