System administration for WebSphere Application Server V7: Part 3: Administering a flexible management topology

This tutorial demonstrates how to set up and manage the new flexible management topologies in IBM WebSphere Application Server V7. You will set up an administrative agent to manage one or more base application server nodes, and register the base application server nodes or the deployment manager cells with the job manager for multi-cell management. You will learn about the flexible management job life cycle, and the job manager interfaces for submitting and managing administrative jobs using wsadmin jython scripting, and the administrative console.

From the IBM WebSphere Developer Technical Journal.

Syed-Muasir Khalil, Software Engineer, IBM

Syed-Muasir Khalil works as a lead in the WebSphere naming functional test team, and is also a function test developer for WebSphere application management components.



Nazgol Sedghi (nsedghi@us.ibm.com), Staff Software Engineer, IBM

Nazgol Sedghi is a Software Engineer in WebSphere Application Server System Management test team in Austin, Texas. She led the functional testing effort of Flexible Management in WebSphere Application Server Version 7, with special focus on usability and security.



Nagesh Mvgv (nagemvgv@in.ibm.com), Senior Staff Software Engineer, IBM

Nagesh M V G V is a senior staff software engineer at IBM India Software Lab, Hyderabad. Nagesh works with the WebSphere FVT team working mainly with system management.



Brian Hanczaryk (hanczar@us.ibm.com), Staff Software Engineer, IBM

Brian Hanczaryk is a staff software engineer at IBM in Research Triangle Park, North Carolina. Brian currently works on the WebSphere System Verification Test team and tests both coexistence and multiple profile creation on distributed platforms.



04 March 2009

Also available in Chinese

Introduction

IBM WebSphere Application Server Version 7.0 contains many significant enhancements in system management functionality over the previous release, Version 6.1. This tutorial explores a newly introduced new style of system management called flexible management. Flexible management topology offers an asynchronous non-blocking job queuing mechanism for administration purposes. This asynchronous topology is not meant to replace the existing Network Deployment deployment manager or base servers. Rather, it offers administrators new management options not available before. For example, you can:

  • Manage multiple base servers, up to a server farm containing hundreds of base servers.
  • Coordinate management actions across multiple Network Deployment deployment managers.
  • Manage branch servers that are geographically dispersed and reachable only through a low bandwidth, high latency network.
  • Submit asynchronous administrative jobs that:
    • Take effect at a specified time.
    • Expire after a specified time.
    • Recur at a specified time interval.
    • Notify the administrator when job is complete through e-mail or JMX notification.

This tutorial will guide you through creating and configuring the various profiles involved in flexible management topology. After completing this tutorial, you will be able to:

  • Create and configure a flexible management topology.
  • Understand the difference between Network Deployment and flexible management topologies.
  • Understand the life cycle of jobs.
  • Understand the available administrative jobs and commands in the flexible management topology.

This tutorial contains six exercises to help you understand the different management options available:

  1. Create a flexible management topology
  2. Understand job manager configuration options
  3. Configure polling intervals for the managed nodes
  4. Create a managed node group
  5. Learn about the job life cycle, step-by-step
  6. Understanding managed node resources and their status

Before you begin

Prerequisites

Read the previous installment of this series, Part 2: New administrative topologies, before you begin this tutorial.

System requirements

The system on which you perform the exercises in this tutorial must have a supported operating system and a proper installation of WebSphere Application Server V7.0. For more information on supported operating systems, see the WebSphere Application Server V7.0 Information Center. In addition, the system must have the following resources available:

  • Approximately 500MB of hard drive space to store the installation images.
  • An appropriate installation of WebSphere Application Server V7.0.
  • Another 500MB of hard drive space for the installed product.

Duration

1.5 hours


Exercise 1: Create a flexible management topology

To take advantage of this new administrative style, you need to create a flexible management topology. At a minimum, you need a job manager to queue up jobs, and an administrative agent with at least one registered node or a deployment manager cell.

For this exercise, you must have WebSphere Application Server V7.0 installed on two systems, which will be referred to here as system1 and system2.

