Reconfigure an instance group in a running cluster - Amazon EMR
Considerations when you reconfigure an instance groupReconfigure an instance group in the consoleReconfigure an instance group using the CLIReconfigure an instance group using the Java SDKTroubleshoot

Reconfigure an instance group in a running cluster

With Amazon EMR version 5.21.0 and later, you can reconfigure cluster applications and specify additional configuration classifications for each instance group in a running cluster. To do so, you can use the Amazon EMR console, the AWS Command Line Interface (AWS CLI), or the AWS SDK. When you update an application configuration for an instance group, Amazon EMR merges the new configuration with the existing configuration to create a new active configuration.

After you submit a reconfiguration request for an instance group, Amazon EMR assigns a version number to the new configuration specification. You can track the version number of a configuration, or the state of an instance group, by viewing the CloudWatch events. For more information, see Monitor CloudWatch Events.

Note

You can only override, and not delete, cluster configurations that were specified during cluster creation. If there are differences between the existing configuration and the file that you supply, Amazon EMR resets manually modified configurations, such as configurations that you have modified while connected to your cluster using SSH, to the cluster defaults for the specified instance group.

Considerations when you reconfigure an instance group

Reconfiguration actions

When you submit a reconfiguration request using the Amazon EMR console, the AWS Command Line Interface (AWS CLI), or the AWS SDK, Amazon EMR checks the existing on-cluster configuration file. If there are differences between the existing configuration and the file that you supply, Amazon EMR initiates reconfiguration actions, restarts some applications, and resets any manually modified configurations, such as configurations that you have modified while connected to your cluster using SSH, to the cluster defaults for the specified instance group.

Note

Amazon EMR performs some default actions during every instance group reconfiguration. These default actions might conflict with cluster customizations that you have made, and result in reconfiguration failures. For information about how to troubleshoot reconfiguration failures, see Troubleshoot instance group reconfiguration.

Amazon EMR also initiates reconfiguration actions for the configuration classifications that you specify in your request. For a complete list of these actions, see the Configuration Classifications section for the version of Amazon EMR that you use. For example, 6.2.0 Configuration Classifications.

Note

The Amazon EMR Release Guide only lists reconfiguration actions starting with Amazon EMR versions 5.32.0 and 6.2.0.

Service disruption

Amazon EMR follows a rolling process to reconfigure instances in the Task and Core instance groups. Only 10 percent of the instances in an instance group are modified and restarted at a time. This process takes longer to finish but reduces the chance of potential application failure in a running cluster.

To run YARN jobs during a YARN restart, you can either create an Amazon EMR cluster with multiple master nodes or set yarn.resourcemanager.recovery.enabled to true in your yarn-site configuration classification. For more information about using multiple master nodes, see High availability YARN ResourceManager.

Application validation

Amazon EMR checks that each application on the cluster is running after the reconfiguration restart process. If any application is unavailable, the overall reconfiguration operation fails. If a reconfiguration operation fails, Amazon EMR reverses the configuration parameters to the previous working version.

Note

To avoid reconfiguration failure, we recommend that you only install applications on your cluster that you plan to use. We also recommend that you make sure all cluster applications are healthy and running before you submit a reconfiguration request.

Types of reconfiguration

You can reconfigure an instance group in one of two ways:

  • Overwrite. Default reconfiguration method and the only one available in Amazon EMR releases earlier than 5.35.0 and 6.6.0. This reconfiguration method indiscriminately overwrites any on-cluster files with the newly submitted configuration set. The method erases any changes to configuration files made outside the reconfiguration API.

  • Merge. Reconfiguration method supported for Amazon EMR versions 5.35.0 and 6.6.0 and later, except from the Amazon EMR console, where no version supports it. This reconfiguration method merges the newly submitted configurations with configurations that already exist on the cluster. This option only adds or modifies the new configurations that you submit. It preserves existing configurations.

Note

Amazon EMR continues to overwrite some essential Hadoop configurations that it needs to ensure that the service is running correctly.

Limitations

When you reconfigure an instance group in a running cluster, consider the following limitations:

  • Non-YARN applications can fail during restart or cause cluster issues, especially if the applications aren't configured properly. Clusters approaching maximum memory and CPU usage may run into issues after the restart process. This is especially true for the master instance group.

  • You can't submit a reconfiguration request when an instance group is being resized. If a reconfiguration is initiated while an instance group is resizing, reconfiguration cannot start until the instance group has completed resizing, and vice versa.

