Newer
Older
docker-analytics / kafka / Dockerfile
FROM analytics/spark

ENV KAFKA_VERSION="2.2.0" \
    SCALA_VERSION="2.12" \
    KAFKA_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/kafka/$KAFKA_VERSION/kafka_$SCALA_VERSION-$KAFKA_VERSION.tgz | tar -xz -C $KAFKA_INSTALL && \
    cd $KAFKA_INSTALL && ln -s kafka_$SCALA_VERSION-$KAFKA_VERSION kafka

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 KAFKA_HOME="$KAFKA_INSTALL/kafka"
#     SPARK_LOCAL_IP="127.0.0.1" \
#     SPARK_MASTER_PORT="7077" \
#     SPARK_MASTER_WEBUI_PORT="8080" \
#     PYSPARK_PYTHON="/usr/local/bin/python3"
COPY start-kafka.sh /usr/local/bin/start-kafka.sh
COPY server.properties $KAFKA_HOME/config/server.properties

# 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-kafka.sh"]