  1. Create Job Manager profile
    1. Assuming WebSphere Application Server is installed on system1 under C:\WebSphere\ND directory, open a command line and change the directory to C:\WebSphere\ND\bin\ProfileManagement.
    2. Run pmt.bat to launch the Profile Management Tool.
    3. Click Launch Profile Management Tool.
    4. Click Create. The Environment Selection panel displays (Figure 1).
      Figure 1. Environment Selection
      Figure 1. Environment Selection
    5. Click Next. The Server Type Selection panel displays (Figure 2).
      Figure 2. Figure Type Selection
      Figure 2. Figure Type Selection
    6. Select Job manager and click Next.
    7. The Profile Creation Options panel displays next (not shown). Select Advance Profile Creation from the menu. Click Next.
    8. Accept the default values on the Optional Application Deployment panel (not shown). Click Next.
    9. The Profile Name and Location panel displays (Figure 3). Click Next.
      Figure 3. Profile Name and Location
      Figure 3. Profile Name and Location
    10. The Node, Host and Cell Names panel displays (Figure 4). Accept the default values or specify a unique node name, the actual host name of the machine, and a unique cell name. Click Next.
      Figure 4. Node, Host and Cell Names
      Figure 4. Node, Host and Cell Names
    11. On the Administrative Security panel (Figure 5), select the Enable administrative security checkbox and enter values for User name, Password and Confirm password fields. Click Next.
      Figure 5. Administrative Security
      Figure 5. Administrative Security
    12. Click Next repeatedly until the Profile Creation Summary panel displays (Figure 6).
      Figure 6. Profile Creation Summary
      Figure 6. Profile Creation Summary
    13. After reviewing the information in the summary, click Create . The profile creation starts, which might take a while to complete. Upon completion, the Profile Creation Complete panel displays (Figure 7).
      Figure 7. Profile Creation Complete
      Figure 7. Profile Creation Complete
    14. Click Finish. You have now successfully created a job manager profile named "jobmanager." A server called "jobmgr" is created for the job manager profile.
    15. Now, start the jobmgr server with startServer command, the same way you start any other server:

      Jobmanager_profile_root\bin\startServer.bat jobmgr

  2. Create the Network Deployment cell
    1. Use system1 to create a Cell profile. Start the Profile Management tool as you did in Step1 and click Create.
    2. In Environment Selection panel, select the Cell as the environment, as shown in Figure 8.
      Figure 8. Environment selection
      Figure 8. Environment selection
    3. Follow the steps on the wizard to complete the creation of a cell environment on system1. This step will create a deployment manager profile, Dmgr01, and an application server profile, AppSrv01.
    4. Start the deployment manager.
    5. Start the nodeagent and server1 of the federated node.
  3. Create administrative agent and base profiles
    1. Use system2 to create the administrative agent and the base profiles.
    2. Start the Profile Management Tool and click Create.
    3. Select the management environment from the Environment Selection panel and click Next.
    4. Select Administrative agent from the Server Type Selection panel (Figure 9) and follow the steps in the wizard until profile creation is complete. Know that you need to start the adminagent server in the administrative agent profile before you can register any base application servers with the administrative agent.
      Figure 9. Server Type Selection
      Figure 9. Server Type Selection
    5. Create an application server profile on the system that you created the administrative agent (system2). You will register it later with the administrative agent that you created above.
  4. Register the base application server nodes with the administrative agent

    You can register a standalone node with an administrative agent by running the registerNode command, which is available under the administrative agent profile bin directory.

    Before you begin registering nodes with a job manager, first make sure that the clocks on the various computer systems in the topology are synchronized. Just as when you federate base nodes with a deployment manager, you might experience invalid security token errors if system clocks are out of sync by more than five minutes (irrespective of time zones). To synchronize system clocks, first stop all WebSphere servers (node agent, administrative agent, deployment manager, job manager, and application servers). Update the clocks and then restart the servers.

