12.2.6 Upgrade installation in a cluster configuration

The following describes the procedure for installing a JP1/AJS3 series program as an upgrade installation on a UNIX host in a cluster configuration.

The installation procedure differs according to whether JP1/AJS3 - Manager is updated from JP1/AJS2 - Manager or from JP1/AJS3 - Manager.

In the procedures that follow, the node where the logical host normally operates is called the primary node, and the other node is called the secondary node.

Organization of this subsection

(1) Upgrade procedure for a program other than JP1/AJS2 - Manager
(2) Upgrade procedure (from JP1/AJS2 - Manager to JP1/AJS3 - Manager in a standard configuration)
(3) Upgrade procedure (from JP1/AJS2 - Manager to JP1/AJS3 - Manager in a compatible ISAM configuration)

(1) Upgrade procedure for a program other than JP1/AJS2 - Manager

To perform an upgrade installation from JP1/AJS3 - Manager, JP1/AJS3 - Agent, or JP1/AJS2 - Agent:

Make sure that the logical host is not operating on the secondary node.

Make sure that JP1/AJS3 or JP1/AJS2 of the logical host is not operating on the secondary node.

No problems are caused if JP1/AJS3 or JP1/AJS2 of the logical host is operating on the primary node.
Perform an upgrade installation on the secondary node.
Make sure that the logical host is not operating on the primary node.

Either fail over the logical host from the primary node to the secondary node or stop the logical host to ensure that JP1/AJS3 or JP1/AJS2 of the logical host is not operating on the primary node.

No problems are caused by failing over the logical host to the secondary node on which an upgrade installation has finished.
Perform an upgrade installation on the primary node.
Important note
- When the logical host has been failed over to the secondary node on which an upgrade installation has finished, do not fail back the logical host to the primary node until the upgrade installation on the primary node has finished.
- After upgrading JP1/AJS3 - Manager in the standard configuration, check whether the version of the embedded database is the same as the version indicated in Installation Procedures and System Generation Precautions in the Release Notes. If these versions are different, upgrade the embedded database.

To Page Top

(2) Upgrade procedure (from JP1/AJS2 - Manager to JP1/AJS3 - Manager in a standard configuration)

Important note: An upgrade installation of JP1/AJS2 - Manager in a cluster configuration requires some setup, such as the migration of data on the shared disk, after completion of the upgrade installation. This setup requires that cluster operation be temporarily stopped. You will not be able to start JP1/AJS3 - Manager operation until this setup is complete. Because the time required for the setup to finish depends on the amount of data and other factors, the upgrade installation and setup must be performed according to a suitable plan.

To perform an upgrade installation from JP1/AJS2 - Manager:

Note: If your system consists of multiple logical hosts, perform steps 6 to 16 and 20 to 23 for each logical host. When using a logical host in a non-cluster environment, perform only the steps that relate to the primary node. You cannot begin using JP1/AJS3 until setup is completed on all physical and logical hosts.

Note: The following procedure assumes that one scheduler service is stored in one embedded database. If multiple scheduler services have been set up, perform steps 8, 11, and 12 for all of the scheduler services. To add an embedded database, perform steps 9, 10, 20, and 21.

Terminate JP1 on both the primary and secondary nodes.

Use the cluster software to stop all JP1 programs and services running on the primary and secondary nodes. For details about how to do so, see the documentation for your cluster software.
Install JP1/Base as an upgrade installation on the primary node.

For details about installing JP1/Base, see the Job Management Partner 1/Base User's Guide.
Install JP1/AJS3 - Manager as an upgrade installation on the primary node.

On the primary node, upgrade JP1/AJS2 - Manager to JP1/AJS3 - Manager.
Start the embedded database or instance of HiRDB you were using before the upgrade on the physical primary node host.

To provide access to the database from before the upgrade, start the embedded database or instance of HiRDB that was used as the database of the scheduler service on the physical host of the primary node.

For details about how to start the embedded database or HiRDB, see the manual for the JP1/AJS2 product or HiRDB you are using.
Perform setup on the physical host of the primary node.

On the physical host of the primary node, execute the jajs_migrate command.

