Controlling a JES2 MAS system through a VTAM link

Figure 33 shows a z/OS JES2 MAS system connected to the Tivoli Workload Scheduler for z/OS host via a VTAM® link.

System A and the systems in the JES2 MAS complex (System B and System C) are each represented by a computer workstation. The destination field for the workstation on System A is left blank so that work is initiated by the controller on that system. The destination field of the workstation descriptions for the MAS complex contains the VTAM application ID of the tracker on System B. The controller sends work to the tracker on System B via the network communication function. The tracker passes the work to the complex, and the work then processes on either System B or System C, depending on installation parameters.

Figure 33. Controlling a JES2 MAS system through a VTAM link
The graphic shows a JES2 MAS system that is controlled by another system via a VTAM link.

A tracker is started on each system in the configuration. An event-reader subtask in the controller reads events from System A. The event-reader on System B reads the event information from System C and passes the events to NCF for transmission to the controller. This event-reader is required because System C does not have its own link to the controller. The event-writer subtask on System B is started with a reader function—EWSEQNO is defined in the EWTROPTS statement. This means that the event writer passes the events for System B to NCF for transfer to the controller at the same time as they are written to the event data set.

Note:
This figure demonstrates the need for an event reader task where System C does not have a direct link to the controller. But if the required resources are available, try to give each tracker its own link to the controller.

Automatic workload restart can be used in this configuration if the controller cannot communicate with the tracker on System B. The status of the workstation for System B is set to offline if z/OS is stopped or fails, if the tracker is stopped or fails, or if the VTAM link is lost. WLR actions are taken according to the WSOFFLINE keyword of JTOPTS. Workload restart is not affected by failures on System C, because the controller has no direct link with this system.

Table 56 shows the initialization statements you can use to create the configuration in Figure 33.

Table 56. Example EQQPARM members for the preceding figure
EQQPARM members for System A

CONTROLR 

OPCOPTS  OPCHOST(YES)
         ERDRTASK(1)
         ERDRPARM(ERDR1)
         NCFTASK(YES)
         NCFAPPL(NCFAPPL1)
ROUTOPTS SNA(NCFAPPL2)

TRACKERA 

OPCOPTS OPCHOST(NO)
        ERDRTASK(0)
        EWTRTASK(YES)
        EWTRPARM(TRKAEW)
TRROPTS HOSTCON(DASD)

ERDR1 

ERDROPTS ERSEQNO(1)

TRKAEW 

EWTROPTS
EQQPARM members for System B

TRACKERB 

OPCOPTS OPCHOST(NO)
        ERDRTASK(1)
        ERDRPARM(ERDR2)
        EWTRTASK(YES)
        EWTRPARM(TRKBEW)
        NCFTASK(YES)
        NCFAPPL(NCFAPPL2)
TRROPTS HOSTCON(SNA)
        SNAHOST(NCFAPPL1)

TRKBEW 

EWTROPTS EWSEQNO(1)

ERDR2 

ERDROPTS ERSEQNO(2)
EQQPARM members for System C

TRACKERC 

OPCOPTS OPCHOST(NO)
        ERDRTASK(0)
        EWTRTASK(YES)
        EWTRPARM(TRKCEW)
TRROPTS HOSTCON(DASD)

TRKCEW 

EWTROPTS HOLDJOB(NO)

Note:
In this example, the controller has VTAM application ID NCFAPPL1, and the tracker on System B has VTAM application ID NCFAPPL2.