Adding cluster members

When you created the deployment environment, a single cluster was created. To use this cluster to balance the workload among the application servers in the managed nodes, you must create the application servers as members of the cluster.

About this task

To create the cluster members, rerun the BPZCNFGD job that you submitted earlier. This operation will again run the BPMConfig.sh script, this time creating the cluster members by using the settings that are specified for the cluster and its members in the BPMConfig properties file. Typically, one cluster member is added on each federated node.

Procedure

From the deployment manager node, submit job BPZCNFGD to invoke the BPMConfig.sh script.
When the job completes, view the job output to verify that the job completed successfully. Also review the log file for the BPMConfig.sh script, which is written to WAS_HOME/logs/config/BPMConfig_timestamp.log on the deployment manager node.
Synchronize these configuration changes across each of the federated nodes either by using the syncNode.sh command or the administrative console:
- From the UNIX System Services command shell, run the syncNode.sh command on each node:
  cd WAS_HOME/bin
  
  ./syncNode.sh dmgr_host_name dmgr_port
- From the administrative console, click System administration > Nodes. Select all the nodes and then click Full Resynchronize.
Recycle the deployment manager.

What to do next

If required, define additional started class profiles in RACF® for the new cluster members in a network deployment environment:

RDEFINE STARTED W8SR%%%A.* STDATA(USER(W8ACRU) GROUP(W8CFG) TRACE(YES))
RDEFINE STARTED W8SR%%%S.* STDATA(USER(W8ASRU) GROUP(W8CFG) TRACE(YES))

SETROPTS RACLIST(STARTED) GENERIC(STARTED) REFRESH

RDEFINE SERVER CB.*.W8SR*.* UACC(NONE)
PERMIT CB.*.W8SR*.* CLASS(SERVER) ID(W8ACRU) ACC(READ)
PERMIT CB.*.W8SR*.* CLASS(SERVER) ID(W8ASRU) ACC(READ)
SETROPTS RACLIST(SERVER) GENERIC(SERVER) REFRESH