  • After reconfiguring an instance group, Amazon EMR restarts the applications to allow the new configurations to take effect. Job failure or other unexpected application behavior might occur if the applications are in use during reconfiguration.

  • If a reconfiguration for an instance group fails, Amazon EMR reverses the configuration parameters to the previous working version. If the reversion process fails too, you must submit a new ModifyInstanceGroup request to recover the instance group from the SUSPENDED state.

  • Reconfiguration requests for Phoenix configuration classifications are only supported in Amazon EMR version 5.23.0 and later, and are not supported in Amazon EMR version 5.21.0 or 5.22.0.

  • Reconfiguration requests for HBase configuration classifications are only supported in Amazon EMR version 5.30.0 and later, and are not supported in Amazon EMR versions 5.23.0 through 5.29.0.

  • Amazon EMR supports application reconfiguration requests on an EMR cluster with multiple master nodes only in Amazon EMR versions 5.27.0 and later.

  • Reconfiguring hdfs-encryption-zones classification or any of the Hadoop KMS configuration classifications is not supported on an EMR cluster with multiple master nodes.

  • Amazon EMR currently doesn't support certain reconfiguration requests for the capacity scheduler that require restarting the YARN ResourceManager. For example, you cannot completely remove a queue.

Reconfigure an instance group in the console

Note

The Amazon EMR console does not support Merge type reconfigurations.

  1. Open the Amazon EMR console at https://console.aws.amazon.com/elasticmapreduce/.

  2. In the cluster list under Name, choose the active cluster that you want to reconfigure.

  3. Open the cluster details page for the cluster, and go to the Configurations tab.

  4. In the Filter drop-down list, select the instance group that you want to reconfigure.

  5. In the Reconfigure drop-down menu, choose either Edit in table or Edit in JSON file.

    • Edit in table - In the configuration classification table, edit the property and value for existing configurations, or choose Add configuration to supply additional configuration classifications.

    • Edit in JSON file - Enter the configuration directly in JSON, or use shorthand syntax (demonstrated in shadow text). Otherwise, provide an Amazon S3 URI for a file with a JSON Configurations object.

    Note

    The Source column in the configuration classification table indicates whether the configuration is supplied when you create a cluster, or when you specify additional configurations for this instance group. You can edit the configurations for an instance group from both sources. You cannot delete initial cluster configurations, but you can override them for an instance group.

    You can also add or edit nested configuration classifications directly in the table. For example, to supply an additional export sub-classification of hadoop-env, add a hadoop.export configuration classification in the table. Then, provide a specific property and value for this classification.

  6. (Optional) Select Apply this configuration to all active instance groups.

  7. Save the changes.

Reconfigure an instance group using the CLI

Use the modify-instance-groups command to specify a new configuration for an instance group in a running cluster.

Note

In the following examples, replace <j-2AL4XXXXXX5T9> with your cluster ID, and replace <ig-1xxxxxxx9> with your instance group ID.

Example – Replace a configuration for an instance group

The following example references a configuration JSON file called instanceGroups.json to edit the property of the YARN NodeManager disk health checker for an instance group.

  1. Prepare your configuration classification, and save it as instanceGroups.json in the same directory where you will run the command.

    
[
   {
      "InstanceGroupId":"<ig-1xxxxxxx9>",
      "Configurations":[
         {
            "Classification":"yarn-site",
            "Properties":{
               "yarn.nodemanager.disk-health-checker.enable":"true",
               "yarn.nodemanager.disk-health-checker.max-disk-utilization-per-disk-percentage":"100.0"
            },
            "Configurations":[]
         }
      ]
   }
]

  2. Run the following command.

    aws emr modify-instance-groups --cluster-id <j-2AL4XXXXXX5T9> \
--instance-groups file://instanceGroups.json

Example – Add a configuration to an instance group

If you want to add a configuration to an instance group, you must include all previously specified configurations for that instance group in your new ModifyInstanceGroup request. Otherwise, the previously specified configurations are removed.

The following example adds a property for the YARN NodeManager virtual memory checker. The configuration also includes previously specified values for the YARN NodeManager disk health checker so that the values won't be overwritten.

  1. Prepare the following contents in instanceGroups.json and save it in the same directory where you will run the command.