    Before registration starts, make sure:

    • The administrative agent is on the same system as the standalone base node being registered.
    • The administrative agent is running.
    • The base node is a standalone node.
    • The server is stopped on the base node.
    Table 1. Options for the registerNode command
    ParameterDescription
    -conntype <JSR160RMI|IPC|RMI|SOAP>(optional) Connector type used to connect to the administrative agent.
    -host host_name(optional) Specifies the host name of the administrative agent.
    -name managed_node_name(optional) Specifies the name of the managed node after the node is registered with the administrative agent.
    -nodepassword node_password (optional) Specifies the password of the node that you are registering.
    -nodeusername node_user_name (optional) Specifies the use name of the node that you are registering.
    -openConnectors connectors (optional) Specifies a list of connectors separated by commas of connectors that the administrative agent will open.
    -port port_number (optional) Specifies the port number of the administrative agent connector port.
    -profilePath profile_path (required) Specifies the path to the profile of the application server node to be registered.
    -trace (optional) Provides tracing output for the registerNode command.
    -portsFile ports_file (optional) Specifies the path to a file that defines port settings for the newly registered node.
    -profileName profile_name (optional) Specifies the profile name of the administrative agent.
    -username username (optional) Specifies the user ID to log onto the administrative agent.
    -password password (optional) Specifies the password to log onto the administrative agent.

    As a minimum, you need to provide profilePath to register the base node. If you enabled security when creating the profiles, you also need to submit the security credentials. Figure 10 shows the registration of base node using the registerNode command.

    Figure 10. registerNode command
    Figure 10. registerNode command

    Repeat this step for each base node to be registered with the administrative agent.

    To deregister the managed node that is managed by an administrative agent, run the deregisterNode command:

    deregisterNode.bat –profilePath BASE_NODE_PROFILE_PATH –username ADMINAGENT_USER_NAME –password ADMIN_AGENT_PASSWORD

  5. Register the base nodes and the deployment manager with the job manager profile

    Each base node that is registered to an administrative agent or a deployment manager node can be registered with the job manager. When a standalone base node is registered with an administrative agent, much of the administrative code that was in the base server is contained by the administrative agent. This results in a significantly smaller and faster starting base server. Each administrative agent can register any of the profiles it manages with the job manager.

    You can register a node with the job manager by running the registerWithJobManager AdminTask command.

    Table 2. Options for registerWithJobmanager command
    ParameterDescription
    -managedNodeName(required) Specifies the name of the node that is registered with the administrative agent. If the node is a deployment manager profile, specify the node name of the deployment manager.
    -host(optional) Specifies the hostname of the job manager.
    -port(optional) Specifies the administrative port number to use. The default secure port number is 9943. The default unsecure port number is 9960.
    -user (optional) Specifies the connector login user name.
    -password (optional) Specifies the password for the connector login user name.
    -alias (optional) Specifies an alias for the node. The job manager uses this name instead of the value of the managedNodeName parameter to register the node. Use this parameter if the new node has the same name of a node that is registered with the job manager.
    -startPolling (optional) Specifies whether to start polling after registering the node. Specify false to disable polling. The default value is true.
    -autoAcceptSigner (required) Specifies whether to automatically accept the signer provided by the server. Specify false to disable this option. The default value is true.

    To register a base node with the job manager:

    1. Navigate to the administrative agent profile bin directory on system2: C:\WebSphere\AppServer\profiles\adminagent\bin.
    2. Launch the wsadmin client to connect to the administrative agent:

      wsadmin –conntype SOAP –port SOAP_PORT –lang jython

    3. Run the AdminTask.registerWithJobManager command in interactive mode. Specify the hostname, port, and security credentials of the job manager, and the name of the base node that is being registered as shown in Figure 11:
      Figure 11. AdminTask.registerWithJobManager command
      Figure 11. AdminTask.registerWithJobManager command

    Repeat this step to register additional base nodes with the job manager.

