18.7.4 Starting and stopping the mutual node switching system for maintenance

This subsection describes how to start and stop the mutual node switching system for maintenance.

The stopping procedure will be described for the following three cases:

When maintenance that does not require restart
When maintenance that requires restart (method of stopping both the nodes at the same time)
When maintenance that requires restart (method of not stopping both the nodes at the same time)

Note that this is the procedure for the case where both the virtual hosts of the executing nodes are already running.

Organization of this subsection

(1) When maintenance that does not require restart
(2) When maintenance that requires restart (method of stopping both the nodes at the same time)
(3) When maintenance that requires restart (method of not stopping both the nodes at the same time)

(1) When maintenance that does not require restart

To start and stop the system for maintenance that does not require system restart:

In the virtual host of the executing node (virtual host of the active node), stop the J2EE applications and resource adapters that are running.

When the batch server, confirm that the batch applications are not running and then stop the resource adapter.

For stopping the J2EE applications and resource adapters, see the following manuals:
- In a system executing J2EE applications
  
  See 4.1.3 Procedure for stopping a system and 4.1.4 Methods of stopping a system in the uCosminexus Application Server Management Portal User Guide.
- In a system executing batch applications
  
  See 6.1.3 Procedure for stopping a system and 6.1.4 Methods of stopping a system in the uCosminexus Application Server Management Portal User Guide.
Perform maintenance using server management commands.

For operations of the server management commands, see 3. Basic Operations of Server Management Commands in the uCosminexus Application Server Application Setup Guide.
Start the J2EE applications and resource adapters that were stopped in step 1.

For the J2EE server, the J2EE applications and resource adapters will start. The resource adapters will start even for the batch server.

For starting the J2EE applications and resource adapters, see the following manuals:
- In a system executing J2EE applications
  
  See 4.1.1 Procedure for starting a system and 4.1.2 Methods of starting a system in the uCosminexus Application Server Management Portal User Guide.
- In a system executing batch applications
  
  See 6.1.1 Procedure for starting a system and 6.1.2 Methods of starting a system in the uCosminexus Application Server Management Portal User Guide.
Check the operations of the J2EE applications and resource adapters as needed.
For the virtual host of the executing node (virtual host of the active node), run the monswap command to switch the nodes.
```
# monswap server-identification-name
```
In the underlined part, specify the server identification name of the virtual host of the executing node specified in the operand 'alias' of the servers file. The servers file exists in application server where the virtual host of the executing node is deployed.

The virtual host of the executing node (virtual host of the active node) switches to the virtual host of the standby node and the virtual host that was the standby node (virtual host of the spare node) becomes the virtual host of the executing node.
Execute steps 1 to 4 in the virtual host of the executing node (virtual host of the spare node) after the nodes are switched.
As needed, run the monswap command again in the executing node host (virtual host of the spare node) after the nodes are switched.
```
# monswap server-identification-name
```
In the underlined part, specify the same server identification name as that specified in the monswap command argument in step 5.

Execute this command to return the virtual host of the active node to the executing node.

Thus, the procedure for starting and stopping the system for maintenance is complete.

To Page Top

(2) When maintenance that requires restart (method of stopping both the nodes at the same time)

To start and stop the system for maintenance that requires system restart:

Run the monend command for the virtual host of the executing node to stop the system.
```
# monend server-identification-name
```
In the underlined part, specify the server identification name of the virtual host of the executing node specified in the operand 'alias' of the servers file in application server where the virtual host of the executing node is deployed.

The virtual host of the executing node will stop. Also, stop instruction is automatically sent to the standby node by the HA monitor and the virtual host of the standby node also stops.
Execute maintenance operations such as changing the definition files for the virtual hosts of both the active node and spare node.
Run the monbegin command for the virtual host of the active node.
```
# monbegin server-identification-name
```
In the underlined part, specify the same server identification name as that specified in the monend command argument in step 1.

As a result, the virtual host of the active node will start as the executing node.
Run the monbegin command for the virtual host of the spare node.
```
# monbegin server-identification-name
```
In the underlined part, specify the same server identification name as that specified in the monend command argument in step 1.

As a result, the virtual host of the spare node will start as the standby node.
As needed, check the operations related to changes in definitions in the virtual host of the executing node.

Reference note: For details on the servers file definition (server environment settings in the HA monitor), see 18.5.6 Server-compliant environment settings.

To Page Top

(3) When maintenance that requires restart (method of not stopping both the nodes at the same time)

To start and stop the system for maintenance that requires system restart:

Execute maintenance operations such as changing the definition files for the virtual host of the standby node.
Run the monswap command for the virtual host of the executing node to switch the nodes.
```
# monswap server-identification-name
```
In the underlined part, specify the server identification name of the virtual host of the executing node specified in the operand 'alias' of the servers file in Application Server where the virtual host of the executing node is deployed.

The virtual host of the executing node will be stopped, and then the standby node will start. As a result, the virtual host for which maintenance is complete becomes the virtual host of the executing node.
As needed, check the operations related to the changes in definitions for the executing node host (virtual host of the spare node) after the nodes are switched.
Execute maintenance operations such as editing the definition file in the standby node host (virtual host of the active node) after the nodes are switched.
As needed, run the monswap command again for the virtual host of the executing node (spare node host) after the nodes are switched.
```
# monswap server-identification-name
```
In the underlined part, specify the same server identification name as that specified in the monswap command argument in step 2.

Execute this command to return the virtual host of the active node to the executing node.
As needed, check the operations related to changes in definitions in the virtual host of the executing node (virtual host of the active node) after the nodes are switched.

Thus, the procedure for starting and stopping the system for maintenance is complete.

To Page Top