8.3 Starting and Stopping Daemons in the HACMP System

This section describes how to start and stop Systemwalker Operation Manager in the HACMP system.

Starting daemons

This section describes the procedure for starting the Systemwalker Operation Manager daemons.

Start the Systemwalker Operation Manager daemons that are not registered with the HACMP system by executing the soperationmgr command on all the servers on both the active and standby nodes. In the case of the 1:1 active/standby configuration (with subsystems and partial cluster operation), the JobScheduler daemons and Job Execution Control daemons of subsystems that are not registered in cluster systems will also be started up.
```
# /opt/systemwalker/bin/soperationmgr
```
Start cluster services if they are not started. If they are already started, go to step 3 to start the resource group.
[Starting cluster services]
1. Enter "smitty hacmp" to open a SMIT session.
```
#smitty hacmp
```
  Refer to the HACMP manual for details on SMIT.
2. Select System Management (C-SPOC)] >> Manage HACMP Services >> Start Cluster Services.
3. In the Start Cluster Services window, enter the field values for nodes on which cluster services are started.
Bring the resource groups registered with the cluster system online to start the Systemwalker Operation Manager daemons registered with the cluster system.
[Starting resource groups]
1. Enter "smitty hacmp" to open a SMIT session.
```
#smitty hacmp
```
  Refer to the HACMP manual for details on SMIT.
2. Select System Management (C-SPOC) >> HACMP Resource Group and Application Management >> Bring a Resource Group Online.
3. In the window that displays the list of resource groups, select the resource group of Systemwalker Operation Manager.
4. Select the node in the Select an Online Node window.

Stopping daemons

This section describes the procedure for stopping the Systemwalker Operation Manager daemons.

Bring the resource groups registered with the cluster system off line to stop the Systemwalker Operation Manager daemons registered with the cluster system.
1. Enter "smitty hacmp" to open a SMIT session.
```
#smitty hacmp
```
  Refer to the HACMP manual for details on SMIT.
2. Select System Management (C-SPOC) >> HACMP Resource Group and Application Management >> Bring a Resource Group Offline.
3. In the window that displays the list of resource groups, select the resource group of Systemwalker Operation Manager.
4. Select a node in the Select an Online Node window.
Stop the Systemwalker Operation Manager daemons that are not registered with HACMP by executing the poperationmgr command. In the case of the 1:1 active/standby configuration (with subsystems and partial cluster operation), the JobScheduler daemons and Job Execution Control daemons of subsystems that are not registered in cluster systems will also be stopped.

Note

Do not move the current directory to the shared disk when executing the soperationmgr or poperationmgr commands. Otherwise, a failover may not be performed correctly.
Do not start or stop any services using the Service control section in the Systemwalker Operation Manager Environment Setup window.
Any resource groups where the Systemwalker Operation Manager application has been registered must be stopped before Systemwalker Operation Manager is stopped (if Systemwalker Operation Manager has been registered with a cluster system).
As a result, the shared disk and logical IP address registered with any such resource group will also stop, along with the Operation Manager application and any applications (such as Oracle or Interstage) registered with it.