Release Notes (10.4.1.3)

Release Notes (10.4.1.3)

Copy .jar Files from the Cluster to the Domain

Copy .jar Files from the Cluster to the Domain

Depending on the run-time engine and the cluster version, copy Hive .jar files from the cluster to the domain, and remove existing .jar files.

Blaze engine

Copy files from a directory on the cluster to a directory on each machine that hosts the Data Integration Service, and then delete a file from the domain directory.
  1. Depending on the cluster version, copy files from the EMR cluster to the target directory on the domain.
    EMR 5.29 Source Files
    • /usr/share/aws/hmclient/lib/aws-glue-datacatalog-hive3-client-3.0.0.jar
    • /usr/share/aws/hmclient/lib/hive-exec-1.2.1-spark2-amzn-1.jar
    EMR 6.1 Source Files
    • /usr/share/aws/hmclient/lib/aws-glue-datacatalog-hive3-client-3.0.0.jar
    • /usr/share/aws/hmclient/lib/hive-exec-3.1.2-amzn-2.jar
    • /usr/lib/hive/lib/hive-common-3.1.2-amzn-2.jar
      Verify that the version of hive-common-3.1.2-amzn-2.jar from the cluster replaces the version of the file that existed in the domain directory.
    Target Directory
    /services/shared/hadoop/EMR_<version>/lib/
  2. Remove the following file from the domain machine: /services/shared/hadoop/EMR_<version>/lib/hive-exec-3.1.2-amzn-0.jar.

Spark engine

Copy files from a directory on the cluster to a directory on each machine that hosts the Data Integration Service, and then delete files from the domain directory.
  1. Depending on the cluster version, copy files from the EMR cluster to the target directory on the domain.
    EMR 5.29 Source Files
    • /usr/share/aws/hmclient/lib/aws-glue-datacatalog-spark-client-3.0.0.jar
    • /usr/lib/spark/jars/hive-exec-1.2.1-spark2-amzn-1.jar
    EMR 6.1 Source Files
    • /usr/share/aws/hmclient/lib/aws-glue-datacatalog-spark-client-3.0.0.jar
    • /usr/lib/spark/jars/hive-exec-2.3.7-amzn-0-core.jar
    • /usr/lib/spark/jars/hive-common-2.3.7-amzn-0.jar
    Target Directory
    /services/shared/spark/lib_spark_3.0.1_hadoop_3.2.0
  2. Remove the following files from the domain machine:
    • /services/shared/spark/lib_spark_3.0.1_hadoop_3.2.0/hive-exec-2.3.7-core.jar
    • /services/shared/spark/lib_spark_3.0.1_hadoop_3.2.0/hive-common-2.3.7.jar