Monitoring Executions

Every time a workflow runs, an execution is created. The execution tracking system gives you full visibility into what your workflows are doing, what succeeded, what failed, and why.

The Executions List

The executions list shows all workflow runs across your deployment. From here you can:

Filter by status — show only Running, Succeeded, Failed, Cancelled, or Suspended executions
Filter by workflow — focus on executions from a specific workflow
Filter by date — narrow down to a time range
Sort — order by most recent, oldest, or execution status

Each entry shows the workflow name, execution status, when it started, and how long it ran.

Execution Statuses

Status	Meaning
Running	The workflow is actively executing steps
Succeeded	All steps completed successfully
Failed	One or more steps failed and no error-handling path caught the failure
Cancelled	The execution was manually cancelled before it completed
Suspended	The workflow is paused, waiting for human input (e.g., an approval from a Human Gate node)

Execution Detail View

Click on any execution to see its full detail. The detail view shows:

Step-by-Step Progress

Every step in the workflow is listed with its current status. For completed executions, you can see the sequence in which steps ran and how long each one took.

Step Inspection

Click on any individual step to see:

Inputs — the exact data that was sent to the step after expression rendering
Outputs — the data the step produced (available to downstream steps)
Timing — when the step started, when it completed, and its total duration
Retry history — if the step was retried, how many attempts and what errors occurred
Error details — for failed steps, the error message and error category

Debugging Failed Workflows

When investigating a failure, start by finding the first failed step in the execution. Check its inputs (was the right data passed in?) and its error details (what went wrong?). Most failures come down to either incorrect data or an external service being unavailable.

ForEach Iterations

For ForEach steps, the detail view shows the aggregated results: how many iterations succeeded, failed, and were cancelled. You can expand the ForEach to see each individual iteration as a child execution.

Cancelling a Running Execution

If you need to stop a running execution:

Open the execution detail view
Click the Cancel button
The execution is marked as Cancelled

Steps that are already running will complete, but no new steps will be dispatched. Any pending Human Gate approval requests are also cancelled.

Practical Monitoring Tips

Set up failure alerts: For critical workflows, add a failure edge to the last step that sends a notification (email, Slack, webhook) when the workflow fails. This way you learn about failures immediately instead of discovering them later.

Check execution timing: If a workflow that normally takes 10 seconds suddenly takes 5 minutes, something may be wrong with an external service. The step-by-step timing helps you identify the bottleneck.

Use tags and filters: When you have many workflows running, use workflow tags and status filters to quickly find what you are looking for.

Next Steps

Error Handling and Retries — understand what happens when steps fail
Triggers and Scheduling — set up automated execution
API Overview — check execution status programmatically