Event Replicator Installation Steps for z/OS Systems

Be sure your system meets the requirements described in System Requirements.

To install Event Replicator for Adabas on z/OS systems, complete the steps described in this document.


Step 1. Install the Replication Load Modules

Install the replication load modules by concatenating the delivered load library in sequence or copying the modules to the library used for running the Adabas nucleus and utilities. The Replication load modules can be found in ARFvrs.LOAD.

Step 2. Apply Necessary Zaps

Apply any necessary zaps, as described in the $README file on the ARFvrs.MVSZAPS library and in any subsequent ARFvrs.MVSZnnn data sets (if they have been provided).

Step 3. (Optional) Create the ADAMQS Load Module for WebSphere MQ Support

If WebSphere MQ will be used, execute a job for linking ADAMQS. To create the required Event Replicator Server load module called ADAMQS, use JCL similar to this:

//LINK     EXEC PGM=IEWL,PARM='LIST,XREF,REUS,RENT'
//SYSPRINT DD SYSOUT=*
//SYSUT1   DD UNIT=SYSDA,SPACE=(CYL,(3,1))
//ADALIB   DD DISP=SHR,DSN="ADABAS-Event-Replicator-Load-Library"
//MQSLIB   DD DISP=SHR,DSN="MQ-Series-SCSQLOAD-Library"
//SYSLMOD  DD DISP=SHR,DSN="User-Load-Library"
//SYSLIN   DD *
 INCLUDE ADALIB(ADAMQT)
 INCLUDE MQSLIB(CSQBSTUB)
 ENTRY ADAMQS
 NAME  ADAMQS(R)
/*

Note:
If you are using IBM WebSphere MQ Series definitions for your DESTINATION or IQUEUE definitions, a S0D3 abend can occur if you run it as a started task and specify the parameter REUSASID=YES. This is a documented IBM WebSphere MQ Series issue (http://www-01.ibm.com/support/docview.wss?uid=swg21410392).

Step 4. Create the Event Replicator Server

Create a database for use as your Event Replicator Server. Read the ADADEF documentation for the new parameters used when defining an Event Replicator Server.

If you want to perform transaction logging or use TLOG destination definitions, be sure to create and activate a CLOG (command log). For more information, refer to your Adabas DBA Tasks documentation.

Note:
User application files should not be loaded on the Event Replicator Server.

Step 5. (Optional) Load a Replicator System File

To use Event Replicator for Adabas and customize its processing, you must supply various replication definitions. You can maintain these definitions in one of two ways:

  • You can specify Event Replicator initialization parameters for the replication definitions in DDKARTE in the Event Replicator Server startup job. For more information about the initialization parameters you can specify in the Event Replicator Server startup job, read Event Replicator Server Initialization Parameters. This method of defining replication definitions does not require the use of a Replicator system file because the definitions are read from the initialization parameters in the startup job.

  • You can maintain your replication definitions in the Replicator system file using the Adabas Event Replicator Subsystem. The Adabas Event Replicator Subsystem is an online interface that must be loaded into Natural before you can use it. Access to the Adabas Event Replicator Subsystem is then available through Natural or from Adabas Online System. For more information, read Using the Adabas Event Replicator Subsystem.

If you determine that you want to maintain your replication definitions in the Replicator system file, you must load the file on the Event Replicator Server.

Start of instruction setTo load a Replicator system file on the Event Replicator Server:

  1. Use the ADALOD utility. A sample ADALOD utility job is provided in member ADALODRP of the ARFvrs.JOBS data set. The only ADALOD utility parameter you should specify in this member is the REPLICATOR parameter. The others are not valid when loading the Replicator system file. See the ADALOD documentation for more information.

  2. Do either of the following in Natural to identify the physical database file to be associated with the system file:

    • Run a Natural NTLFILE macro for the file. For complete information, refer to your Natural LFILE parameter documentation, found in the Natural for Mainframe documentation on Software AG's Empower web site.

    • When you start the Adabas Event Replicator Subsystem, be sure to set the Natural LFILE parameter as described in Accessing the Adabas Event Replicator Subsystem. The LFILE parameter can be specified either as a dynamic parameter or inside a Natural SYSPARM profile.

Step 6. (Optional) Load an SLOG System File

If you intend to use the subscription logging (SLOG) facility, you will need an SLOG system file, which is an Adabas system file on the Event Replicator Server. To set this up, please read Setting Up Subscription Logging. For complete information about the SLOG facility, read Using the Subscription Logging Facility.

Step 7. (Optional) Load the Adabas Event Replicator Subsystem Application Into Natural

If you intend to use the Adabas Event Replicator Subsystem to maintain replication definitions in the Replicator system file (see Step 4), it must be loaded into Natural. To do this, use Natural's INPL utility to load the ARFvrs.INPL data set into the Natural system file libraries.

Step 8. Set Up the Event Replicator Server Startup JCL

Set up the startup JCL and parameters for the Event Replicator Server job. Use the JCL for an existing Adabas nucleus as a starting point for creating the Event Replicator Server JCL. For complete information on the ADARUN parameters pertinent to the Event Replicator Server, read Pertinent ADARUN Parameters .

  1. Specify an initial setting for the ADARUN parameter RPLPARMS to "NONE". The RPLPARMS parameter identifies where the replication definitions (initialization parameters) should be read from. A setting of "NONE" allows you to run the Event Replicator Server when you have no definitions set up, providing you an opportunity to specify them.

  2. Set ADARUN parameter LRPL to the size of the Event Replicator Server replication buffer. For performance reasons, Software AG recommends setting LRPL to a relatively large value (e.g. LRPL=40M).

  3. Set ADARUN parameter LU to 167,000 or greater.

  4. Set ADARUN parameter NAB to a value greater than or equal to:

    41 * 10 * the-number-of-Adabas-nuclei-sending-data-to-the-Event-Replicator-Server

    For example, if one Adabas nucleus will be sending data to the Event Replicator Server, set the NAB parameter greater than or equal to 410 (for example NAB=420).

  5. Set ADARUN parameter NT to a value greater than or equal to 15.

Step 9. Start the Event Replicator Server Job

Run the Event Replicator Server job you set up in the previous step.

Note:
If you are using IBM WebSphere MQ Series definitions for your DESTINATION or IQUEUE definitions, a S0D3 abend can occur if you run it as a started task and specify the parameter REUSASID=YES. This is a documented IBM WebSphere MQ Series issue (http://www-01.ibm.com/support/docview.wss?uid=swg21410392).

Step 10. Implement Replication

The previous steps in this section allowed you to install Event Replicator for Adabas, but they did not implement or start replication. How replication is implemented varies by site, but general implementation steps are described in Post-Installation Replication Implementation Steps (All Platforms).