Reputation: 3711
I'm using gettyimages as the spark master container, while at the same time I've got a spark image that's going to be launching a slave node. Here is my corresponding docker file.
FROM debian:jessie
RUN apt-get update \
&& apt-get install -y locales \
&& dpkg-reconfigure -f noninteractive locales \
&& locale-gen C.UTF-8 \
&& /usr/sbin/update-locale LANG=C.UTF-8 \
&& echo "en_US.UTF-8 UTF-8" >> /etc/locale.gen \
&& locale-gen \
&& apt-get clean \
&& rm -rf /var/lib/apt/lists/*
# Users with other locales should set this in their derivative image
ENV LANG en_US.UTF-8
ENV LANGUAGE en_US:en
ENV LC_ALL en_US.UTF-8
RUN apt-get update \
&& apt-get install -y curl unzip \
&& apt-get clean \
&& rm -rf /var/lib/apt/lists/*
# JAVA
ARG JAVA_MAJOR_VERSION=8
ARG JAVA_UPDATE_VERSION=92
ARG JAVA_BUILD_NUMBER=14
ENV JAVA_HOME /usr/jdk1.${JAVA_MAJOR_VERSION}.0_${JAVA_UPDATE_VERSION}
ENV PATH $PATH:$JAVA_HOME/bin
RUN curl -sL --retry 3 --insecure \
--header "Cookie: oraclelicense=accept-securebackup-cookie;" \
"http://download.oracle.com/otn-pub/java/jdk/${JAVA_MAJOR_VERSION}u${JAVA_UPDATE_VERSION}-b${JAVA_BUILD_NUMBER}/server-jre-${JAVA_MAJOR_VERSION}u${JAVA_UPDATE_VERSION}-linux-x64.tar.gz" \
| gunzip \
| tar x -C /usr/ \
&& ln -s $JAVA_HOME /usr/java \
&& rm -rf $JAVA_HOME/man
# HADOOP
ENV HADOOP_VERSION 2.7.2
ENV HADOOP_HOME /usr/hadoop-$HADOOP_VERSION
ENV HADOOP_CONF_DIR=$HADOOP_HOME/etc/hadoop
ENV PATH $PATH:$HADOOP_HOME/bin
RUN curl -sL --retry 3 \
"http://archive.apache.org/dist/hadoop/common/hadoop-$HADOOP_VERSION/hadoop-$HADOOP_VERSION.tar.gz" \
| gunzip \
| tar -x -C /usr/ \
&& rm -rf $HADOOP_HOME/share/doc \
&& chown -R root:root $HADOOP_HOME
# SPARK
ENV SPARK_VERSION 2.0.1
ENV SPARK_PACKAGE spark-${SPARK_VERSION}-bin-without-hadoop
ENV SPARK_HOME /usr/spark-${SPARK_VERSION}
ENV SPARK_DIST_CLASSPATH="$HADOOP_HOME/etc/hadoop/*:$HADOOP_HOME/share/hadoop/common/lib/*:$HADOOP_HOME/share/hadoop/common/*:$HADOOP_HOME/share/hadoop/hdfs/*:$HADOOP_HOME/share/hadoop/hdfs/lib/*:$HADOOP_HOME/share/hadoop/hdfs/*:$HADOOP_HOME/share/hadoop/yarn/lib/*:$HADOOP_HOME/share/hadoop/yarn/*:$HADOOP_HOME/share/hadoop/mapreduce/lib/*:$HADOOP_HOME/share/hadoop/mapreduce/*:$HADOOP_HOME/share/hadoop/tools/lib/*"
ENV PATH $PATH:${SPARK_HOME}/bin
RUN curl -sL --retry 3 \
"http://d3kbcqa49mib13.cloudfront.net/${SPARK_PACKAGE}.tgz" \
| gunzip \
| tar x -C /usr/ \
&& mv /usr/$SPARK_PACKAGE $SPARK_HOME \
&& chown -R root:root $SPARK_HOME
WORKDIR $SPARK_HOME
CMD ["bin/spark-class","org.apache.spark.deploy.worker.Worker", //TODO: Figure out what this should be]
I'm wondering how could I get the slave to be able to access the masters host and port if I'm setting this up through docker compose.
Upvotes: 1
Views: 1843
Reputation: 5275
Lets say you have docker-compose
something like:
version: '2'
services:
spark-master:
image: spark-master
ports:
- "7077:7077"
- "8080:8080"
spark-slave1:
image: spark-slave
ports:
- "8081:8081"
depends_on:
- spark-master
in the Dockerfile
for your spark slave you need to define master like:
./bin/spark-class org.apache.spark.deploy.worker.Worker spark://IP:PORT
but using IP address is not very good idea. So you can use hostname from docker-compose (spark-master
is the hostname and it added to /etc/hosts
):
CMD ["bin/spark-class","org.apache.spark.deploy.worker.Worker", "spark://spark-master:7077"]
now you can go to:
DOCKER_IP:8080
-> and see 1 worker in "Workers"
DOCKER_IP:8081
-> and see details of worker
In case you want to have more workers you can add another services to docker-compose
. Following docker-compose
will create 2 workers - details of 1. worker on port 8081
and second on 8082
.
version: '2'
services:
spark-master:
image: spark-master
ports:
- "7077:7077"
- "8080:8080"
spark-slave1:
image: spark-slave
ports:
- "8081:8081"
depends_on:
- spark-master
spark-slave2:
image: spark-slave
ports:
- "8082:8081"
depends_on:
- spark-master
Upvotes: 1