Run administrative jobs across multiple nodes

Use this topic to run administrative jobs across multiple nodes in a flexible management environment .

Configure a job manager, administrative agent, and register nodes with the job manager to set up a flexible management environment.

After configuring the flexible management environment, we can submit, monitor, and manage jobs for the nodes that are registered with the job manager. Your ID at the job manager must be authorized for the administrator role or the operator role to submit jobs. When you submit a job, we can specify a user name and password for authentication and authorization at the target node or nodes. If we do not provide a user name and password in the job parameters, the credentials for the job submitter at the job manager are used for this purpose.

Note: When you submit a job to multiple target nodes, the user name and password or the credentials for the submitter must be applicable to all of the job targets.

In the following example, a company has four branch offices located in Hursley, Berlin, Warsaw, and Paris, with an application server at each location. An application that runs on each application server frequently experiences a memory leak issue. While a development team fixes the application, it might be necessary for each location to frequently stop and restart the application server. Use the following steps to create a node group consisting of the application servers from each location, and schedule weekly recurring jobs to stop and restart the four application servers.

1. Launch wsadmin.sh.

   Navigate to the app_server_root/bin directory and use the following command to connect wsadmin.sh to the job manager process:

   wsadmin -profileName myJobManager -lang jython

2. Create a node group.

   Use the createManagedNodeGroup command to create a new node group in the flexible management configuration. The following example commands creates the European_Branch_Offices node group:

   AdminTask.createManagedNodeGroup('-groupName European_Branch_Offices -description "Management group for all European branch offices"')

3. Add nodes to the node group.

   Use the addMemberToManagedNodeGroup command to add nodes to the new group. Each node can belong to one or more groups. The following example commands add the Hursley_office, Berlin_office, Warsaw_office, and Paris_office nodes to the European_Branch_Offices node group:

   AdminTask.addMemberToManagedNodeGroup('-groupName European_Branch_Offices -managedNodeNameList "[Hursley_Node01][Berlin_Node01][Warsaw_Node01][Paris_Node01]"')

4. Schedule an administrative job on the job manager.

   Use the submitJob command to submit the future administrative job. Job submissions consist of the following information:

   Job type

   The job type specifies the type of job to perform. Many jobs exist in the flexible management environment including application management, product maintenance, configuration, and application server runtime control jobs.

   Job target list and target group

   The job target list and target group specifies the nodes and node group where the job runs.

   Job specific parameters

   Most administrative jobs require configuration information in addition to the job type and target in order to run the job. Job parameters are specific to each job type.

   When you submit a job to multiple target nodes, the user name and password or the credentials for the submitter must apply to all the job targets.

   Optional generic parameters

   In addition to the job specific parameters, we can include any of the following optional parameters with the job submission:

   descriptions. Run a submitJob command with its required parameters and with optional parameters appropriate

   Parameter	Description	Type
   username	Username to use to submit the job when security is enabled.	String
   password	Password for the username to use to submit the job when security is enabled.	String
   description	Specifies a description for the job.	String
   activationDateTime	Date and time to activate the job in the format "2006-05-03T10:30:45-0000". The -0000 section of the activationDateTime parameter value represents RFC 822 format. We can specify Z as a shortcut for Greenwich Mean Time (GMT), such as "2006-05-03T10:30:45Z". If we do not specify the time zone, the system uses the time zone of the server.	String
   expirationDateTime	Expiration date for the job, in the format "2006-05-03T10:30:45-0000". The -0000 section of the activationDateTime parameter value represents RFC 822 format. We can specify Z as a shortcut for Greenwich Mean Time (GMT), such as "2006-05-03T10:30:45Z". If we do not specify the time zone, the system uses the time zone of the server.	String
   expireAfter	Amount of time, in minutes, to wait before the job expires.	Integer
   executionWindow	Recurring interval for the job.	String
   executionWindowUnit	Recurring interval unit of measure for the value set by the executionWindow parameter. Specify DAILY to run the job daily, WEEKLY to run the job weekly, MONTHLY to run the job monthly, YEARLY to run the job annually, or CONNECTION to run the job each time the node connects and polls for jobs.	String
   email	Email address where the system sends job notifications.	String

   The following example schedules two weekly recurring jobs. For this example, an application that runs on application servers at 4 branch offices frequently experiences a memory leak issue. While a development team fixes the application, it might be necessary to frequently stop and restart the application server. The following command examples schedule the job manager to stop and restart the server once a week, and notifies the system administrator when the server stops and restarts:

   AdminTask.submitJob('-jobType stopServer -group European_Branch_Offices   -jobParams [serverName server1] -activationDateTime 2006-05-03T10:30:45Z
    -executionWindowUnit DAILY -executionWindow 13:00:00-14:00:00
    -email system_admin@company.com')
    AdminTask.submitJob('-jobType startServer -group European_Branch_Offices
    -jobParams [serverName server1] -activationDateTime 2006-05-03T10:40:45Z
    -recurringIntervalUnits DAILY -recurringInterval 13:00:00-14:00:00
    -email system_admin@company.com')

Results

The jobs have been submitted to the queue and will run at the date and time specified by the command.

What to do next

Use the commands in the AdministrativeJobs command group to manage and query for administrative jobs in the flexible management configuration.

Related tasks

Registering nodes with the job manager using commands

AdministrativeJobs (AdminTask)

ManagedNodeAgent (AdminTask)

JobManagerNode (AdminTask)