Red Hat Training

A Red Hat training course is available for Red Hat JBoss Enterprise Application Platform

Chapter 20. Configuring Batch Applications

JBoss EAP 7 introduced support for Java batch applications as defined by JSR-352. You can configure an environment for running batch applications and manage batch jobs using the batch-jberet subsystem.

For information on developing batch applications, see Java Batch Application Development in the JBoss EAP Development Guide.

20.1. Configuring Batch Jobs

You can configure settings for batch jobs using the batch-jberet subsystem, which is based on the JBeret implementation.

The default batch-jberet subsystem configuration defines an in-memory job repository and default thread pool settings. 

<subsystem xmlns="urn:jboss:domain:batch-jberet:1.0">
    <default-job-repository name="in-memory"/>
    <default-thread-pool name="batch"/>
    <job-repository name="in-memory">
        <in-memory/>
    </job-repository>
    <thread-pool name="batch">
        <max-threads count="10"/>
        <keepalive-time time="30" unit="seconds"/>
    </thread-pool>
</subsystem>

By default, any batch jobs stopped during a server suspend will be restarted upon server resume. You can set the restart-jobs-on-resume property to false to leave jobs in the STOPPED state instead. 

/subsystem=batch-jberet:write-attribute(name=restart-jobs-on-resume,value=false)

You can also configure the settings for batch job repositories and thread pools.

20.1.1. Configure Batch Job Repositories

This section shows you how to configure in-memory and JDBC job repositories for storing batch job information using the management CLI. You can also configure job repositories using the management console by navigating to the Batch subsystem from the Configuration tab and selecting either In Memory or JDBC from the left-hand menu.

Add an In-memory Job Repository

You can add a job repository that stores batch job information in memory. 

/subsystem=batch-jberet/in-memory-job-repository=REPOSITORY_NAME:add
Add a JDBC Job Repository

You can add a job repository that stores batch job information in a database. You must specify the name of the datasource for connecting to the database. 

/subsystem=batch-jberet/jdbc-job-repository=REPOSITORY_NAME:add(data-source=DATASOURCE)
Set a Default Job Repository

You can set an in-memory or JDBC job repository as the default job repository for batch applications. 

/subsystem=batch-jberet:write-attribute(name=default-job-repository,value=REPOSITORY_NAME)

This will require a server reload. 

reload

20.1.2. Configure Batch Thread Pools

This section shows you how to configure thread pools and thread factories to be used for batch jobs using the management CLI. You can also configure thread pools and thread factories using the management console by navigating to the Batch subsystem from the Configuration tab and selecting Thread Pools or Thread Factories from the left-hand menu.

Configure a Thread Pool

When adding a thread pool, you must specify the max-threads, which should always be greater than 3 as two threads are reserved to ensure partition jobs can execute as expected.

  1. Add a thread pool. 

    /subsystem=batch-jberet/thread-pool=THREAD_POOL_NAME:add(max-threads=10)

  2. If desired, set a keepalive-time value. 

    /subsystem=batch-jberet/thread-pool=THREAD_POOL_NAME:write-attribute(name=keepalive-time,value={time=60,unit=SECONDS})
Use a Thread Factory

  1. Add a thread factory. 

    /subsystem=batch-jberet/thread-factory=THREAD_FACTORY_NAME:add

  2. Configure the desired attributes for the thread factory.

    • group-name - The name of a thread group to create for this thread factory.
    • priority - The thread priority of created threads.

    • thread-name-pattern - The template used to create names for threads. The following patterns may be used:

      • %% - A percent sign
      • %t - The per-factory thread sequence number
      • %g - The global thread sequence number
      • %f - The factory sequence number
      • %i - The thread ID

  3. Assign the thread factory to a thread pool. 

    /subsystem=batch-jberet/thread-pool=THREAD_POOL_NAME:write-attribute(name=thread-factory,value=THREAD_FACTORY_NAME)

    This will require a server reload. 

    reload
Set a Default Thread Pool

You can set a different thread pool as the default thread pool. 

/subsystem=batch-jberet:write-attribute(name=default-thread-pool,value=THREAD_POOL_NAME)

This will require a server reload. 

reload
View Thread Pool Statistics

You can view runtime information about a batch thread pool using the read-resource management CLI operation. You must use the include-runtime=true parameter in order to see this runtime information. 

/subsystem=batch-jberet/thread-pool=THREAD_POOL_NAME:read-resource(include-runtime=true)
{
    "outcome" => "success",
    "result" => {
        "active-count" => 0,
        "completed-task-count" => 0L,
        "current-thread-count" => 0,
        "keepalive-time" => undefined,
        "largest-thread-count" => 0,
        "max-threads" => 15,
        "name" => "THREAD_POOL_NAME",
        "queue-size" => 0,
        "rejected-count" => 0,
        "task-count" => 0L,
        "thread-factory" => "THREAD_FACTORY_NAME"
    }
}

You can also view runtime information for batch thread pools using the management console by navigating to the Batch subsystem from the Runtime tab.

20.2. Managing Batch Jobs

The batch-jberet subsystem resource for deployments allows you to start, stop, and restart batch jobs. You can also view the details of job executions.

Restart a Batch Job

You can restart a job that is in a STOPPED or FAILED state by providing its execution ID and optionally any properties to use when restarting the batch job. 

/deployment=DEPLOYMENT_NAME/subsystem=batch-jberet:restart-job(execution-id=EXECUTION_ID,properties={PROPERTY=VALUE})

The execution ID must be the most recent execution of the job instance.

Start a Batch Job

You can start a batch job by providing the job XML file and optionally any properties to use when starting the batch job. 

/deployment=DEPLOYMENT_NAME/subsystem=batch-jberet:start-job(job-xml-name=JOB_XML_NAME,properties={PROPERTY=VALUE})

Stop a Batch Job

You can stop a running batch job by providing its execution ID. 

/deployment=DEPLOYMENT_NAME/subsystem=batch-jberet:stop-job(execution-id=EXECUTION_ID)

View Batch Job Execution Details

You can view the details of batch job executions. You must use the include-runtime=true parameter in order to see this runtime information. 

/deployment=DEPLOYMENT_NAME/subsystem=batch-jberet:read-resource(recursive=true,include-runtime=true)
{
    "outcome" => "success",
    "result" => {"job" => {"import-file" => {
        "instance-count" => 2,
        "running-executions" => 0,
        "execution" => {
            "2" => {
                "batch-status" => "COMPLETED",
                "create-time" => "2016-04-11T22:03:12.708-0400",
                "end-time" => "2016-04-11T22:03:12.718-0400",
                "exit-status" => "COMPLETED",
                "instance-id" => 58L,
                "last-updated-time" => "2016-04-11T22:03:12.719-0400",
                "start-time" => "2016-04-11T22:03:12.708-0400"
            },
            "1" => {
                "batch-status" => "FAILED",
                "create-time" => "2016-04-11T21:57:17.567-0400",
                "end-time" => "2016-04-11T21:57:17.596-0400",
                "exit-status" => "Error : org.hibernate.exception.ConstraintViolationException: could not execute statement",
                "instance-id" => 15L,
                "last-updated-time" => "2016-04-11T21:57:17.597-0400",
                "start-time" => "2016-04-11T21:57:17.567-0400"
            }
        }
    }}}
}