You can add or remove data sources to a multi data source while the data source is deployed (referred to as dynamically changing the data source list in the multi data source).
The data sources that you add to a multi data source must be deployed on the same targets on which you intend to deploy the multi data source. You cannot include data sources in a multi data source that are deployed on different servers or clusters. They must all be deployed on the same deployment targets.
To add or remove a JDBC multi data source:
You can make all data source and multi data source changes in a single edit session. You do not have to activate your changes between steps.
The order in the list determines the order in which the multi data source uses to route connection requests. For multi data sources that use the Failover algorithm, the first data source in the list is considered the primary data source. Others are considered secondary, tertiary, and so forth.