Agents

With an agent, you can define actions to perform once or recurring at a scheduled time.

An agent is defined by:

Adding data sources used by actions as sources of information.
Adding actions used to perform dfferent kinds of tasks.
Defining a schedule.

Create an agent

Open Genus Studio.
In the Directory tree, click the Agents folder, and then on the toolbar, click New .
In the Name box, type a name for the Agent.
Optionally type a description in the Description box.
If you want to receive an alert each time the Agent has finished running, select the Notify owner each time the agent has finished running check box. To change the owner for an agent, see below.
Follow the instructions described below for adding data sources, adding actions, defining a time schedule, and enabling the Agent.

If there are more than one Data Set, the agent schedules one execution for each data set. To limit the execution to one data set, do the following:

In the File menu, click Properties.
In the Data Set Binding section, click Data Set and then select the data set from the list.

If the Agent fails to execute, the owner or the delegate for the Agent will receive an alert. When you create a new Agent, you would automatically be assigned the ownership. To change the owner or delegate the Agent to another user account, do the following:

In the File menu, click Properties.
Click the Details tab.
In the Owner or Delegate box, select a user account.

Add Data Sources to an Agent

Click the Data Sources tab.
Follow the instructions in the step-by-step procedure for adding a new data source.

Add Actions to an Agent

Click the Actions tab.
Follow the instructions in the step-by-step procedure for adding a new action.

Define a Schedule for an Agent

Scheduling of Agents are done in the Execution Monitor. You can add new schedules or copy existing schedules from other agents.

Add Schedule

To schedule an Agent do the following:

Open Genus Studio.
In the Directory tree, expand the Agents folder, and then click the Execution Monitor folder.
In the list of Agents, right-click the Agent to schedule, and click Schedule.
In the Schedule Type list, select the type of schedule. You can choose between a daily, weekly or monthly schedule, or if the Agent should be run only once.
In the Start Time box, enter the time to run the Agent.

Depending on the type of schedule selected, the information about when to run the Agent will vary:

Schedule once	1. Enter the date the Agent should run on.
Schedule daily	1. Enter a day interval, that is, if the Agent should run every day, every 2 days, every 3 days and so on
Schedule weekly	1. Enter a week interval, that is, if the Agent should run every week, every 2 weeks, every 3 weeks and so on. 2. Choose which weekdays to run the Agent.
Schedule monthly	1. Select which day of the month, which weekday to run the Agent, or the last day of the month. 2. Select which month(s) to run the Agent.

The next run time for the Agent is acquired from the Application Server when you save the schedule, and is displayed in the Next Run Time column.

Advanced scheduling options

You can specify a period an Agent should run, and schedule it to repeat in an interval less than one day:

Click Advanced.
In the Start Date box, enter the date the agent should start.
To specify when the agent should stop running, select the End Date check box, and enter the date of the last day the agent should run.
Select the Repeat Agent check box.
Specify the number of minutes or hours in which you want the Agent to be repeated.
Specify the time to end repeating the Agent or the duration to repeat the Agent.

Add multiple schedules:

Select the Show multiple schedules check box.
To add a new schedule, click New. To remove a schedule, select the schedule and click Delete.

Specify max queue time:

Select the Max Queue Time check box, and specify the maximum number of seconds from the scheduled start time that the Agent can be queued before it is executed. The instant an Agent is scheduled to run it is placed in a queue and executed as soon as possible. For Agents which repeats in an interval less than one day, it may be critical for the result that the execution is not delayed. If this time limit is exceeded, the Agent is removed from the queue.

Specify max running time:

This setting allows you to limit the amount of time an Agent is allowed to run. Select the Max Running Time check box, and specify the maximum number of seconds the Agent is allowed to run. If the Agent is forced to stop when this time limit is exceeded, the owner or the delegate for the Agent will receive an alert.

The agent may be interrupted before max running time is exceeded, due to the App Services Recycling settings on the application server.

Copy Schedule

To copy an existing schedule from an Agent, do the following:

Open Genus Studio.
In the Directory tree, expand the Agents folder, and then click the Execution Monitor folder.
In the list of Agents, right-click the Agent to copy the schedule from, and click Copy.
Click the Agent to copy the schedule to, and click Paste Schedule.

Enable or Disable an Agent

To enable or disable an agent, or to see the current status of an agent, open the Execution Monitor. Every agent is listed with information on whether is is enabled or disabled, current state, time and date of the last execution, time and date of next planned execution, the schedule, and which data set it belongs to.

To disable or enable an agent do the following

Open Genus Studio
In the Directory tree, expand the Agents folder, and then click the Execution Monitor folder.
If the agent is disabled, right-click the agent and click Enable.
If the agent is enabled, right click the agent and click Disable.

View the Execution History for an Agent

Open Genus Studio.
In the Directory tree, expand the Agents folder, and then click the Execution Monitor folder.
Right-click the Agent, and then on the shortcut menu, click Execution History.
To view more details for an execution, select the execution in the log and click View Event.

In the Execution Monitor, stopped agents are listed in gray font color, failed agent executions are listed in red font color, successful executions with warnings are listed in blue font color.

You can also view the executions for an Agent under the History entry in the navigation pane, by clicking the Execute History entry. Here it is possible to view the execution history for all Agents.

An execution will report one of the following statuses:

Stopped. The agent is stopped, either temporarily by the system, or after being disabled by a user.
Scheduled. An agent is scheduled for execution after having been stopped.
Queued. The next run time is passed, and the Agent is placed in queue for execution.
Timed Out. The Agent has been queued for a period longer than the defined maximum queue time.
Running. The Agent is still running.
Succeeded. Execution of the Agent finished successfully.
Failed. Execution of the Agent failed. You should inspect the details for the execution to find the source of the error.

By default only the last 100 executions are displayed. If you want to change the number of executions displayed, do one of the following:

Click Display the last executions and enter the number of executions.
Click Display all executions since and enter a date.

The amount of information traced for each execution can be specified in the history setup for the Agent. To change this setting do the following:

Open Genus Studio.
In the Directory tree, expand the Agents folder.
Right-click the Agent, and then click Open.
In the File menu, click Properties.
Click the History tab.
Select the Save detailed execution trace in history check box if you want to trace more detailed information for each execution. By default this setting is not activated. Note that activating this setting may generate a large amount of data for each execution.

Test Run an Agent

If you have changed the definition for the Agent, you must deploy these changes to your computer (i.e. Deploy Directory to This Computer).
Start the Genus desktop client.
In the File menu, point to Test Run and then click Agent.
Follow the instructions in the Test Run wizard.

Note that it is important to follow the steps above in the given sequence. Do not change the definition of the agent or related parts of your application model after the deploy, and then do the test run. In such circumstances the test run might fail of subtle technical reasons.