The agent return code retrieval

The IBM i programming model was originally based on an early object orientation model in which programs communicated through message passing, rather than using return codes. The introduction of the Integrated Language Programming (ILE) model lead to the definitions of common areas to exchange data as return codes in the same job environment: the user return codes and the system end codes.

For information about user return codes, see Controlling the job environment with the user return code.

When the IBM i agent verifies that a submitted command or job is completed, it assigns a return code to the job based on the job status of the completed job. The return code is set depending on the completion message of the command or job. If the command or job completes successfully, the return code is set to 0. If the command or job does not complete successfully, the return code is set to the value of the severity of the message related to the exception that caused the abnormal end of the job. The IBM i agent can also set the return code to the value of the user return code when it is returned by the submitted command. If retrieved, the user return code is used as the value to set the return code.

The return code value assigned to the job is included in the IBM i agent joblog for the job and sent back to the scheduler user interface (WEB UI or z/OS ISPF panels) as return code, for compatibility reasons with agents on other operating systems.