Installing the Database Synchronization Component

Installation Options › Database Synchronization Component › Installing the Database Synchronization Component

Installing the Database Synchronization Component

To install the Database Synchronization Component (DSC), you need the Database Synchronization service machine. The name of this machine can be any name you choose. Specify the name in the DBSYNC VMO record. If not specified in the DBSYNC VMO record, the default is ACFSYNC. This guide uses ACFSYNC as the name of the Database Synchronization Component service machine in all examples.

To install the Database Synchronization Component:

Add the ACFSYNC service machine to the VM directory. You can use the supplied file ACFSYNC DIRECT as a sample directory. Edit this file and make any changes necessary.
The three MDISK statements required for the ACFSYNC machine are commented out in the supplied ACFSYNC DIRECT. You must change these MDISK statements before you add ACFSYNC to the VM directory or you must add them just after you add ACFSYNC to the VM directory, depending on your directory maintenance software and site standards and procedures.

The three minidisks are:
- 100-ACF2DSC PARMS file and the ACFSYNC nucleus
- 191-PROFILE EXEC for the ACFSYNC machine
- 600-ACFSYNC recovery file.
Add the CA ACF2 for z/VM logonid record for ACFSYNC and give it the ACCOUNT, SECURITY, and AUTONOPW privileges. Also add any other changes your site standards and procedures require.
Modify the DBSYNC VMO record to indicate which databases you want to synchronize, and the name of the ACFSYNC and CCIVM service machines. See the Administrator Guide for information about changing the DBSYNC VMO record.
If you want to allow access to the ACFSYNC minidisks, write an access rule for ACFSYNC now.
Logon to the ACFSYNC service machine and format the minidisks as follows:
Use the following command to format the ACFSYNC 100 disk as a standard CMS minidisk to prepare it for the ACFSYNC nucleus:
```
FORMAT 100 Z
```
The above command prompts to make sure the format is appropriate. It also prompts for the DISK label. The disk label can be any name (for example, DSC100).
Issue the following command to reserve the first cylinder of the two‑cylinder 100 disk as available for use by CMS files.
```
FORMAT 100 Z 1 (RECOMP
```
This command also leaves the last cylinder of the disk for the ACFSYNC nucleus.
Format the ACFSYNC 191 disk as a standard CMS minidisk with the following command:
```
FORMAT 191 A
```
This command prompts you to make sure the format is correct. It also prompts for the DISK label. You can name the label any appropriate name (for example, DSC191).
Link to the CAIMAINT minidisk where you installed CA ACF2 for z/VM, normally the CAIMAINT 291 disk. Access this disk as the B‑disk.
```
LINK CAIMAINT 291 291 RR
```
```
ACCESS 291 B
```
Create the PROFILE EXEC for the ACFSYNC machine with the following command:
```
COPYFILE ACFSYNC PROFILE B PROFILE EXEC A
```
You can edit the new PROFILE EXEC file on the A‑disk. The only change that is required is if your site uses a disk other than the CAIMAINT 291 disk. This profile IPLs the nucleus on the ACFSYNC 100 disk if the ACFSYNC machine is running disconnected. If the ACFSYNC machine is logged on normally, it assumes maintenance is being done and stays running CMS and accesses the CAIMAINT 291 disk.
Create the ACF2DSC PARMS file on the ACFSYNC 100 disk with the following command:
```
ACCESS 100 Z
```
```
COPYFILE ACF2DSC PARMSAMP B ACF2DSC PARMS Z
```
Edit the ACF2DSC PARMS file and make any local modifications that are needed. Refer to the section titled Parameter Files for more information.
Prepare the ACFSYNC 600 disk for use as the ACFSYNC recovery file. This file saves unprocessed database synchronization transactions until they can be processed. Issue the following commands:
```
DSCCATDK 600 DSCFIL
```
```
DSCMAI DSC
```
Note: The 600 disk must have a VOLSER of DSCFIL, and the DSCCATDK command makes it DSCFIL. DSCMAI uses the name from the DSN= statement in the DSC MAIDATA file as the data set name to use when initializing the recovery file. If you use a name other than the default of CAI.ACF2.DSC.RECOVERY.FILE, you must modify the DSC MAIDATA file (on the CAIMAINT 291 disk) before running the above commands.

If you are using the RULELONG feature to use CA ACF2 for z/VM rules over 4 Kb in length, you must modify the BLOCKSIZE=5120 statement in the DSC MAIDATA file. To modify the BLOCKSIZE, specify the VSAM CISIZE+256, up to a maximum of 32512. For example, if you are using a CISIZE of 8192 (8K), change the statement to read BLOCKSIZE=8448. The BLOCKSIZE= cannot be set to less than 5120 or more than 32512.

The DSCGEN EXEC creates the Database Synchronization Component nucleus. This nucleus is created by a process similar to the CP or CMS nucleus. It punches a standard load deck back to the virtual reader. The DSGEN EXEC IPLs 00C to create the actual nucleus, assuming the load deck was created WITH RC=0.

To generate the nucleus, issue the following commands:

ACCESS 291 B
DSCGEN

Make sure that, when the DSCGEN finishes, it finishes with a disabled wait PSW with the second word as all zeroes. Also, there must not be any other error messages issued, such as unresolved references or text files not found, even if the PSW has a second word of zeroes.

You must IPL CMS (or log off and log back on) after the DSCGEN to get out of the disabled wait. Then, read in the LOADMAP with the following command:

DSCRDMAP