    To register a deployment manager node with the job manager:

    1. Navigate to the deployment manager profile bin directory on system1: C:\WebSphere\ND\profiles\Dmgr01\bin.
    2. Launch the wsadmin client to connect to the deployment manager:

      wsadmin –conntype SOAP –port SOAP_PORT –lang jython

    3. Run the AdminTask.registerWithJobManager command in interactive mode. Specify the hostname, port, and security credentials of the job manager, and the name of the deployment manager node that is being registered, as shown in Figure 12.
      Figure 12. AdminTask.registerWithJobManager command
      Figure 12. AdminTask.registerWithJobManager command

    The unique id (UUID) of the job manager (in this example, “jobmanager-JOB_MANAGER-01c54275-62f0-4419-9a98-fbcf36105cd6”) is returned when the registration with the job manager is complete.

    If you want to unregister the managed node with the job manager, run the AdminTask command:

    AdminTask.unregisterWithJobManager(‘[-host JOBMANAGER_HOST_NAME –user JOBMANAGER_USER_NAME –password JOBMANAGER_PWD –managedNodeName dmgrManagedNode]’)

Review

In this exercise, you learned how to create a new job manager and administrative agent profiles using the profile management tool. You also learned how to register the base and the deployment manager nodes with the job manager.


Exercise 2: Understand job manager specific administrative console panels

You can find the job manager's configuration page in the administrative console under System administration => Job manager. This panel will enable you to configure and view the job manager properties. You can configure the default job expiration, maximum database results, job manager URL, mail session JNDI name, notification e-mail sender's address, and indicate whether to start components as needed. The panel also provides a button that enables you to stop the job manager from the administrative console. If you choose to stop the job manager from the administrative console, you will be logged out of your current HTTP session.

This exercise navigates you through the different configuration options available on the job manager administrative console.

Assuming you have completed Exercise 1, open the administrative console application of the job manager profile using this URL: http://hostname:jobmanager-administrative-port/ibm/console. The panel should look similar to Figure 13. Table 3 describes the fields displayed on this panel.

Figure 13. Job manager configuration page
Figure 13. Job manager configuration page

Table 3. Job manager configuration options

PropertyDescription
Default job expiration(required) Specifies the default job expiration time in days, with the default being 1 day.
Maximum database results(required) Corresponds to the maximum number of records that can be retrieved during a job manager query/find operation for jobs, managed nodes, and managed node resources. By default, the maximum database results is 10000. Be aware that changing this parameter would adversely affect the performance of the job manager query operations. The higher this value is set, the higher the impact on the performance.
Job manager URL(optional) Specifies the URL of the job manager that the administrative agent uses to fetch jobs. The URL that you specify is used only when the job manager is configured as a proxy server. The specified URL overrides the default URL.
Mail provider JNDI name(optional) Set the mail session JNDI name to be used for job completion e-mail notifications. (Note: “Mail provider” is incorrectly named in the console; it should be “Mail session.”)
Notification e-mail sender's address(optional) Specifies the e-mail address for the sender of the job completion notification messages. This setting is required if you specify a JNDI mail provider, otherwise it is optional.
Start components as needed(optional) Select this checkbox if you want the server components to dynamically start as applications need them. When this checkbox is not selected, all of the server components are started during the startup process. Selecting this property might shorten startup time because fewer components are started during the startup process.

A change to any of above options requires a restart of the job manager server process to take affect.

Review

In this exercise, you learned about the configuration options available on the job manager administrative console.


Exercise 3: Configure polling intervals for the managed nodes

Not only can you use the administrative console (of the administrative agent or the deployment manager) to register and unregister a managed node with the job manager, but you can also change the polling interval for each managed node. The polling interval determines how often the administrative agent or the deployment manager fetches jobs from the job manager for that particular managed node. If a managed node’s specific polling interval is not defined, then this value is set to 30 seconds by default. Polling interval should be tuned as the number of managed node changes in the topology.

At the end of this exercise, you will know how to update the polling interval of a managed node of type Base.

Assuming you have completed Exercise 1, open the administrative console application of the administrative agent profile using this URL: http://hostname:adminagent-administrative-port/ibm/console.

