Configure High Availability for SparkMaster

Configure High Availability for the Spark Master so that the master does not become the single point of failure.

By using ZooKeeper to provide leader election and some state storage, you can launch multiple masters in your cluster that are connected to the same ZooKeeper instance. Zookeeper elects one master to be the “leader” and the others remain in standby mode. If the leader goes down, Zookeeper elects another master, recovers the old master’s state, and resumes scheduling.

1. Set SPARK_DAEMON_JAVA_OPTS in spark-env.sh with the appropriate ZooKeeper information for the cluster.

   export SPARK_DAEMON_JAVA_OPTS="-Dspark.deploy.recoveryMode=ZOOKEEPER  
   -Dspark.deploy.zookeeper.url=<zookeeper1:5181,zookeeper2:5181,...> 
   -Djava.security.auth.login.config=/opt/mapr/conf/mapr.login.conf -Dzookeeper.sasl.client=false

2. Restart Spark Master and Spark HistoryServer services:

   maprcli node services -nodes <node-ip> -name spark-master -action restart
   maprcli node services -nodes <node-ip> -name spark-historyserver -action restart

3. On the master node, restart the Spark slaves as the map user:

   /opt/mapr/spark/spark-<version>/sbin/stop-slaves.sh
   /opt/mapr/spark/spark-<version>/sbin/start-slaves.sh