Internetwork dependencies overview
Before you specify an internetwork dependency, you must create a workstation definition for the network agent. A network agent is an HCL Workload Automation workstation that handles follows dependencies between its local network and a remote HCL Workload Automation network.
In the local HCL Workload Automation network there can be more than one network agent, each representing a specific HCL Workload Automation remote network where jobs and job streams referring to locally defined internetwork dependencies are defined. Internetwork dependencies are assigned to jobs and job streams in the same way as local follows dependencies, with the exception that the network agent's name is included to identify the followed job or job stream.
A special job stream named EXTERNAL is automatically created by HCL Workload Automation for each network agent in the local network. It contains placeholder jobs to represent each internetwork dependency.
- The script file name, which identifies the remote job or job stream the local job or job stream is dependent on.
- The date the local job stream containing the internetwork dependency is planned to start. If the dependency is defined in a job within the job stream the date the job stream is planned to start is taken into account.
In case of two jobs belonging to different job streams and referring to the same internetwork dependency, as one of their job streams is released and the job starts the internetwork dependency is checked and possibly released. In this case when the second job starts to check its internetwork dependency it finds the dependency already solved but not necessarily on the expected day. If you want to prevent this situation from occurring you must rerun the job representing the internetwork dependency after it is solved the first time.
HCL Workload Automation checks the status of the referred jobs and job streams in the remote network and maps their status in the jobs representing the internetwork dependencies in the EXTERNAL job stream. The status of these jobs and job streams is checked over a fixed time interval until the remote job or job stream reaches the SUCC, CANCL, or ERROR state.