Dockerfile

FROM debian:9

ENV HADOOP_VERSION 3.2.1
ENV HADOOP_URL https://www.apache.org/dist/hadoop/common/hadoop-$HADOOP_VERSION/hadoop-$HADOOP_VERSION.tar.gz
ENV HADOOP_PREFIX /usr/local/hadoop

RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \
      openjdk-8-jdk \
      net-tools \
      curl \
      netcat \
      gnupg \
      procps \
    && rm -rf /var/lib/apt/lists/*

ENV JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64

VOLUME /mnt/hadoop/

# Download hadoop.
RUN set -x && curl -fSL "$HADOOP_URL" -o /hadoop.tar.gz

RUN set -x && tar -xzf /hadoop.tar.gz -C /usr/local/ \
     && cd /usr/local \
     && ln -s ./hadoop-$HADOOP_VERSION hadoop \
     && rm -rf /var/lib/apt/lists/*


ENV HADOOP_COMMON_HOME $HADOOP_PREFIX
ENV HADOOP_HDFS_HOME $HADOOP_PREFIX
ENV HADOOP_MAPRED_HOME $HADOOP_PREFIX
ENV HADOOP_YARN_HOME $HADOOP_PREFIX
ENV HADOOP_CONF_DIR $HADOOP_PREFIX/etc/hadoop
ENV YARN_CONF_DIR $HADOOP_PREFIX/etc/hadoop

WORKDIR $HADOOP_PREFIX

RUN sed -i '/^export JAVA_HOME/ s:.*:export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64\nexport HADOOP_PREFIX=/usr/local/hadoop\nexport HADOOP_HOME=/usr/local/hadoop\n:' $HADOOP_PREFIX/etc/hadoop/hadoop-env.sh  \
  && sed -i '/^export HADOOP_CONF_DIR/ s:.*:export HADOOP_CONF_DIR=/usr/local/hadoop/etc/hadoop/:' $HADOOP_PREFIX/etc/hadoop/hadoop-env.sh \
  && chmod +x /usr/local/hadoop/etc/hadoop/*-env.sh

COPY core-site.xml.template hdfs-site.xml.template $HADOOP_PREFIX/etc/hadoop/

COPY bootstrap.sh fence.sh /etc/

# NameNode                Secondary NameNode  DataNode                     JournalNode  NFS Gateway    HttpFS         ZKFC
EXPOSE 8020 50070 50470   50090 50495         50010 1004 50075 1006 50020  8485 8480    2049 4242 111  14000 14001    8019