• No results found

Configuring the Agent Plug-in

Chapter 4: Configuring the Agent Plug-in

This section contains the following topics:

How to Configure Agent Parameters (see page 33)

CA WA Agent for Informatica Parameters in the agentparm.txt File (see page 34)

How to Set Up an Alias for the Agent Plug-in (see page 37)

How to Configure Agent Parameters

You configure agent parameters by editing the agentparm.txt file, located in the agent installation directory. When you install the agent, the installation program adds

frequently-configured agent parameters to the file. Other agent parameters exist, which you must manually add to the agentparm.txt file to configure the agent. For any

configuration changes to take effect, always stop and restart the agent. For some agent parameters, such as the agent name and communication parameters, also configure the parameters on the scheduling manager.

To configure agent parameters, do the following: 1. Configure agent parameters on the agent.

2. Configure agent parameters on the scheduling manager (see page 34).

Configure Agent Parameters on the Agent

Use the following procedure to configure agent parameters on CA WA Agent for UNIX, Linux, or Windows.

Follow these steps:

1. Change to the agent installation directory.

2. Stop the agent. At the command prompt, enter the following command:

■ On Linux:

./cybAgent -s

■ On Windows:

cybAgent -s

34 Implementation Guide

3. Open the agentparm.txt file located in the agent installation directory. 4. Edit the parameters to make the required changes.

5. Save and close the agentparm.txt file.

6. Start the agent. At the command prompt, enter the following command:

■ On Linux:

./cybAgent &

■ On Windows:

cybAgent -a

The agent starts and the parameters are configured.

Configure Agent Parameters on the Scheduling Manager

When you change an agent parameter in the agentparm.txt file that is also defined on the scheduling manager, such as the agent name, configure the agent parameter on the scheduling manager.

Note: For detailed instructions to configure agent parameters on the scheduling manager, see the documentation for your scheduling manager.

CA WA Agent for Informatica Parameters in the agentparm.txt

File

The agent plug-in installation program adds the parameters listed below to the agentparm.txt file. The file is located in the agent installation directory. You can open the agentparm.txt file in any standard text editor.

Note: You must save the file after making edits to persist the changes. communication.alias_n

Defines the alias name for the agent. The n suffix increments sequentially for each alias agent.

Note: To enable an alias on the agent, verify that the comment character (#) is removed from the parameter line.

CA WA Agent for Informatica Parameters in the agentparm.txt File

Chapter 4: Configuring the Agent Plug-in 35 informatica.appendRunLog

Specifies whether the workflow or the task logs of the current execution is captured and stored in the spool file for the startWorkflow, restartWorkflow, and

monitorWorkflow CLI operations. The following values are available: ALWAYS

Indicates that the workflow or the task log is always stored in the spool file. FAILURE

Indicates that the workflow or the task log is stored in the spool file only when the job fails.

NONE

Indicates that no logs are stored in the spool file for a workflow run. Default: FAILURE

Note: The logs of a failed task are only added if the failed task is of the SESSION type.

informatica.concurrentinstances.enabled

Indicates whether to enable the execution of concurrent workflows with the same instance name.

true

Enables the execution of concurrent workflows with the same instance name. Note: Due to a known issue in Informatica, this option is not supported so you cannot run workflow instances in parallel.

false

Disables the execution of concurrent workflows with the same instance name. Default: false

informatica.connection.pool.size

Specifies the maximum number of simultaneously open connections in a database pool.

36 Implementation Guide

informatica.controltask.enabled

Indicates whether to retrieve the run details of CONTROL tasks within a workflow using the Informatica API or a database query.

true

Retrieves the run details of CONTROL tasks within a workflow using the Informatica API.

Note: Due to a known issue in Informatica, this option is not supported. false

Retrieves the run details of CONTROL tasks within a workflow using a database query.

