Creating the Target Group
In a clustered environment, the only nodes from which you can select servers for the target group are those that are part of the primary configured port. To ensure that the servers are available for the target group, configure all servers in the target group to use the remote alias of the primary port.
Perform the following procedure to create a target group for a clustered environment.
To create a target group in a clustered environment
1. In Deployer, perform one of the following:
For... | Perform the following... |
Integration Server or Trading Networks | Go to Target Groups > IS & TN and click Create IS & TN Target Groups. |
process models | Go to Create BPM(ProcessModel) Target Group and click Create BPM(ProcessModel) Target Group. Note:
You should not use concurrent deployment when deploying BPM process models to BPM target groups. |
2. In the Name box, type the name to use for the target group. The name can be up to 32 characters long and cannot contain spaces or the following illegal characters:
$ ~ / \ # & @ ^ ! % * : ; , + = > < ‘ ’ "
3. In the Description box, type a description for the target group. The description length has no limit and can include any characters.
4. In the Version box, enter the version of the target group.
5. Click Create.
6. In the left-hand pane, click the name of the target group from the Group Name column.
7. On the Configure Target Group pane, set Roll Back All on Failure to Yes.
Note:
Roll Back All on Failure is valid for runtime-based deployment only. Deployer ignores this setting for repository-based deployment.
8. The Available Servers list shows the cluster name as a top-level node in the tree, and then all the servers in that cluster as child nodes under the cluster name. Select the cluster name node, and then click Add. The entire cluster tree moves to the Selected Servers list. Click Save.
Note:
If you select individual nodes of the cluster and not the entire cluster, when you deploy, the nodes in the cluster will no longer be identical. Tasks will not run equally well on all servers in the cluster, which could cause errors and failures.
If a server in the cluster is... | Then... |
Defined as a remote server alias | The child node for that server shows the remote server alias, and you can select it. |
Not defined as a remote server | The child node shows the server host and port, but you cannot select it. |
Not running | That server does not appear in the list at all. Ensure that every server in the cluster is defined as a remote server and is up and running. |
9. You can add other clusters or individual servers to the target group.
10. Map the project to the target group.
11. Checkpoint, deploy, and, if necessary, roll back the project as you would in an unclustered environment.