How an interactive job starts

When a user signs on to the system, the subsystem gathers information from several system objects before the interactive job is ready.

  1. The subsystem looks in the workstation entry for the job description in order to get the attributes for the interactive job. If the workstation entry specifies *USRPRF for the job description, the job will use the information from the user profile.
    Note: This flexibility allows you to specify whether the job's attributes are tied to the workstation or to the individual user.
  2. After the subsystem knows which job description to use, it might not find all of the job attributes in the job description. Some attributes might be in the user profile. If the user profile does not have the information, the subsystem looks at the system value.
    Note: The user profile contains job attributes that allow you to tailor certain things specifically for that user.
  3. After the subsystem gathers all of the job's attributes, it determines whether a new interactive job can start or if an error message should be posted on the signon screen. The subsystem checks whether the maximum number of jobs allowed by the subsystem or by the workstation entry has been reached. Then it verifies that a valid user profile name has been supplied, that the user profile name is an enabled user profile, and that the supplied password (if required) is valid. Next, it verifies that the user has the proper authorities to the job description, the subsystem description, the workstation device description, and the output queue and library. Finally, the subsystem checks whether the user has reached the limits for allowed signons for that user profile. If any validation errors are encountered, the signon screen displays with an appropriate message. Otherwise, the process of starting the interactive job continues.
  4. After the subsystem validates that the interactive job can start, it checks the job description for the routing data. The subsystem uses the routing data to find a routing entry in the subsystem description. The routing entry provides information about which pool the job will use, which routing program will be used, and from which class the job will get its run time attributes.
  5. When all of these pieces are obtained, the routing program runs. IBM® supplies a routing program called QCMD, which you can use for all types of work. QCMD knows if the job is an interactive job and checks the user profile for an initial program to run. If the initial program finishes running, QCMD displays the initial menu.
Related concepts
Disconnect interactive jobs
I/O error for job requester device
Interactive jobs and routing steps
Programs that control the routing step
Workstation versus user based routing
When jobs end at the same time
Related tasks
Avoid a long-running function from a workstation