Default: false

informatica.monitor.workflow.sleep.time.seconds

Specifies the job monitor polling interval for outstanding jobs in seconds. Default: 20

informatica.passonsuccessonly.default

Indicates whether the status of the workflow is retrieved directly from Informatica or inferred depending on the successful execution of all the tasks.

true

Returns the workflow status based on the status of tasks in the Informatica workflow. If all tasks in the workflow complete successfully, the agent reports the status of the workflow as successful. If any task in the workflow fails, the agent reports the status of the workflow as failed.

false

Returns the workflow status as reported by Informatica. In Informatica, a workflow can succeed even if one or more tasks in the workflow fail. Default: false

Note: You can override the value in the job definition. informatica.server.target.default

Specifies the name of the Informatica target (infaTarget). The

install_dir/config/informatica/infaTarget directory contains the properties for

connecting to the Informatica server and one or more database repositories. Note: You can identify the Informatica target server in a job definition to override the default value.

How to Set Up an Alias for the Agent Plug-in

Chapter 4: Configuring the Agent Plug-in 37 informatica.thread.pool.size

Specifies the maximum number of threads in a database pool that is used for connecting to Informatica.

Default: 20 java.security.egd

Sets the source for random number generation. The agent sets this parameter to file:///dev/urandom to overcome a known issue on Linux with random number generation that can cause applications to be blocked.

jdbc.drivers

Specifies the driver class name for the JDBC driver. Separate multiple drivers with a colon (:).

Example:

oracle.jdbc.driver.OracleDriver:com.microsoft.sqlserver.jdbc.SQLServerDriver plugins.start_internal_n

Specifies the agent plug-in to start by the core Java agent. n

Denotes an integer assigned to the agent plug-in, starting at 1. The n suffix must increase sequentially for each agent plug-in.

How to Set Up an Alias for the Agent Plug-in

When you install the agent plug-in, you are prompted to create a default alias, which you can change after installation. If you enable an alias on the agent plug-in, you must also configure the alias on the scheduling manager.

To set up an alias for the agent plug-in, follow these steps: 1. Create an alias for the agent plug-in (see page 38).

38 Implementation Guide

Create an Alias for the Agent Plug-in

An alias lets you create a unique agent name for an agent plug-in. Each agent plug-in has a default alias, which you can enable or change.

To create an alias for the agent, configure the following agent parameter on the agent: communication.alias_n

Defines the alias name for the agent. The n suffix increments sequentially for each alias agent.

Note: To enable an alias on the agent, verify that the comment character (#) is removed from the parameter line.

Configuring an Alias on the Scheduling Manager

When you create an alias for an agent plug-in, you must also configure the alias on the scheduling manager. You must define the alias on the scheduling manager with the same address, port number, and encryption key as the agent where the agent plug-in is installed.

Note: For detailed instructions to configure an alias on the scheduling manager, see the documentation for your scheduling manager.

Chapter 5: Troubleshooting the Agent Plug-in 39

Chapter 5: Troubleshooting the Agent

Plug-in

This section contains the following topics:

Unable to Establish Connection with Informatica Server (see page 39)

Unable to Establish Connection with Informatica Server

Symptom:

When I enter the machine name instead of the IP address of the machine the

Informatica server runs on in the Information host details CLI, I get the following error message:

Error Code: -109 Error Message: Unable to establish connection with Informatica Server Message Context: IJLMDriver::getConnection (01hw310783, 6014, 360). Note: In the error message, the machine name is 01hw310783, the Informatica listener port for a client is 6014, and the timeout parameter required for the Informatica APIs is 360.

Solution:

The machine name 01hw310783 is unknown to UNIX or Linux and must be added in the hosts file with a mapping IP address.

To correct this problem

1. Enter the following command at the operating system prompt to edit the hosts file:

vi/etc/hosts

The hosts file opens.

2. Add the new host name, for example:

172.26.193.114 01hw310783

A new host is added in the hosts file.

Related documents