Summary instructions for this step

In this step, do the following to create a new database (all jobs are described in more detail below):

1. Run the batch job B97DBDEF.

   This job allocates and formats the Adabas Audit Data Retrieval definition database and uploads the data dictionary.

2. Run the batch job B97DBFOR.

   This job updates the database definition file with values you provided when running the installation REXX, allocates and formats the Adabas Audit Data Retrieval VSAM databases, and loads customized data.

Required (at the latest before running B97DBFOR):

• The Beta SVC must be installed.
• The security environment must have been initialized (B97INIT).
• A valid license file must be available.

B97DBDEF

Job B97DBDEF has one step and does the following:

1. It allocates the VSAM cluster BETA97.DB.DEF, which serves as database definition file (B97DEF).
2. It loads the Adabas Audit Data Retrieval data dictionary as provided with the installation files into BETA97.DB.DEF.

   The loaded data provides the structure information (databases, tables, keys, and fields) on the entire Adabas Audit Data Retrieval database.

B97DBDEF must run to completion with RC=0.

B97DBFOR

Job B97DBFOR requires the Beta SVC and must run to completion with RC=0.

B97DBFOR has three steps:

1. Step FILLDEF updates the database definition file with customized data from member S#97UDEF, which is tailored at the end of the installation REXX.
2. Step FORMDB allocates and formats the other Adabas Audit Data Retrieval VSAM datasets according to the specifications (size, volumes, etc.) made during the installation REXX.
3. Step INSERT uploads data into the Adabas Audit Data Retrieval database (subsystem options record from tailored member S#97ISYS and more data from other tailored members).

S#97UDEF: Keywords and syntax

The BSA program BST05UPF updates the database definition file with customized data from member S#97UDEF.

S#97UDEF contains one definition statement for each Adabas Audit Data Retrieval database component. Continuation signs ( - ) are used if the definition statement spans several lines.

This is the syntax of the definition statement:

DEFINE UPDATE FILE logicalfilename -
DSNAME datasetname -
VOLUME volser -
UNIT unit -
BUFFER nn -
SPACE no_of_cyl -
HWM high_water_mark_in_percent

Use the following syntax for SMS-managed databases:

DEFINE UPDATE FILE logicalfilename -
DSNAME datasetname -
STORAGECLASS storclas -
MANAGEMENTCLASS mgmtclas -
DATACLASS dataclas -
BUFFER nn -
SPACE no_of_cyl -
HWM high_water_mark_in_percent

For more information on BST05UPF and its parameters, see "BST05UPF: Updating file information in the database definition file" in BSA Installation and System Guide.

If B97DBFOR ends with RC=16

If there is a problem when defining the Adabas Audit Data Retrieval database, for example, not enough space, job B97DBFOR ends with RC=16. In this case, find out what caused the problem and proceed as follows:

1. Run job B97DBDEL to delete any parts of the Adabas Audit Data Retrieval database which have been defined during the previous run of B97DBDEF and B97DBFOR.
2. Rerun job B97DBDEF to allocate the VSAM cluster BETA97.DB.DEF and to load the database definition file.
3. If necessary, correct the entries in member S#97UDEF, for example, change the volume name, SMS class, or amount of space.
4. Rerun job B97DBFOR to upload the modified definitions of member S#97UDEF into BETA97.DB.DEF (B97DEF) and to format and initialize the other Adabas Audit Data Retrieval VSAM datasets.

Adabas Audit Data Retrieval database components

The following table shows which datasets are created for the Adabas Audit Data Retrieval database. It describes their purpose and shows default dataset names and size.

Dataset name	Short name	3390 space	Purpose
BETA97.DB.DEF	B97DEF	5 cyl	Database definition
BETA97.DB.MAIN BETA97.DB.MAIN.KEY	B97MAIN B97KEY	20 cyl 10 cyl	Main database containing definitions
BETA97.DB.LIST BETA97.DB.LIST.KEY	B97LIST B97LKEY	100 cyl 50 cyl	List index database
BETA97.DB.MSG BETA97.DB.MSG.KEY	B97MSG B97MKEY	20 cyl 10 cyl	Message log database
BETA97.DB.NOTES BETA97.DB.NOTES.KEY	B97NOTES B97NKEY	10 cyl 5 cyl	Notes database
BETA97.DB.ARC BETA97.DB.ARC.KEY	ARCDATA ARCKEY	60 cyl 25 cyl	Archive database
BETA97.DB.SPOOLn	SPnnnnnn	100 cyl	Lists
BETA97.DB.CACHEn	CACHEn	100 cyl	Reloaded 4-MB blocks of list data
BETA97.DB.INDEXn	INDEXn	100 cyl	Indexes
BETA97.DB.GLOBLn	GLOBLn	100 cyl	Global indexes
BETA97.DB.RELODn	RELODn	100 cyl	Reloaded lists
BETA97.DB.SFR BETA97.DB.SFR.KEY	B97SFR B97SFRK	10 cyl 3 cyl	Spool file recovery information
BETA97.DB.LGF BETA97.DB.LGF.KEY	LGFDATA LGFKEY	20 cyl 10 cyl	Information used with _beta smf
BETA97.DB.LOG	LOGFILE	20 cyl	Temporary processing information
BETA97.DB.SYNC	SYNCFILE	1 cyl	Synchronization file

Notes on databases

• There is no need to save the SYNC database.
• If no changes have been made during installation, the 3390 space values are used as defaults.
• A standard VSAM dataset can have a total maximum size of 4 GB and an extended VSAM dataset can have a total maximum size of 28 GB.
• Job B97DBFOR allocates one spool file of each type (SPOOL, CACHE, INDEX, etc.). Define additional spool file models for each type to enable Adabas Audit Data Retrieval to allocate and format additional spool files on the fly when available spool files are full. For additional information on spool files, see "Spool files" in Adabas Audit Data Retrieval Administrator Guide.
• For the best performance, we recommend the following for CIsize:
  • The CIsize of the LOG database component should be 24K on type 3390 DASD units.
  • The CIsize of all other database components should be 4K.

  If you use different values, you should ensure that spool files of one type (SPOOL, INDEX, CACHE, or GLOBAL) have the same CIsize.

Calculating required space for databases

The size of a database depends on the maximum number of records that are to be stored in this database. To calculate the required space for a database, you need to know the structure of the Adabas Audit Data Retrieval database, for example:

• The tables contained in a database
• The length of the records of a table
• The number of related records created for each entry in a table

Because the structure of the Adabas Audit Data Retrieval database is similar to the structure of the _beta doc|z database, refer to the _beta doc|z Installation and System Guide for information on calculating the required space for a database.

Maximum number of open spool files

The maximum number of spool files (cache, spool, and index) that can be opened by Adabas Audit Data Retrieval is 32000.

Important: The operating system may have a lower limit for the maximum number of open files. For more information, see the description of the task I/O table (TIOT) in the IBM publication MVS Initialization and Tuning Reference, chapter "ALLOCxx (allocation system defaults)".