The Node log on a run’s Run Details screen records each node’s outcome. You can review its status, duration, message, and saved Input and Output details. These records help you check a node’s configuration and result when data does not reach a later step as expected.
Open the node log
- In the WordPress dashboard, open Agent Workflows → Runs.
- Find the run you want to investigate. Use the status views or Filter by workflow if needed.
- Select the run ID or choose View to open Run Details.
- Scroll to the Node log table. Each row represents a recorded node outcome for that run.
Screenshot to add: [Run Details screen with the Node log table]
Understand the Node log columns
- Node: The node label saved when the node ran. If no label was recorded, the table uses the saved node type or (unknown node).
- Status: Success means the node reported success, Error means it reported a failure, and Skipped means the node was not run as a normal successful step.
- Duration: The recorded node execution time. If no duration was saved, the display may not contain a measured value.
- Message: The node’s recorded error message, when one is available. A successful node may have an empty message.
- Details: Choose View to expand the saved Input and Output data when available. If neither was recorded, the cell shows a dash instead.
Inspect Input and Output
- In the relevant row, select View under Details.
- Read the formatted JSON under Input. This is the node configuration recorded when it ran; it is not necessarily a complete record of every runtime value resolved by the workflow engine.
- Read the formatted JSON under Output. This is the node’s recorded execution result. Check whether it contains the fields and values expected by the next step.
- Compare the saved output with the configuration of the downstream node in the workflow editor. Correct the mapping or configuration, then run a new test to confirm the change.
Screenshot to add: [Expanded Node log Details showing Input and Output JSON]
Use errors and skipped nodes to diagnose a run
Start with the earliest relevant Error entry and read its Message. Then expand Details to review the saved configuration and result. A missing output can mean the node did not produce a result that was stored. If the entire run has no node entries yet, the page displays This run has no node log entries yet.
The default failure policy is to stop the run at the first failing node. To review or change the global behavior, open Agent Workflows → Settings → General and check When a node fails. The available choices are to stop the run at the first failing node or continue running the remaining nodes. A node can also be skipped by workflow execution logic, so read its status and message in context.
Know what the records preserve
The node log stores the node label and type as snapshots from execution time, along with the recorded configuration, result, message, duration, and log timestamp. The snapshot helps identify a node even if the live workflow node is later changed or removed. It does not make the current workflow editor a historical copy of the entire workflow.
Run history and node logs follow the configured finished-run retention period. The default is 14 days. To review it, open Agent Workflows → Settings → Logging & Retention. Set Keep finished run history for to 0 to keep finished records indefinitely, or use Purge Now to remove eligible older finished runs and their logs immediately. Deleting a run also removes its saved history, so retain the details you need before deleting it.
Inputs, outputs, trigger payloads, and error messages can contain sensitive data. Share diagnostic details only with people who are authorized to see them, and avoid copying secrets or customer information into public support requests.