Use this task to create a server cluster, which is a set of servers
that are managed together and participate in workload management.
Before you begin
Before you can create a server cluster, you must have created a
managed node in a deployment manager cell. You can do this by federating either
a stand-alone server (as the first managed node) or a custom node to the deployment
manager..
Why and when to perform this task
If you want to balance workload, such as service requests, over
a set of servers, you can create a server cluster, then add servers as members
of that cluster. You can also create a backup cluster, to provide
failover support for the server cluster that it is assigned to.
To create a server cluster,
use the administrative console to complete the following steps:
Steps for this task
- In the navigation pane, click .
- Click New.
- On the Enter basic cluster information. page,
specify the following details:
- For Cluster name, type a name for the cluster.
- Optional: To disable node-scoped routing optimization,
clear Prefer local.
By default, the Prefer
local option is selected. This indicates that, if possible, Enterprise
JavaBean (EJB) requests are routed to the client node. If you enable this
feature, performance is improved because client requests are sent to local
enterprise beans.
- Optional: To create a replication domain for this
cluster, select Create a replication domain for this cluster.
Use replication domains to transfer data, objects, or events for
session manager, dynamic cache, or stateful session beans among the application
servers in a cluster. Create a separate replication domain to use with each
component that acts as a consumer of the replication. For example, you can
configure one replication domain to use with a session manager and another
domain to use with dynamic cache. The replication domain name that is created
is identical to the cluster name.
For more information about using replication
domains, see Replicating data across application
servers in a cluster.
- Click Next.
- On the Create cluster members page, specify
details of any cluster members that you want to create with the cluster. You can create cluster members (servers in the cluster) at this time,
or add new members to the cluster after creating the cluster.
For each new
cluster member, perform the following actions:
- For Member name, type the name of a new server
to add as a member of the cluster.
- For Select node, select the node on which the
server runs.
- For Weight, specify the server
weight.
The weight value controls the amount of work directed
to the server. If the weight value for the server is greater than the weight
values assigned to other servers in the cluster, then the server receives
a larger share of the workload. The value can range from 0 to 20.
- Make sure that Generate Unique Http Ports is
selected. By generating unique HTTP ports for the server, you avoid
potential port collisions and configurations that are not valid.
- Under Select template:, select the template
that you want to use to create all members of the cluster. The
template can either be a default server template or an existing server that
was created from the required template. If you want the cluster to support
service applications for mediation modules on an enterprise service bus, select
one of the following from the drop down lists provided:
- The server template defaultESBServer (distributed platforms) or defaultESBServerZOS
(z/OS platform)
- An existing application server that you know was created from the server
template defaultESBServer (distributed platforms) or defaultESBServerZOS (z/OS
platform).
For more information about the templates, see Server templates.
- Click Apply to finish the cluster member. You can add
more cluster members now or later. All cluster members you add are based on
the same server template, so there are less steps to complete.
Each cluster member that you specify now is added to the list
displayed. You can use the Edit function to edit the properties of a cluster
member in the list. You can use the Delete function to remove a cluster member
from the list.
- Click Next.
- Check the details on the View the summary page.
To create the cluster click Finish.
- Check that the server template indicated is correct; for example: Clone
Template = defaultESBServer
- If more than one node group bounds the selected node, the node group DefaultNodeGroup
is chosen by the system. After creating the cluster, you can change the bounding
node group in the Server cluster configuration view.
If any of the details that you selected are incorrect, you can click Previous,
to return to earlier pages and change the details.
- Save your changes to the master configuration. As part
of saving the change to the configuration, you can select Synchronize
changes with Nodes before clicking Save on
the Save page.
- Before you can start the cluster, the configuration needs to be
synchronized to the nodes. If you selected Synchronize changes with
Nodes when saving your configuration in the previous step, you can
ignore this step.
Otherwise, take one of the following actions:
- If you are using automatic synchronization, wait until synchronization
runs.
- Run manual synchronization to get the configuration files moved to the
nodes.
- Click
- On the Nodes page, select the node name.
- Click Synchronize or Full resynchronize.
The Nodes page displays status indicating whether the node is synchronized.
What to do next
If the cluster is to support mediation modules, use the Configure
your Network Deployment Environment guided activity to configure
the cluster for the SCA runtime, as described in Configuring servers or clusters for the SCA runtime.
You
can display the properties and local topology of the cluster. Click .
You
can change the configuration of the cluster. Note that if you have not clicked Save and
saved your administrative configuration, you only see the Configuration and Local
Topology tabs. To see the Runtime tab, you
must save your administrative configuration. Also, ensure that changes are
synchronized to the nodes as described above.
You can create
more cluster members or Starting clustersstart a server group. For
more information about cluster configuration options, see Balancing
workloads with clusters.
You can use scripting to automate the
task of creating clusters. For information about using scripting for clusters,
see Creating
clusters using scripting.
To
manage a server cluster, you can use the Server
cluster collection.
You can use this page to display the properties and topology
of a cluster, and to make changes as needed.
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)