RestartPlus Supervisor

This chapter describes the functions and requirements of the RestartPlus Supervisor.

The RestartPlus Supervisor is a continuously running started task that maintains the RestartPlus Journal. It acquires storage in extended CSA. All other components of RestartPlus that wish to update the Journal communicate the data they wish to add, modify, or delete to the RestartPlus Supervisor by accessing this extended CSA storage.

The RestartPlus Supervisor also uses the service CSVDYNEX to dynamically install its own IEFACTRT SMF exit to accumulate SMF information on JOBs it is tracking for its Journal. The exit module name will be UCRACTnn where nn is a unique suffix for each Supervisor active on the system. The SMF exit also communicates to the Supervisor by using the extended CSA storage.

The RestartPlus Supervisor monitors the CSA it has acquired for requests and processes them as they are detected.

A RestartPlus Supervisor must be active on all systems that wish to use any RestartPlus components.

We recommend that the RestartPlus Supervisor started task be normally started automatically at system IPL time. This ensures that the Supervisor is active so that all RestartPlus components can function properly.

Only one RestartPlus Supervisor job can be active in an MVS image at a time for a specific Journal. To segregate data for different environments on a machine, you can have multiple RestartPlus Supervisors and Journals active on one image. In other words, each RestartPlus Supervisor job can use only one Journal and each Journal can be controlled by only one Supervisor on a given image. If more than one RestartPlus Supervisor job using the same Journal is mistakenly started, a message will be issued and the second job will terminate.

Journals can be shared among MVS images or machines. In this case, the shared Journal must be on a DASD device that is shared among all images that will share the Journal. Each image can then have its own Supervisor that uses the shared Journal. The Supervisors control access to the shared Journal using system enqueues.