    
[
   {
      "InstanceGroupId":"<ig-1xxxxxxx9>",
      "Configurations":[
         {
            "Classification":"yarn-site",
            "Properties":{
               "yarn.nodemanager.disk-health-checker.enable":"true",
               "yarn.nodemanager.disk-health-checker.max-disk-utilization-per-disk-percentage":"100.0",
               "yarn.nodemanager.vmem-check-enabled":"true",
               "yarn.nodemanager.vmem-pmem-ratio":"3.0"
            },
            "Configurations":[]
         }
      ]
   }
]

  2. Run the following command.

    aws emr modify-instance-groups --cluster-id <j-2AL4XXXXXX5T9> \
--instance-groups file://instanceGroups.json

Example – Add a configuration to an instance group with Merge type reconfiguration

When you want to use the default Overwrite reconfiguration method to add a configuration, you must include all previously specified configurations for that instance group in your new ModifyInstanceGroup request. Otherwise, the Overwrite removes the configurations that you previously specified. You don't need to do this with Merge reconfiguration. Instead, you must ensure only that the new configurations are included.

The following example adds a property for the YARN NodeManager virtual memory checker. Because this is a Merge type reconfiguration, it does not overwrite previously specified values for the YARN NodeManager disk health checker.

  1. Prepare the following contents in instanceGroups.json and save it in the same directory where you will run the command.

    
[
   {"InstanceGroupId":"<ig-1xxxxxxx9>",
    "ReconfigurationType" :"MERGE",
      "Configurations":[
         {"Classification":"yarn-site",
            "Properties":{
               "yarn.nodemanager.vmem-check-enabled":"true",
               "yarn.nodemanager.vmem-pmem-ratio":"3.0"
            },
            "Configurations":[]
         }
      ]
   }
]

  2. Run the following command.

    
aws emr modify-instance-groups --cluster-id <j-2AL4XXXXXX5T9> \
--instance-groups file://instanceGroups.json

Example – Delete a configuration for an instance group

To delete a configuration for an instance group, submit a new reconfiguration request that excludes the previous configuration.

Note

You can only override the initial cluster configuration. You cannot delete it.

For example, to delete the configuration for the YARN NodeManager disk health checker from the previous example, submit a new instanceGroups.json with the following contents. 


[
   {
      "InstanceGroupId":"<ig-1xxxxxxx9>",
      "Configurations":[
         {
            "Classification":"yarn-site",
            "Properties":{
               "yarn.nodemanager.vmem-check-enabled":"true",
               "yarn.nodemanager.vmem-pmem-ratio":"3.0"
            },
            "Configurations":[]
         }
      ]
   }
]
Note

To delete all of the configurations in your last reconfiguration request, submit a reconfiguration request with an empty array of configurations. For example,

[
   {
      "InstanceGroupId":"<ig-1xxxxxxx9>",
      "Configurations":[]
   }
]

Example – Reconfigure and resize an instance group in one request

The following example JSON demonstrates how to reconfigure and resize an instance group in the same request.


[
   {
      "InstanceGroupId":"<ig-1xxxxxxx9>",
      "InstanceCount":5,
      "EC2InstanceIdsToTerminate":["i-123"],
      "ForceShutdown":true,
      "ShrinkPolicy":{
         "DecommissionTimeout":10,
         "InstanceResizePolicy":{
            "InstancesToTerminate":["i-123"],
            "InstancesToProtect":["i-345"],
            "InstanceTerminationTimeout":20
         }
      },
      "Configurations":[
         {
            "Classification":"yarn-site",
            "Configurations":[],
            "Properties":{
               "yarn.nodemanager.disk-health-checker.enable":"true",
               "yarn.nodemanager.disk-health-checker.max-disk-utilization-per-disk-percentage":"100.0"
            }
         }
      ]
   }
]

Reconfigure an instance group using the Java SDK

Note

In the following examples, replace <j-2AL4XXXXXX5T9> with your cluster ID, and replace <ig-1xxxxxxx9> with your instance group ID.

The following code snippet provides a new configuration for an instance group using the AWS SDK for Java.


AWSCredentials credentials = new BasicAWSCredentials("access-key", "secret-key");
AmazonElasticMapReduce emr = new AmazonElasticMapReduceClient(credentials);

Map<String,String> hiveProperties = new HashMap<String,String>();
hiveProperties.put("hive.join.emit.interval","1000");
hiveProperties.put("hive.merge.mapfiles","true");
        
Configuration configuration = new Configuration()
    .withClassification("hive-site")
    .withProperties(hiveProperties);
    
InstanceGroupModifyConfig igConfig = new InstanceGroupModifyConfig()
    .withInstanceGroupId("<ig-1xxxxxxx9>")
    .withReconfigurationType("MERGE");
    .withConfigurations(configuration);

ModifyInstanceGroupsRequest migRequest = new ModifyInstanceGroupsRequest()
    .withClusterId("<j-2AL4XXXXXX5T9>")
    .withInstanceGroups(igConfig);

emr.modifyInstanceGroups(migRequest);

The following code snippet deletes a previously specified configuration for an instance group by supplying an empty array of configurations.


