Configuring for network deployment

Network deployment allows you to manage a group of servers centrally from a deployment manager. A deployment manager manages the configuration and state for all of the managed nodes in its cell and deploys applications to any managed node in the cell.

Before you begin

Before configuring for network deployment, ensure that you have completed the following steps:
  • Decide which network deployment scenario you are creating.
  • Decide whether you want to use both distributed and z/OS nodes in the same cell.
  • Decide whether you want to use multiple cells.
  • Decide whether or not to use server clusters.

Why and when to perform this task

If you have decided that your solution requires network deployment, then you require a deployment manager and managed nodes.

To configure for network deployment, complete the following steps:

Steps for this task

  1. Create and start the deployment manager, ready to add managed nodes to the cell.
    1. On one system, configure a deployment manager.
    2. Start the deployment manager. For more information about how to start the deployment manager, see Starting or stopping the server.
    3. On the deployment manager, start the administrative console. For more information about how to do this, see Starting and stopping the administrative console.
  2. Add managed nodes to the cell.
    1. Optional: Federate a stand-alone server profile to the deployment manager, as the first managed node. After using a stand-alone server to validate the use of service applications, you can federate that profile into a new deployment manager cell. This enables you to evolve the use of those service applications into a distributed group of servers and server clusters, on one or more nodes.

      This step adapts the SCA* buses of the stand-alone server to be the SCA* buses of the deployment manager cell.

    2. Configure a managed node, as a custom profile, on a different system to the deployment manager. When configuring a custom profile, you can federate it to the deployment manager, as an empty managed node.
    3. If you did not federate the custom profile when creating a managed node, you must now federate that custom profile to the deployment manager. For more information about how to do this, see Federating custom nodes to a deployment manager. You now have an empty managed node.
  3. Optional: Configure your own choice of data store configuration for messaging engines on the SCA* buses.

    When you configure a server cluster to host destinations for the SCA runtime, a new messaging engine is created on the SCA* buses. You must explicitly configure both the data store and the JDBC data source that the messaging engine uses to interact with its data store.

    Similarly, if you configure a server for the SCA runtime, a new messaging engine is created on the SCA* buses and configured to use a Cloudscape data source by default. This default configuration enables the messaging engine to run without needing any additional configuration by the Administrator. If you do not want to use the default data source configuration, you can change the configuration parameters. For example, you can change the data source or you can configure the data store to use a different JDBC provider.

    For more information about configuring your choice of data store configuration for messaging engines, see Setting up the data store for a messaging engine.

  4. Configure a server cluster for each managed node.
    1. Create a server cluster. For more information about how to do this, see Creating a server cluster.
    2. Configure the cluster for the SCA runtime required by mediation modules, as described in Configuring servers or clusters for the SCA runtime. Before deploying any mediation modules onto the new cluster, configure the cluster to indicate where queue destinations used by the SCA runtime for the module are to be hosted.
    3. Start the cluster. For more information about how to do this, see Managing available server clusters
    Repeat this step for each managed node that is to support a server cluster.
  5. Configure any servers to run outside clusters. It is not common to add single servers to a network deployment environment. Generally, it is advisable to have clusters in your network deployment environment to provide high availability.
    1. On a managed node, create a server. For more information about how to do this, see Creating a server.
    2. Configure the server for the SCA runtime required by mediation modules, as described in Configuring servers or clusters for the SCA runtime. Before deploying any mediation modules onto the new server, configure the server to indicate where queue destinations used by the SCA runtime for the module are to be hosted.
    Repeat this step for each managed node that is to support a server.

Last updated: Wed 06 Dec 2006 07:08:08

(c) Copyright IBM Corporation 2005, 2006.
This information center is powered by Eclipse technology (http://www.eclipse.org)