47 lines
1.6 KiB
Docker
47 lines
1.6 KiB
Docker
ARG HADOOP_VERSION=2.8.4
|
|
ARG HIVE_VERSION=2.3.3
|
|
FROM varadarb/hudi-hadoop_${HADOOP_VERSION}-hive_${HIVE_VERSION}
|
|
|
|
ENV ENABLE_INIT_DAEMON true
|
|
ENV INIT_DAEMON_BASE_URI http://identifier/init-daemon
|
|
ENV INIT_DAEMON_STEP spark_master_init
|
|
|
|
ARG SPARK_VERSION=2.3.1
|
|
ARG SPARK_HADOOP_VERSION=2.7
|
|
|
|
ENV SPARK_VERSION ${SPARK_VERSION}
|
|
ENV HADOOP_VERSION ${SPARK_HADOOP_VERSION}
|
|
|
|
COPY wait-for-step.sh /
|
|
COPY execute-step.sh /
|
|
COPY finish-step.sh /
|
|
|
|
RUN echo "Installing Spark-version (${SPARK_VERSION})" \
|
|
&& wget http://apache.mirror.iphh.net/spark/spark-${SPARK_VERSION}/spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz \
|
|
&& tar -xvzf spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz \
|
|
&& mv spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION} /opt/spark \
|
|
&& rm spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz \
|
|
&& cd /
|
|
|
|
#Give permission to execute scripts
|
|
RUN chmod +x /wait-for-step.sh && chmod +x /execute-step.sh && chmod +x /finish-step.sh
|
|
|
|
# Fix the value of PYTHONHASHSEED
|
|
# Note: this is needed when you use Python 3.3 or greater
|
|
ENV PYTHONHASHSEED 1
|
|
|
|
ENV SPARK_HOME /opt/spark
|
|
ENV SPARK_INSTALL ${SPARK_HOME}
|
|
ENV SPARK_CONF_DIR ${SPARK_HOME}/conf
|
|
ENV PATH $SPARK_INSTALL/bin:$PATH
|
|
|
|
ENV SPARK_DRIVER_PORT 5001
|
|
ENV SPARK_UI_PORT 5002
|
|
ENV SPARK_BLOCKMGR_PORT 5003
|
|
|
|
EXPOSE $SPARK_DRIVER_PORT $SPARK_UI_PORT $SPARK_BLOCKMGR_PORT
|
|
|
|
# Without this spark-shell fails - Download if it is not already there in $SPARK_INSTALL
|
|
RUN wget -nc -q -O "${SPARK_INSTALL}/jars/jersey-bundle-1.19.4.jar" "http://repo1.maven.org/maven2/com/sun/jersey/jersey-bundle/1.19.4/jersey-bundle-1.19.4.jar"
|
|
|