Debug and Troubleshoot Workflows

The Debug panel shows:

• Execution status per node.
• Input/output values at each step.
• Error messages.
• Timing metrics.
• Iteration details

Change Log

The change log tracks every edit made to a workflow’s node configuration. It records the timestamp, the user responsible, and a description of each change.

To access it: Click the History/Log icon in the top-right corner of the workflow canvas.

Filter the change log by:

• Date: Select a date range using the calendar icon.
• User: View changes made by a specific team member.
• Node type: Filter by the type of node that was changed.

---

Run and Inspect a Workflow

1. On the workflow canvas, click the Run flow icon in the top-right corner. The Run dialog opens with Input, Flow log, and Output sections.
2. Click the Debug icon to open the debug log. The log populates in real time as the workflow executes.
3. When the run completes:
   • Success: Copy the output using the copy icon. The total execution time is shown.
   • Failure: An error message appears. The output key is empty, and the output is shown in JSON format.

You can stop a running workflow at any time and restart it by clicking Run flow again.

Debug Log

The debug log captures a detailed record of each step in the workflow execution. It shows:

• Flow input values: The values passed when the run was triggered.
• Flow-level log: Initiation and progress details at the workflow level.
• Node-level information: Success or failure status for each node.
• Node success / failure links: Links to additional details per node outcome.
• Tool calling details (AI nodes only): Logs of any workflows called during execution, including inputs (JSON), responses, and errors. A separate trace panel shows step-by-step execution of each called workflow.
• Node metrics for each node:
  • Initiated on: When the node was triggered.
  • Executed on: When the node finished.
  • Total time taken: Duration of node execution.
  • Tokens (AI nodes only): Token usage for the node.

Expand the debug panel to full screen for a cleaner layout. In full-screen mode, all nodes align to the left, and clicking any node shows its input, output, and metrics in a side-by-side view.

Time Metrics for API and AI Nodes

The debug log shows timing breakdowns that help identify performance bottlenecks.

API nodes — synchronous mode:

Metric	Description
Node processing time	Time taken by the node to process and complete.
API response time	Time spent waiting for a response from the external API.

API nodes — asynchronous mode:

Metric	Description
Node paused at	Timestamp when the node paused waiting for the async response.
Node resumed at	Timestamp when the node resumed after receiving the response.
Total wait time	Duration between pause and resume.
Node processing time	Time the platform spent processing the node after it resumed.

AI nodes:

Metric	Description
Node processing time	Time taken by the node to complete execution.
LLM response time	Time taken for the connected AI model to return a response.

View Parallel and Sequential Execution in Logs

Sequential flows: Nodes appear in the debug panel one after another in execution order. This view helps trace linear flow progression.

Parallel flows: Each parallel branch appears as an indented block under the parent node, labeled A, B, C, and so on. You can expand or collapse each branch. All branches run simultaneously, but the workflow waits for all of them to complete before proceeding to the next step.

Debug Loop Nodes

When a loop completes, the debug panel shows:

• Inputs received by the Loop node.
• Per-iteration outputs from child nodes.
• Aggregated results in the output field.
• Errors encountered during any iteration.

Click the icon next to the Loop node in the debug panel to open the Loop Runs view:

• View all iterations with statuses: Running, Completed, or Failed.
• Failed iterations are highlighted in red.
• Click any iteration to inspect its step-by-step node execution.

---

Troubleshoot Failures

Workflow Run Errors

In the All runs tab, click a failed run to view error details. Each error entry includes:

• The HTTP status code returned.
• A message describing the error.
• Suggestions for resolving it.

Error categories:

Category	Description
Authorization	API key authorization failed for the workflow.
Data Validation	Input field or API call data did not pass validation.
Content Filter	An AI node violated a guardrail threshold.
Internal Server Error	A technical issue occurred on the platform server.
Network	A connectivity issue caused the request to fail or time out.

Common error scenarios:

Error scenario	Category	HTTP status
Mandatory input field is missing	Data Validation	400 Bad Request
Incorrect data type for an input field	Data Validation	400 Bad Request
Empty input value	Data Validation	400 Bad Request
Request payload exceeds size limit	Data Validation	413 Payload Too Large
Server-side failure	Internal Server Error	500 Internal Server Error
Request timeout	Network	408 Request Timeout
Guardrail threshold exceeded at an AI node	Content Filter	403 Forbidden

Timeout Behavior and Run Status

The way timeouts affect runs depends on whether the workflow and its API nodes are configured as synchronous or asynchronous.

Workflow mode	API node mode	Behavior
Sync	Sync	Request is fulfilled immediately. Status shows In Progress while running.
Sync	Async (API node timeout < Sync timeout)	Workflow pauses with Waiting status until the external system responds, then resumes to In Progress.
Async	Sync	Workflow executes and sends the response to the callback URL. Status shows In Progress while running.
Async	Async	Workflow pauses with Waiting status. Resumes when the external system responds. If the external system retries the same callback URL, it receives a notification that the request was already fulfilled.

Trigger Issues

Issue	Fix
Output variable is undefined	Open the Start node and define all required output variables. Save and re-run the workflow.
Trigger is inactive or has an authentication error	Redeploy the workflow to refresh connections. Retest the auth profile in Settings > Security & Control > Authorization profiles. Verify the trigger is active before running.