  1. Connect to the administrative agent

    Connect to the administrative agent administrative console and select the appropriate administrative agent radio button on the profile selection panel, then click Continue.

    Figure 14. Integrated Solutions Console
    Figure 14. Integrated Solutions Console

    If you have enabled security, you will be directed to the login page where you will need to provide user ID and password.

  2. Navigate to managed nodes

    Navigate to System administration => Administrative agent => Nodes. You will see a list of registered managed nodes to this administrative agent along with their unique UUIDs. You can also register and unregister a managed node with the job manager from this panel.

    Figure 15. Managed nodes
    Figure 15. Managed nodes
  3. Navigate to the job manager panel

    To set the polling interval for each of the managed nodes, click the link for a specific managed node. You will be directed to the panel where the general properties of the managed node are listed. Click the Job managers link to see which job managers this particular managed node is registered with.

    Figure 16. Job managers
    Figure 16. Job managers

    Each managed node can be registered to multiple job managers. You will find a list of all the job managers with which this managed node is registered on this panel. Notice that from this page you can also unregister and register your managed node with the job manager.

  4. Change the polling interval

    Click the UUID link of the job manager with which this managed node is registered. After changing the polling interval, click OK. You will go back to the previous panel to review the changes and save them.

    Figure 17. Change polling interval
    Figure 17. Change polling interval

Review

In this exercise, you learned how to tune the polling interval of a managed node using the administrative console of the administrative agent.


Exercise 4: Create a managed node group

You can create managed node groups using the job manager administrative console to make the job submission to multiple managed nodes easier. You can define managed groups to logically group the managed nodes in the flexible management topology. For example, the testing team can create a managed node group for all the servers they use. In this exercise, you learn how to create a managed node group.

Assuming you have completed Exercise 1, open the administrative console application of the job manager profile using this URL: http://hostname:jobmanager-administrative-port/ibm/console.

  1. Navigate to node groups

    Navigate to Jobs => Groups of nodes. The Groups of nodes panel is shown in Figure 18.

    Figure 18. Group of nodes
    Figure 18. Group of nodes
  2. Create a new group

    Click New . The General Properties panel of the new managed node group that is going to be created displays (Figure 19).

    Figure 19. Node group general properties
    Figure 19. Node group general properties
    1. Enter Basenode_Group in the group name input box.
    2. Enter base nodes in the description input box.
    3. Type node1 in the Member list input box and click Add. Repeat this step for all the other managed nodes you want add to this group. If you do not remember the names of the nodes to add, press Find to search for the node names.
    4. Click OK to create the new managed node group, Basenode_Group.
  3. Navigate back to the managed node groups panel

    You should see the newly created group there (Figure 20).

    Figure 20. Revised node group
    Figure 20. Revised node group

Review

In this exercise, you learned how to create a managed node group using the job manager administrative console.


Exercise 5: Learn about the job life cycle, step-by-step

A job, such as startServer or installApplication, is an administrative operation that is queued for asynchronous execution on the nodes that are registered to a job manager. To submit a job, you must have at least operator access on the job manager. Once a managed node (base server or deployment manager) is registered with a job manager, you can submit jobs for that base server or deployment manager target. The submitted jobs that are available for execution are queued up in the job manager. The deployment managers or the administrative agent polls for jobs from the job manager periodically. You can submit a job either from the command line using wsadmin client and through the administrative console.

The information in Table 4 needs to be provided during the job submission:

Table 4. Required job information
ParameterDescription
Job typeDistinguishes between different kinds of jobs; for example startServer vs. stopServer jobs. (See Resources for a complete list of built-in jobs.)
Job targetsOne or more managed nodes or a group of managed nodes, where the job is executed.
Username and passwordRefers to the username and password of the managed node(s) where the job is executed.
Job specific parametersSome jobs, like inventory or status jobs, do not have any parameters; others, such as startServer, have specific parameters that you should specify when submitting a job.
Optional generic job parametersThese optional parameters include description, activationDateTime, expirationDateTime, executionWindowUnit, executionWindow, and e-mail where job status notification will be sent if you have configured the job manager mail session properly.