For example, execute the command as follows:
```
jajs_migrate -convert
```
For details about the jajs_migrate command, see jajs_migrate in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Confirm that the shared disk and logical IP address are available to the primary node.

Make sure that the shared disk is mounted and the logical IP address is valid so that the data on the shared disk can be migrated. If the shared disk is not mounted or the logical IP address is invalid, use the cluster software or volume manager software to mount the shared disk and enable the logical IP address. For details about how to do so, see the documentation for your cluster software.
Start the embedded database or instance of HiRDB you were using before the upgrade on the logical host of the primary node.

To provide access to the database from before the upgrade, start the embedded database or instance of HiRDB that was used as the database of the scheduler service on the logical host of the primary node.

For details about how to start the embedded database or HiRDB, see the manual for the JP1/AJS2 product or HiRDB you are using.
Execute the ajscnvdbexport command on the primary node to back up the contents of the database.

For example, execute the command as follows:
```
ajscnvdbexport -mh LHOST -F AJSROOT2 -b /tmp/work
```
If you were using the embedded database as the scheduler database before the upgrade, set the following environment variables before you execute the command. If you fail to do so, the message KAVS0990-E is output and the command ends abnormally.
- In HP-UX
  
  SHLIB_PATH=installation-directory-of-embedded-database-before-migration/client/lib
- In Solaris or Linux
  
  LD_LIBRARY_PATH=installation-directory-of-embedded-database-before-migration/client/lib
- In AIX
  
  LIBPATH=installation-directory-of-embedded-database-before-migration/client/lib
For details about the ajscnvdbexport command, see ajscnvdbexport in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajsembdbinstl command on the primary node to install the JP1/AJS3 database.

For example, execute the command as follows:
```
ajsembdbinstl -s /opt/jp1ajs2/tools/AJS3DB -id _JF1 -mh LHOST
```
For details about the ajsembdbinstl command, see ajsembdbinstl in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajsembdbbuild command on the primary node to build the environment for the database.

For example, execute the command as follows:
```
ajsembdbbuild -s -r -d /shdsk/node0/jp1ajs2/embdb/_JF1 
-ld /var/opt/jp1ajs2/embdb/_JF1/dbarea -mh LHOST -eh hostA 
-p 22221 -i /opt/jp1ajs2/embdb/_JF1 -id _JF1
```
For details about the ajsembdbbuild command, see ajsembdbbuild in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajsembdbsetup command on the primary node to set up the environment for the database.

For example, execute the command as follows:
```
ajsembdbsetup -mh LHOST -F AJSROOT2 -p 22221 -id _JF1 -ru s -convert
```
For details about the ajsembdbsetup command, see ajsembdbsetup in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajscnvdbimport command on the primary node to import the contents of the database.

For example, execute the command as follows:
```
ajscnvdbimport -mh LHOST -F AJSROOT2 -b /tmp/work
```
For details about the ajscnvdbimport command, see ajscnvdbimport in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Perform setup on the primary node.

Execute the jajs_migrate command on the primary node to set up the environment for running JP1/AJS3.

For example, execute the command as follows:
```
jajs_migrate -convert -h LHOST -r -S
```
For details about the jajs_migrate command, see jajs_migrate in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Stop the embedded database, which was started automatically during the setup process.

For example, execute the command as follows:
```
ajsembdbstop -id _JF1
```
For details about the ajsembdbstop command, see ajsembdbstop in 2. Commands in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 1.
Execute the jbsgetcnf command on the primary node to back up the common definition information to a file.

Execute the command as follows:
```
jbsgetcnf -h logical-host-name > backup-file-name
```
Copy the file you created in step 15 to the secondary node.
Install JP1/Base as an upgrade installation on the secondary node.

For details about installing JP1/Base, see the Job Management Partner 1/Base User's Guide.
Install JP1/AJS3 - Manager as an upgrade installation on the secondary node.

On the secondary node, upgrade JP1/AJS2 - Manager to JP1/AJS3 - Manager.
Perform setup on the physical host of the secondary node.

Execute the jajs_migrate command on the physical host of the secondary node.

For example, execute the command as follows:
```
jajs_migrate -convert
```
For details about the jajs_migrate command, see jajs_migrate in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajsembdbinstl command on the secondary node to install the JP1/AJS3 database.

