Newer
Older
docker-analytics / spark / Dockerfile
FROM python:3.6-alpine3.9

ENV SPARK_VERSION="2.4.3" \
    HADOOP_VERSION="2.7" \
    SPARK_INSTALL="/usr/local"

RUN apk add --no-cache \
        bash \
        openjdk8 \
        tini \
        zeromq

RUN apk add --no-cache --virtual .fetch-deps \
        curl \
        tar

RUN curl -s https://www-us.apache.org/dist/spark/spark-$SPARK_VERSION/spark-$SPARK_VERSION-bin-hadoop$HADOOP_VERSION.tgz | tar -xz -C $SPARK_INSTALL && \
    cd $SPARK_INSTALL && ln -s spark-$SPARK_VERSION-bin-hadoop$HADOOP_VERSION spark

RUN apk del .fetch-deps

COPY start-master.sh /usr/local/bin/start-master.sh
COPY start-worker.sh /usr/local/bin/start-worker.sh

ENV SPARK_HOME="$SPARK_INSTALL/spark" \
    SPARK_LOCAL_IP="127.0.0.1" \
    SPARK_MASTER_PORT="7077" \
    SPARK_MASTER_WEBUI_PORT="8080" \
    PYSPARK_PYTHON="/usr/local/bin/python3"

# Spark doesn't seem to respond directly to SIGTERM as the exit status is
# for SIGKILL (137), after a pause. Presumably docker-compose down times out.
# Using tini gives immediate exit with status 143 (SIGTERM).
ENTRYPOINT ["/sbin/tini", "--"]

CMD ["/usr/local/bin/start-master.sh"]