Each submitted job has an associated state and a status. A job state shows where it is in the execution process from the job manager perspective. The predefined states are shown in Table 5.

Table 5. Job states
Job stateDescription
PendingYou have submitted the job, but the job is not available yet to be executed on the targets.
ActiveOne or more targets have started execution of the job.
ExpiredJob has expired. If target has already started to execute the job when it expired, execution continues. No new target will execute the job.
SuspendedJob has been suspended. If target has already started to execute the job when it is suspended, job execution continues. No new target will execute the job.

The status of a job forms a history of job execution against a specific managed node that is saved in a database. Status values are shown in Table 6.

Table 6. Job status
Job statusDescription
NOT_ATTEMPTEDThe agent has not received the job.
DISTRIBUTEDThe agent has received the job.
ASYNC_IN_PROGRESSThe agent is executing the job concurrent with other jobs.
FAILEDThe job execution failed.
REJECTEDThe agent has rejected the job, because it does not support the job type, for example.
SUCCEEDEDJob completed successfully.
PARTIALLY_SUCCEEDEDOnly applicable to start and stopCluster jobs where the cluster may have multiple cluster members, and to start and stopApplication jobs where the application is installed on multiple targets (like an application that is installed on a cluster). If only a portion of the cluster members are started or stopped or the application does not start on all application targets, the status of the job will be PARTIALLY_SUCCEEDED.

After you submit a job, you can query, suspend, resume, or delete a job. You can also check the state and status of a job through either command line (wsadmin client) or administrative console. (See Resources for a list of available commands and a list of predefined and supported job types.)

Job submission is divided into five steps in the job manager administrative console application. In this exercise, you will learn how to submit an install application job to multiple base managed nodes. Before you can submit the install application job, you need to distribute the application ear file using the distribute file job. When the distribute file job is successfully executed, the downloadedContent directory is created in the administrative agent or deployment manager user install root (depending on the type of job target) and the application EAR file is transferred to this directory. The location of the transferred application EAR will be used here as part of the install application job parameters.

Assuming you have completed Exercise 1, open the administrative console application of the job manager profile using this URL: http://hostname:jobmanager-administrative-port>/ibm/console.

In preparation for this exercise:

  1. It is assumed that the bnode2 and node1 (alias for bnode1) nodes are registered to the job manager.
  2. Create a directory called "JobManager" in the jobmanager_user_install_root/config/temp directory.
  3. Copy the jobmanager_user_install_root/installableApps/CacheMonitor.ear file to the jobmanager_user_install_root/config/temp/JobManager directory.

You can now begin to distribute the file and install the application.

A. Distribute the file

  1. Choose job type

    Navigate to Jobs => Submit. Select Distribute file from the Job type drop down menu (Figure 21). Click Next to display the Choose job targets panel (Figure 22).

    Figure 21. Choose job type
    Figure 21. Choose job type
    Figure 22. Choose job targets
    Figure 22. Choose job targets
  2. Choose job targets

    In this step of the job submission process, choose the Node names option (Figure 22) and click Find to see a list of available managed nodes. Find Nodes panel displays (Figure 23).

    Figure 23. Find nodes
    Figure 23. Find nodes
    1. Click Find to see all the nodes that are registered with this job manager.
    2. Choose node1 and bnode2 and click OK. You will be redirected back to the Choose job targets panel (Figure 22) where the list of nodes that you selected are displayed.
    3. Now enter the Node Authentication information and click Next to display the Specify the job parameters panel (Figure 24).
  3. Step 3: Specify job parameters

    Enter file:\CacheMonitor.ear as the value for Source parameter, and CacheMonitor.ear as the value for Destination parameter. Click NEXT.

    Figure 24. Specify job parameters
    Figure 24. Specify job parameters
  4. Schedule the job

    In the Schedule the job panel (Figure 25), enter your e-mail address to receive the job completion notification about the job, if you have already configured the mail session and sender’s e-mail address (see Exercise 2). Accept the defaults for the remaining options and click NEXT.

