All Products
Search

This Product

    E-MapReduce:Export and Import Service Configurations

    Document Center

    E-MapReduce:Export and Import Service Configurations

    Last Updated:Jul 17, 2026

    Alibaba Cloud E-MapReduce (EMR) allows you to export service configurations in XML or JSON format for backup, migration, and recovery. You can import JSON-formatted service configurations into a new cluster as a preset configuration.

    Export Service Configuration

    1. Log on to the E-MapReduce console.

    2. In the top menu bar, select the appropriate region and resource group.

    3. On the EMR on ECS page, in the Actions column of the target cluster, select more > Export Service Configuration.

    4. In the Export Service Configuration dialog box, select the service configuration files that you want to export.

      Parameters

      Note

      Export Mode

      Selected by default. When selected, only custom or modified configurations are exported. When cleared, all service configurations in the current cluster are exported.

      Export Format

      Select an export format based on your needs.

      • XML: Back up, migrate, and restore EMR service configurations.

      • JSON: Use as a software preset configuration for a new cluster. For more information, see Import Service Configurations.

    5. Click OK.

    Import Service Configurations

    You can use a JSON-formatted service configuration as a preset configuration for a new cluster during cluster creation.

    1. On the EMR on ECS page, click CREATE_CLUSTER.

    2. In Software Configuration, in the Advanced Settings section, turn on the Custom Software Configuration switch.

    3. Copy the content of the exported JSON-formatted service configuration into the configuration box.

      After you click the Turn on switch in the Custom Software Configuration area, paste the JSON into the code editor. This JSON contains two configuration objects, both for the hadoop-env file of HADOOP-COMMON: the first sets EXTRA_ENVS to export HADOOP_CLASSPATH=$HADOOP_CLASSPATH:/opt/apps/JINDOSDK/jindosdk-current/lib/* to add the JindoSDK dependency, and the second uses HADOOP_CLIENT_BUILDIN_OPTS to set client heap memory parameters when HADOOP_HEAPSIZE is empty.

    4. Click Next: Hardware Configuration to continue creating the cluster. For more information, see Create Cluster.