Although you can manually run your harvesting routine and transform the data, most organizations prefer to schedule the process to run automatically. This way, InterAction users can be assured that they always have the most up-to-date information.
The frequency of scheduled updates should be determined when you analyze your users’ needs and the frequency of changes in the external system.
There are two separate processes that you need to schedule:
- The harvesting routine.
- The transformation process.
How you schedule the harvesting routine depends on your environment and the external system. For example, if your external system runs on a Windows Server, you may be able to use the Windows Task Scheduler to schedule harvesting. If the external system is on a UNIX server, you need to use a UNIX utility to do this.
To schedule the transformation process, you can use command line parameters within a batch file to launch Application Collaboration and transform a data source or data set. You can then use InterAction Process Manager or a utility, such as the Windows Task Scheduler, to schedule this batch file.
For details about command line parameters, see Transforming Data Sets from the Command Line
Scheduling Recommendations
Before you attempt to schedule Application Collaboration and run it unattended, be sure to thoroughly test the harvesting routine and transformation process!
Keep the following in mind when automating Application Collaboration:
- You should schedule a backup before harvesting and again after a successful transformation.
- Always schedule the transformation to take place after hours. Running transformation during the day may affect InterAction performance for end users.
- Someone should review the Summary and SQL Results logs on a regular basis - preferably after each scheduled transformation. A single error (such as an outstanding claim) can cause many more errors for related data, so resolving the problem right away helps keep the errors to a minimum.