    Figure 25. Schedule the job
    Figure 25. Schedule the job
  5. Review job summary

    The job summary panel will display the information you selected in previous steps for the job (Figure 26). Review the summary and click Finish when you're done to submit the job.

    Figure 26. Job summary
    Figure 26. Job summary

    On the Job status panel (Figure 27), each submitted job has a unique job ID (for example, "122763380912576341" was created for the job submitted by this tutorial) which you can use to query, suspend, resume, or delete the job. When you click a Job ID, you see the specific properties of that job, including activation and expiration time of the job and its status. If you click the job status link, you will see the job history for each job target.

    Figure 27. Job status
    Figure 27. Job status

You are now ready to install the application CacheMonitor, since the required file has been distributed successfully to the downloadedContent directory in the administrative agent user install root.

B. Install the application

  1. Choose job type

    Navigate to Jobs => Submit. Select Install application from the Job type drop down menu (Figure 28) and click Next to display the Choose job targets panel (Figure 22).

    Figure 28. Choose a job type
    Figure 28. Choose a job type
  2. Choose job targets
    1. Choose the Node names option (Figure 22) and click Find to see a list of available managed nodes.
    2. On the Find Nodes panel (Figure 23), click Find to see all the nodes that are registered with this job manager.
    3. Choose node1 and bnode2 and click OK. You will be redirected back to the Choose job targets panel (Figure 22) where the list of nodes which you selected are displayed.
  3. Specify job parameters

    On the Specify the job parameters panel (Figure 29), enter CacheMonitor as the value of Application name parameter. If no Application location is provided, then the Application location defaults to downloadedContent/<Application name>.ear of the target managed node. Click Find to select the servers on the target nodes, then click Next.

    Figure 29. Specify job parameters
    Figure 29. Specify job parameters
  4. Schedule the job

    In the Schedule the job panel (Figure 25), enter your e-mail address to receive the notification about the job status, accept the default values for the remaining options, and click Next.

  5. Job summary

    The Job summary panel displays the information selected in previous steps to install the application on server1 of both node1 and bnode2 (Figure 30). Click Finish after reviewing the summary to submit the job.

    Figure 30. Review summary and submit job
    Figure 30. Review summary and submit job

    When the CacheMonitor application is installed successfully on server1 or on both bnode2 and node1, the status summary displays in green (Figure 31) to indicate successful job execution.

    Figure 31. Job status
    Figure 31. Job status

Review

In this exercise, you have learned the different steps involved in submitting a job using administrative console of the job manager.


Exercise 6: Understanding managed node resources and their status

In a flexible management environment, you can view server, application, and cluster resources associated with nodes and node groups registered to the job manager. You can also view the status of specific resources at each node, and view properties for a particular node resource as a name-value pair. The type of resources you can view depends on your topology. For example, you cannot view clusters if a deployment manager that you registered to the job manager does not have a defined cluster.

Job manager caches the managed node resource IDs and their attributes. This information is used to display status, and helps you construct job parameters. For example, for server related jobs, you need to identify an actual server. The list of cached servers helps you define to which servers you can submit the job. The managed node cache is updated via two different jobs:

  • The inventory job takes complete inventory, and updates information about the managed nodes, all managed resources, and their attributes. The inventory job also updates information about jobs types and parameters.
  • The status job updates only those managed resources with a “status” attribute. It takes less time to complete than an inventory job.

Be aware that the status of a managed node resource might not be up to date all the time; this is especially true for the status of the servers and applications, as they are subject to more frequent changes. You need to run the inventory or status job to refresh the cached information about a managed node resource that was subject to change.

This exercise provides the steps for you to view managed node resources in your flexible management environment and check their status. You can view a managed node resources from Nodes resources, or from Nodes resources panel.

Assuming you have completed Exercise 1, open the administrative console application of job manager profile using this URL http://hostname:jobmanager-administrative-port/ibm/console.

