Training Create a Data Flow
What you'll learn: how to create a data flow, wire nodes in the flow designer, and deploy it. You need: Administrator or Developer access type, an Application Development application. Time: 10 minutes.
Pick the flow Type — it determines how the flow is triggered:
| Type | Use it for |
|---|---|
| System | General backend logic |
| Screen | Logic called from screens |
| Integration | Connection-triggered exchanges |
| Edge | Logic running on Edge Gateways |
| Document | Document generation |
- Go to Left menu → Flows & Scripts → Data Flows. (Fallback: header Search.)
-
Click +. In Create a new Data Flow, type the Name (e.g.
Training Hello Flow), pick the Type (System) and Module, keep Initial Version Number 0.0.1 → click the orange +. You land on the flow's detail page (tabs: DETAILS, VERSIONS, DEPLOYMENTS, SCHEDULES, MODELS, DATA CHANGES).
- Click the flow-designer icon (⑂) in the toolbar. The flow designer opens: canvas left, Toolbox with node categories (Debugging, Fuuz, Transformations, Scripts, Conditionals, Context, Flow Control, Events, Integration, Device Gateway, Validation, Notification), Console right.
- Drag a Source node (Debugging) onto the canvas — it provides a test payload for runs.
- Drag a Log node (Debugging) below it.
-
Drag from the Source node's Output port to the Log node's Input port to connect them with a wire. Clicking a node opens its editor (Name, Note, Configuration — e.g. the Source's test Payload JSON).
- Click File → Save (💾). Toast: "Data Flow Version updated successfully."
-
Click the deploy arrow (⬆). In Deploy Data Flow, note Force Stop Previous Versions (for replacing a running version) → click ⬆.
The Data Flows table shows the flow Active ✓ with Current Version 0.0.1. Executions will appear in Logs → Data Flow Logs and on the Execution Metrics dashboard.
Open the flow designer again, click the Source node, put {"hello": "world"} in its Payload, and use the Log node to see it in the Console when the flow runs. Then look at Flows & Scripts → Flow Schedules to schedule the flow.
A flow can chain many nodes. For example, drag Source → Query → Log and wire Source.Output → Query.Input, then Query.Output → Log.Input — the Query node fetches data and the Log node records the result. Connections are drawn by dragging from an Output port to an Input port.
The Toolbox groups nodes by category — Debugging (Source, Log, Echo, File Source), Fuuz (Query, Mutate, Aggregate, Execute Flow, Render Document, System Email, Saved Query, EDI Translation, Inter-Tenant Request), Transformations, Scripts, Conditionals, Context, Flow Control, Events, Integration, Device Gateway, Validation, and Notification.
Training: Create a Data Model · Training: Investigate Application Logs · Training: Save Queries, Scripts, and Data Mappings