For example, execute the command as follows:
```
ajsembdbinstl -s /opt/jp1ajs2/tools/AJS3DB -id _JF1 -mh LHOST
```
For the -id option, specify the same value that you used when installing the database on the primary node.

For details about the ajsembdbinstl command, see ajsembdbinstl in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the ajsembdbbuild command on the secondary node to build the environment for the database.

For example, execute the command as follows:
```
ajsembdbbuild -s -f -d /shdsk/node0/jp1ajs2/embdb/_JF1 
-ld /var/opt/jp1ajs2/embdb/_JF1/dbarea -mh LHOST -eh hostA 
-p 22221 -i /opt/jp1ajs2/embdb/_JF1 -id _JF1
```
For the -id, -d, and -p options, specify the same value that you used when building the environment on the primary node.

For details about the ajsembdbbuild command, see ajsembdbbuild in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Execute the jbssetcnf command on the secondary node to apply the common definition information.

Execute the command as follows:
```
jbssetcnf name-of-file-copied-in-step-16
```
Perform setup on the secondary node.

Execute the jajs_migrate command on the secondary node to set up the environment for running JP1/AJS3.

For example, execute the command as follows:
```
jajs_migrate -convert -h LHOST -f -S
```
For details about the jajs_migrate command, see jajs_migrate in 2. Commands Used during Setup in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 2.
Unregister the embedded database or instance of HiRDB you were using with JP1/AJS2 before the upgrade from the cluster software.

For details about how to do so, see the documentation for your cluster software.
Start JP1 on the primary node.

Use the cluster software to start the JP1 programs and services on the primary node. This starts JP1/AJS3 operation in a cluster environment.

To Page Top

(3) Upgrade procedure (from JP1/AJS2 - Manager to JP1/AJS3 - Manager in a compatible ISAM configuration)

This subsection describes how to upgrade JP1/AJS2 - Manager to JP1/AJS3 - Manager in a compatible ISAM configuration.

If multiple logical hosts have been set up, perform step 5 in the procedure for all of the logical hosts.

To upgrade JP1/AJS2 - Manager:

Terminate JP1 on both the primary and secondary nodes.

Use the cluster software to stop all JP1 programs and services running on the primary and secondary nodes. For details about how to do so, see the documentation for your cluster software.
Install JP1/Base as an upgrade installation on the primary node.

For details about installing JP1/Base, see the Job Management Partner 1/Base User's Guide.
Install JP1/AJS3 - Manager as an upgrade installation on the primary node.

On the primary node, upgrade JP1/AJS2 - Manager to JP1/AJS3 - Manager.

For details about an upgrade installation of a JP1/AJS3 series program, see 12.2.1(2)(b) About upgrade installations of JP1/AJS3 - Manager and JP1/AJS3 - Agent.
Confirm that the shared disk and logical IP address are available to the primary node.

Make sure that the shared disk is available so that the data on the shared disk can be migrated. If the shared disk is not mounted, use the cluster software or volume manager software to mount the shared disk. For details about how to do so, see the documentation for your cluster software.

Also, use the ping or jp1ping command to confirm that the logical IP address is available. For details about the jp1ping command, see the Job Management Partner 1/Base User's Guide.
On the primary node, execute the jajs_convert_to_spmd command to convert the startup process definition file for system management.

For example, execute the command as follows:
```
/opt/jp1ajs2/tools/jajs_convert_to_spmd -h LHOST -c ISAM
```
Install JP1/Base as an upgrade installation on the secondary node.

For details about installing JP1/Base, see the Job Management Partner 1/Base User's Guide.
Install JP1/AJS3 - Manager as an upgrade installation on the secondary node.

On the secondary node, upgrade JP1/AJS2 - Manager to JP1/AJS3 - Manager.

For details about an upgrade installation of a JP1/AJS3 series program, see 12.2.1(2)(b) About upgrade installations of JP1/AJS3 - Manager and JP1/AJS3 - Agent.
Start JP1 on the primary node.

Use the cluster software to start the JP1 programs and services on the primary node. This starts JP1/AJS3 operation in a cluster environment.

To Page Top