Important YARN entries in dss.properties
Last updated
Last updated
# SPECIFY CORRECT JAVA_HOME
yarn.mode.java.home=/usr/jdk64/jdk1.8.0_60/
yarn.mode.java.xmx=-Xmx2048m
# SPECIFY QUEUE (Check correct queue, if you are using your own hadoop, use default)
yarn.mode.queue.name=default
# SPECIFY CONTAINER AND SPARK RESOURCES
yarn.mode.container.resource.memory=2500
yarn.mode.container.resource.cores=1
spark.driver.memory=1g
spark.driver.cores=1
spark.executor.memory=2g
spark.executor.cores=1
spark.executor.instances=1
# SPECIFY NAME NODE AND RESOURCE MANAGER CONTEXT
yarn.mode.name.node.root.context=hdfs://hklpadhas01.hk.standardchartered.com:8020
yarn.mode.resource.manager.context=
# SPECIFY user.name FOR SECURE ACCESS
yarn.api.additional.request.params=?user.name=ec2-user
# SPECIFY CORRECT PATHS FOR TDSS LIBS / SPARK FILES / UDF FILES IN HADOOP
yarn.mode.app.jar.path=<base dir>/tdss/lib
yarn.mode.other.lib.path=<base dir>/tdss/spark
yarn.mode.spark.lib.path=<base dir>/tdss/spark/spark-assembly-1.6.1-hadoop2.6.0.jar
yarn.mode.udf.dep.path=<base dir>/tdss/udf
# SPECIFY correct egg file for db access (MySQL / jaydebeapi etc)
yarn.mode.db.egg.file=JayDeBeApi-1.1.1-py2.7.egg
# CLASS PATH ENTRIES. Verify hadoop lib paths from admin managing hadoop cluster. Below replace /home/ownappsg with correct hadoop path.
yarn.mode.hadoop.home=/home/ownappsg/hadoop/etc/hadoop/share/hadoop/common/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/common/lib/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/hdfs/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/hdfs/lib/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/yarn/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/yarn/lib/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/mapreduce/*<CPS>/home/ownappsg/hadoop/etc/hadoop/share/hadoop/mapreduce/lib/*
yarn.mode.hadoop.conf=/home/ownapsg/hadoop/etc/hadoop/conf
# SPECIFY correct hadoop version
yarn.mode.hadoop.version=2.6.5
# SPECIFY user used to launch yarn jobs
yarn.mode.hadoop.user.name=ownappsg
# BELOW THIS COPY PROPERTIES GENERATED BY RUNNING get_hdfs_files_timestamp_size.sh above OUTPUT TO tdss_properties