Best practices for Serverless RPA flows

You must follow these recommendations when designing and executing RPA flows in a serverless environment:

• Use Log Execution Message for visibility: Log informational, warning, and error messages during execution to support troubleshooting and auditing.
• Use Linux-compatible file paths: Use relative paths under /tmp/. Avoid Windows-based paths that are not supported. For example, C:\Folder.
• Validate activity compatibility: Ensure all activities used in the flow are supported for serverless execution to prevent runtime failures.
• Avoid desktop-based applications: Do not use desktop application activities, as applications cannot be installed in the serverless container.
• Use cloud-based Microsoft 365 services: Replace desktop office activities with Microsoft graph-based activities for services such as Outlook, OneDrive, SharePoint, and Excel.
• Do not rely on local data persistence: Each execution runs in a new container. Data and files are not retained after the job completes.
• Use cloud storage for file persistence: Use S3 activities to store and retrieve files across job executions.
• Adopt an API-first approach: Use ION API activities to interact with applications instead of UI-based automation where possible.
• Use ION Scripting for Python execution: Use the ION Scripting /execute API endpoint instead of unsupported Python invocation activities.
• Implement structured exception handling: Use Try/Catch blocks and include logging within Catch blocks for better diagnostics.
• Follow well-architected framework guidelines: Design flows in accordance with Infor’s well-architected best practices.