copy bashrc; run not interactive

Author: vangj

spark-cluster/base/Dockerfile

@@ -33,6 +33,9 @@ RUN mkdir /var/run/sshd && \
 # setup ssh
 COPY ubuntu/root/.ssh/* /root/.ssh/
 
+# setup bash
+COPY ubuntu/root/.bashrc /root/.bashrc
+
 # setup hadoop
 RUN wget -q https://dlcdn.apache.org/hadoop/common/hadoop-3.4.1/hadoop-3.4.1.tar.gz -O /tmp/hadoop.tar.gz \
     && tar -xzf /tmp/hadoop.tar.gz -C /usr/local/ \
@@ -65,6 +68,7 @@ RUN wget -q https://repo.anaconda.com/archive/Anaconda3-2024.10-1-Linux-x86_64.s
 RUN rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* \
     && mkdir /tmp/spark-events
 
+# setup supervisor
 COPY ubuntu/etc/supervisor/supervisor.conf /etc/supervisor/supervisor.conf
 COPY ubuntu/etc/supervisor/conf.d/ssh.conf /etc/supervisor/conf.d/ssh.conf
 

spark-cluster/base/Makefile

@@ -5,8 +5,13 @@ PLATFORM=linux/amd64
 build:
 	docker build --platform=$(PLATFORM) -t $(IMAGE_NAME) .
 
-interactive:
-	docker run -it --rm --platform=$(PLATFORM) $(IMAGE_NAME)
+run:
+	docker run \
+		-it \
+		--rm \
+		-p 23:22 \
+		--platform=$(PLATFORM) \
+		$(IMAGE_NAME)
 
 purge: clean
 	docker rmi $(IMAGE_NAME) || true

spark-cluster/base/ubuntu/root/.bashrc

@@ -0,0 +1,18 @@
+export JAVA_HOME=/usr/lib/jvm/java-8-openjdk-amd64
+export HDFS_NAMENODE_USER=root
+export HDFS_DATANODE_USER=root
+export HDFS_SECONDARYNAMENODE_USER=root
+export YARN_RESOURCEMANAGER_USER=root
+export YARN_NODEMANAGER_USER=root
+export YARN_PROXYSERVER_USER=root
+export HADOOP_HOME=/usr/local/hadoop
+export HADOOP_YARN_HOME=${HADOOP_HOME}
+export HADOOP_CONF_DIR=${HADOOP_HOME}/etc/hadoop
+export HADOOP_LOG_DIR=${HADOOP_YARN_HOME}/logs
+export HADOOP_IDENT_STRING=root
+export HADOOP_MAPRED_IDENT_STRING=root
+export HADOOP_MAPRED_HOME=${HADOOP_HOME}
+export SPARK_HOME=/usr/local/spark
+export CONDA_HOME=/usr/local/conda
+export PYSPARK_MASTER=yarn
+export PATH=${CONDA_HOME}/bin:${SPARK_HOME}/bin:${HADOOP_HOME}/bin:${PATH}