Create a new cluster of Sametime® Gateway Servers
by running the Cluster Configuration Wizard. If you are upgrading
an existing IBM® Sametime Gateway cluster,
you can skip this step because there is no need to create a new cluster.
Before you begin
Expected state: the deployment manager is running and nodes
are stopped.
About this task
The instructions that follow describe steps for setting up
a horizontal cluster, the most common cluster configuration. The primary
node already has the primary server installed, so no additional server
is needed on that computer. To add servers to the horizontal cluster,
create one cluster member for each secondary node (computer).For
additional information about default paths, see Directory conventions.
Procedure
- On the deployment manager, open a command window, navigate
to the
stgw_server_root\config
directory,
and run the following command: Linux:
./configwizard.sh
Windows:
configwizard.bat
Note: To
run this program in console mode (instead of using the graphical interface), add the
-console
argument to the command line; for
example:
configwizard.bat -console
-
View the Welcome page, and then click Next.
-
For each secondary node, complete the following steps:
-
Select a secondary node from the Node list and type a description in the
Server Name field. The description you specify is for your use, it does not
map back to an application server.
-
Click Add Member.
- When you have finished adding the secondary nodes, click Next.
- Type the Schema user ID and Schema
password for the database.
These credentials
have appropriate permissions to create tables in the database. You
may need to get this information from the database administrator.
The schema user ID is often the same as the application user ID for
the database.
- Read the summary and click Configure.
When finished, you can view the configuration log at You can review
the configuration wizard log at
stgw_server_root\logs\configwizard.log
.
-
Stop and then start the deployment manager by running the following commands in sequence:
Note: Stopping the deployment manager requires that you include the WebSphere administrator's
credentials on the command
line.
Linux:
./stopManager.sh -username WAS_admin_username -password WAS_admin_password
./startManager.sh
Windows:
stopManager.bat -username WAS_admin_username -password WAS_admin_password
startManager.bat
- Complete the following steps on every node in the cluster,
including the primary node:
-
Log in to the node's operating system as the WebSphere administrator.
- Navigate to the
profile_root\bin
directory.
-
Start the node agent on the node by running the following command:
Linux:
./startNode.sh
Windows:
startNode.bat
Note: During
installations, the node agent on the primary and secondary nodes might, causing the
startnode
command to result in the following error:
Conflict detected
on port 8878. Likely causes: a) An instance of the server node agent is already running b) some
other process is using port 8878.Confirm the node agent status by running the
command serverstatus nodeagent
from the
profile_root\bin
directory. When prompted, supply the WebSphere
administrator credentials. Verify that the node agent status displays The Node Agent
"nodeagent" is STARTED, and then continue to the next step.
- When all the node agents are started, verify that the cluster
configured properly by performing the following steps:
- Log into the Integrated Solutions Console using your
administrative user ID and password on the deployment manager computer.
- Click , and verify that SametimeGatewayCluster appears
in the table.
- Click SametimeGatewayCluster,
and then in the "Additional properties" section, click Cluster
members to view the cluster members that you created.