List<Configuration> configurations = new ArrayList<Configuration>();

InstanceGroupModifyConfig igConfig = new InstanceGroupModifyConfig()
    .withInstanceGroupId("<ig-1xxxxxxx9>")
    .withConfigurations(configurations);

ModifyInstanceGroupsRequest migRequest = new ModifyInstanceGroupsRequest()
    .withClusterId("<j-2AL4XXXXXX5T9>")
    .withInstanceGroups(igConfig);

emr.modifyInstanceGroups(migRequest);

Troubleshoot instance group reconfiguration

If the reconfiguration process for an instance group fails, Amazon EMR reverts the reconfiguration and logs a failure message using an Amazon CloudWatch event. The event provides a brief summary of the reconfiguration failure. It lists the instances for which reconfiguration has failed and corresponding failure messages. The following is an example failure message.

The reconfiguration operation for instance group ig-1xxxxxxx9 in Amazon EMR cluster j-2AL4XXXXXX5T9 (ExampleClusterName) 
failed at 2021-01-01 00:00 UTC and took 2 minutes to fail. Failed configuration version is example12345. 
Failure message: Instance i-xxxxxxx1, i-xxxxxxx2, i-xxxxxxx3 failed with message "This is an example failure message".

To gather more data about a reconfiguration failure, you can check the node provisioning logs. Doing so is particularly useful when you receive a message like the following.

i-xxxxxxx1 failed with message “Unable to complete transaction and some changes were applied.”
On the node

To access node provisioning logs by connecting to a node

  1. Use SSH to connect to the node on which reconfiguration has failed. For instructions, see Connect to your Linux instance in the Amazon EC2 User Guide for Linux Instances.

  2. Navigate to the following directory, which contains the node provisioning log files.

    /mnt/var/log/provision-node/

  3. Open the reports subdirectory and search for the node provisioning report for your reconfiguration. The reports directory organizes logs by reconfiguration version number, universally unique identifier (UUID), Amazon EC2 instance IP address, and timestamp. Each report is a compressed YAML file that contains detailed information about the reconfiguration process.

    The following is an example report file name and path.

    /reports/2/ca598xxx-cxxx-4xxx-bxxx-6dbxxxxxxxxx/ip-10-73-xxx-xxx.ec2.internal/202104061715.yaml.gz

  4. You can examine a report using a file viewer like zless, as in the following example.

    zless 202104061715.yaml.gz
Amazon S3

To access node provisioning logs using Amazon S3

  1. Sign in to the AWS Management Console and open the Amazon S3 console at https://console.aws.amazon.com/s3/.

  2. Open the Amazon S3 bucket that you specified when you configured the cluster to archive log files.

  3. Navigate to the following folder, which contains the node provisioning log files:

    DOC-EXAMPLE-BUCKET/elasticmapreduce/<cluster id>/node/<instance id>/provision-node/

  4. Open the reports folder and search for the node provisioning report for your reconfiguration. The reports folder organizes logs by reconfiguration version number, universally unique identifier (UUID), Amazon EC2 instance IP address, and timestamp. Each report is a compressed YAML file that contains detailed information about the reconfiguration process.

    The following is an example report file name and path.

    /reports/2/ca598xxx-cxxx-4xxx-bxxx-6dbxxxxxxxxx/ip-10-73-xxx-xxx.ec2.internal/202104061715.yaml.gz

  5. To view a log file, you can download it from Amazon S3 to your local machine as a text file. For instructions, see Downloading an object.

Each log file contains a detailed provisioning report for the associated reconfiguration. To find error message information, you can search for the err log level of a report. Report format depends on the version of Amazon EMR on your cluster.

The following example shows error information for Amazon EMR release versions earlier than 5.32.0 and 6.2.0.

- !ruby/object:Puppet::Util::Log
      level: !ruby/sym err
      tags: 
        - err
      message: "Example detailed error message."
      source: Puppet
      time: 2021-01-01 00:00:00.000000 +00:00

Amazon EMR release versions 5.32.0 and 6.2.0 and later use the following format instead.

- level: err
  message: 'Example detailed error message.'
  source: Puppet
  tags:
  - err
  time: '2021-01-01 00:00:00.000000 +00:00'
  file: 
  line: