错误为：-ModuleNotFoundError: No module named 'pyspark' While running Pyspark in docker

Question

Getting the error as:

Traceback (most recent call last): File “/opt/application/main.py”, line 6, in from pyspark import SparkConf, SparkContext ModuleNotFoundError: No module named 'pyspark'

While running pyspark in docker.

And my dockerfile is as follows:

FROM centos
ENV DAEMON_RUN=true
ENV SPARK_VERSION=2.4.7
ENV HADOOP_VERSION=2.7
WORKDIR /opt/application
RUN yum -y install python36
RUN yum -y install wget
ENV PYSPARK_PYTHON python3.6
ENV PYSPARK_DRIVER_PYTHON python3.6
RUN ln -s /usr/bin/python3.6 /usr/local/bin/python
RUN wget https://bootstrap.pypa.io/get-pip.py
RUN python get-pip.py
RUN pip3.6 install numpy
RUN pip3.6 install pandas
RUN wget --no-verbose 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${HADOOP_VERSION} spark \
      && rm spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz
ENV SPARK_HOME=/usr/local/bin/spark
RUN yum -y install java-1.8.0-openjdk
ENV JAVA_HOME /usr/lib/jvm/jre
COPY main.py .
RUN chmod +x /opt/application/main.py
CMD ["/opt/application/main.py"]

Answer 1

You forgot to install pyspark in your dockerfile.

FROM centos
ENV DAEMON_RUN=true
ENV SPARK_VERSION=2.4.7
ENV HADOOP_VERSION=2.7
WORKDIR /opt/application
RUN yum -y install python36
RUN yum -y install wget
ENV PYSPARK_PYTHON python3.6
ENV PYSPARK_DRIVER_PYTHON python3.6
RUN ln -s /usr/bin/python3.6 /usr/local/bin/python
RUN wget https://bootstrap.pypa.io/get-pip.py
RUN python get-pip.py
RUN pip3.6 install numpy
RUN pip3.6 install pandas
RUN pip3.6 install pyspark  # add this line.
RUN wget --no-verbose 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${HADOOP_VERSION} spark \
      && rm spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz
ENV SPARK_HOME=/usr/local/bin/spark
RUN yum -y install java-1.8.0-openjdk
ENV JAVA_HOME /usr/lib/jvm/jre
COPY main.py .
RUN chmod +x /opt/application/main.py
CMD ["/opt/application/main.py"]

Edit : dockerfile improvment:

FROM centos
ENV DAEMON_RUN=true
ENV SPARK_VERSION=2.4.7
ENV HADOOP_VERSION=2.7
WORKDIR /opt/application
RUN yum -y install python36 wget java-1.8.0-openjdk  # you could install python36 and wget in once
ENV PYSPARK_PYTHON python3.6
ENV PYSPARK_DRIVER_PYTHON python3.6
RUN ln -s /usr/bin/python3.6 /usr/local/bin/python
RUN wget https://bootstrap.pypa.io/get-pip.py \
    && python get-pip.py \
    && pip3.6 install numpy==1.19 pandas==1.1.5 pyspark==3.0.2  # you should also pin the version you need, pandas 1.2.x does not support python 3.6
RUN wget --no-verbose 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${HADOOP_VERSION} spark \
      && rm spark-${SPARK_VERSION}-bin-hadoop${HADOOP_VERSION}.tgz
ENV SPARK_HOME=/usr/local/bin/spark
ENV JAVA_HOME /usr/lib/jvm/jre
COPY main.py .
RUN chmod +x /opt/application/main.py
CMD ["/opt/application/main.py"]