  1. Navigate to the Node resource collection

    Navigate to Jobs => Node resources (Figure 32).

    Figure 32. Node resources
    Figure 32 Node resources

    Remember that the first time you accessed the Node resource collection panel, no node resources were listed. Enter parameters for the Find option to obtain a list of node resources based on the parameter information that you provide. The next time you view this panel, a list of node resources will be displayed based on the parameters you last specified for the Find option from this console panel.

  2. Get all managed resources

    Select All from the Type drop down menu and click Find. All the available managed resources in your topology will be displayed.

    Figure 33. Get all managed resources
    Figure 33. Get all managed resources
  3. View the status of servers in base nodes

    Select server/server1. The resource ID, the node name this resource belongs to, and its status are displayed in Figure 34.

    Figure 34. Node resource
    Figure 34. Node resource

    As you can see, both nodes bnode2 and node1 (alias for bnode1) have server1. Both servers are stopped.

  4. View node resource properties

    Click the server/server1 resource ID that belongs to bnode2. You will see the properties of this server in the form of name-value pairs (Figure 35).

    Figure 35. Node resource properties
    Figure 35. Node resource properties
  5. Navigate to the Nodes collection panel

    Navigate to Job => Nodes. You can view specific managed node resources from the Nodes collection panel as well (Figure 36).

    Figure 36. Nodes collection
    Figure 36. Nodes collection
  6. View application resources of node1

    Click Display Resources, select Application from the drop down menu, then select node1 (Figure 37).

    Figure 37. Display resources
    Figure 37. Display resources

    A list of all resources of type application that belong to node1 will be displayed (Figure 38).

    Figure 38. Node resources
    Figure 38. Node resources
  7. Select application/DefaultApplication

    The resource ID, the node name this resource belongs to, and its status are displayed (Figure 39). As explained in Step 4 of this exercise, you can click the resource ID and see a list of DefaultApplication properties in form of value-pair.

    Figure 39. Node resource
    Figure 39. Node resource

Review

In the course of this exercise, you learned how to view the resources of managed nodes from both Nodes and Nodes resources collection panels.


Summary

This tutorial introduced you to the flexible management feature in WebSphere Application Server V7.0. While following these exercises, you have learned about the flexible management topologies and the flexible management life cycle, and you have used the tools and facilities available in WebSphere Application Server V7.0 to learn how to:

  • Create a job manager profile and start the job manager.
  • Create an administrative agent profile and start the administrative agent.
  • Create a base profile.
  • Register the base node with the administrative agent.
  • Register the base node with the job manager.
  • Tune a polling interval.
  • Create a new managed node group.
  • Navigate through and understand managed node resources using the job manager administrative console.
  • Use available job manager administrative console explore options.
  • Use the available AdminTask commands for administering flexible management topology.

Resources

Learn

Get products and technologies

Comments

developerWorks: Sign in

Required fields are indicated with an asterisk (*).


Need an IBM ID?
Forgot your IBM ID?


Forgot your password?
Change your password

By clicking Submit, you agree to the developerWorks terms of use.

 


The first time you sign into developerWorks, a profile is created for you. Information in your profile (your name, country/region, and company name) is displayed to the public and will accompany any content you post, unless you opt to hide your company name. You may update your IBM account at any time.

All information submitted is secure.

Choose your display name



The first time you sign in to developerWorks, a profile is created for you, so you need to choose a display name. Your display name accompanies the content you post on developerWorks.

Please choose a display name between 3-31 characters. Your display name must be unique in the developerWorks community and should not be your email address for privacy reasons.

Required fields are indicated with an asterisk (*).

(Must be between 3 – 31 characters.)

By clicking Submit, you agree to the developerWorks terms of use.

 


All information submitted is secure.

Dig deeper into WebSphere on developerWorks


static.content.url=http://www.ibm.com/developerworks/js/artrating/
SITE_ID=1
Zone=WebSphere
ArticleID=373874
ArticleTitle=System administration for WebSphere Application Server V7: Part 3: Administering a flexible management topology
publish-date=03042009