Run a flow and read output#

Select the run control on the canvas toolbar to run the current flow. The execution log fills as blocks run, and the inspector shows each block's result. This page covers what happens during a run and how to read it.

Start the run#

The run control starts the flow from its Start trigger. It turns into a cancel control while the flow runs.

The engine works out the order from your connections and runs each block once its inputs are ready. Blocks with no dependency on each other can run together. If the connections loop back on themselves, the engine cannot order them. The run stops and records an error in the log.

During the run#

Three things update live while a flow runs.

• The execution log fills with a line each time a block starts, finishes or fails.
• The inspector shows the selected block's output as soon as it is ready.
• The toolbar shows a cancel control in place of run.

The status next to the execution log heading reads "running" while the flow is in progress.

After the run#

When the flow finishes, the cancel control returns to the run control. The status next to the log heading reads "completed" if every block ran cleanly, or "error" if a block failed. A short notification confirms the outcome, including how many blocks failed if any did.

You can run the flow again at any time. Each run starts fresh. The engine clears the previous outputs before the new run begins.

Read one block's output#

Select any block to fill the inspector. The Output section shows that block's most recent result, with the time it took above it. See inspector for the full layout.

Cancel a run#

Select the cancel control to stop a run. Blocks that already finished keep their output. Blocks still waiting do not run, and the log records the cancellation.

For the log itself, see execution log.