Allocating the VSAM data sets

Perform this task if you are installing a controller.

Table 23 shows the VSAM data sets and their characteristics. The JCL procedure for the controller uses all of these data sets except for EQQLDDS and EQQLTBKP, which are used only in the planning batch jobs. Allocate all these VSAM data sets for a controller.

You can allocate the VSAM data sets by submitting the sample listed in Table 23. Alternatively, you can allocate one or more of the VSAM data sets by running a job like this:

Allocating a VSAM data set
//ALOCVSAM JOB STATEMENT PARAMETERS
//*----------------------------------*
//* ALLOCATE AN OPC VSAM DATA SET *
//*----------------------------------*
//ALLOC   EXEC PGM=IDCAMS,REGION=512K
//SYSPRINT DD  SYSOUT=Q
//EQQVOL1  DD  DISP=OLD,VOL=SER=volser,UNIT=3390
//SYSIN    DD  *
  DEFINE +
    CLUSTER ( +
      NAME('OPC.INST.AD') UNIQUE +
      SPANNED +
      SHR(3) VOL(volser) CYLINDERS(2 2) +
            ) +
    DATA    ( +
      NAME('OPC.INST.ADDATA') +
      KEYS(25 0) RECORDSIZE(1000 132072) +
            ) +
    INDEX   ( +
      NAME('OPC.INST.ADINDEX') +
            )
/*

This example allocates the application description database.

You can allocate VSAM data sets on different device types.

Allocate enough space for your data sets, depending upon the amount of work Tivoli Workload Scheduler for z/OS processes at your installation. You can use Table 24 as a guide to allocate space for VSAM data sets.

Consider the following information when allocating VSAM data sets.

Application description data set (EQQADDS)

The application description data set contains application descriptions and JCL variable tables. This data set is allocated as a spanned data set by EQQPCS01. It has a default maximum record size of 131 072. This allocation limits the variable definitions in a variable table to 275   (131 072/476 = 275), provided there are no variable dependencies. If you also use variable dependencies, the number of variables in a JCL variable table is less than 275.

If you will use a greater number of variable definitions in a variable table, allocate the application description data set with a greater record size. To calculate how great the record size should be, use this method:

  LRECL = 86 + (maximum number of variables in one table * 476) +
  (number of variable dependencies * 88)

where 86 is the length of the header record, 476 is the length of each variable record and 88 is the length of each variable dependency record.

This VSAM data set must be allocated with share option set to 3 SHR(3). Do not use share option 2 or 1.

Current plan data sets (EQQCPnDS)

The current plan VSAM files are opened and closed many times by Tivoli Workload Scheduler for z/OS during normal processing. If Tivoli Workload Scheduler for z/OS is unable to open one of the files, for example if the file is already opened by another job or TSO user, the normal mode manager (NMM), is terminated. The NMM issues message EQQN027E which reports the reason for the unexpected termination. You can issue a MODIFY command to restart the NMM subtask.

It is recommended that you do not access the current plan files from outside the Tivoli Workload Scheduler for z/OS address space. Backups of current plan information should be taken from the new current plan (EQQNCPDS). Shut down the controller address space if full-pack backups are taken of the volumes where the data sets reside.

Data sets for extended data (EQQXDnDS)

The extended data VSAM files are opened and closed by Tivoli Workload Scheduler for z/OS together |with the current plan VSAM files. For this reason, the same considerations |for the current plan data sets apply to the data sets for the extended |data.

Current plan backup copy data set (EQQSCPDS)

During the creation of the current plan, the SCP data set is used as a CP backup copy for the production of the Symphony file and for the integration with IBM Tivoli Monitoring.

This VSAM data set must be allocated with share option set to 3 SHR(3). Do not use share option 2 or 1.

JCL repository data sets (EQQJSnDS)

Take special care when allocating the JCL repository data sets. The following information describes the allocation and use of these data sets.

Tivoli Workload Scheduler for z/OS maintains its own copy of JCL in the JCL repository data set for every job that it submits in the current plan. It uses a primary and alternate data set for the JCL repository, EQQJS1DS and EQQJS2DS. It reorganizes the JCL repository data set that is in use by copying it to the alternate data set and then switching over to use the newly copied data set. The value you specify on the MAXJSFILE keyword defines whether the JCL repository should be automatically copied and determines how often the automatic copy process should occur.

Use the EQQPCS01 sample job created by the EQQJOBS installation aid to allocate the JS data sets. This job allocates the JS data sets with the SPANNED attribute and maximum record size 180 000. This limits the maximum number of JCL statements to 2 249 for any one job. If you run jobs with a greater number of JCL statements, increase the record size. Calculate the required record size, in bytes, by multiplying the number of JCL statements in your largest job by 80, and add an extra 80 bytes for the header record. If you define your JS file without SPANNED, the greatest maximum record size that you can specify is 32 760 bytes. This lets you store a job with up to 408 JCL records. If you define the JS data sets with SPANNED, the maximum record size you can specify is slightly less than a control area (CA). If you use the EQQUX002 exit, the largest job that can be returned by this exit is 7 599 JCL records. Consider this, when you define the maximum record size of the JS data sets.

|

An extended-format |data set for VSAM can be allocated for JS data sets that exceed 4 |GB.

Operator Instruction data set (EQQOIDS)

The operator instruction (OI) database contains operator instructions, each of which corresponds to an operation in the AD database and provides specific instructions about how this operation has to be handled.

This VSAM data set must be allocated with share option set to 3 SHR(3). Do not use share option 2 or 1.