Flow execution failures

Diagnose and fix Flow execution failures using logs, error patterns, and debugging techniques.

Identifying the problem

1. Go to Logs
2. Filter by Status: Failed
3. Click the failed execution
4. Review error details

Common Flow failures

Invalid JSON in request body

Error: Unexpected token in JSON at position X

Cause: Malformed JSON in API call or transform step

Fix:

1. Validate JSON syntax (remove trailing commas, quote strings)
2. Use JSON.stringify() in transform steps
3. Test with online JSON validator

Missing required parameter

Error: Missing required field: {fieldName}

Cause: Flow input missing expected field

Fix:

1. Add validation at Flow start
2. Provide default values: {{input.field || "default"}}
3. Update callers to include required fields

API call returns 500 error

Error: External API returned 500

Cause: Third-party API encountered server error

Fix:

1. Add retry logic with delay steps
2. Implement error handling and fallback responses
3. Contact API provider if persistent

Record not found

Error: No Record found matching criteria

Cause: Retrieve Record step found no match

Fix:

1. Add conditional check: {{retrieveRecord.output != null}}
2. Provide fallback behavior in else branch
3. Verify lookup field and value are correct

Transform step JavaScript error

Error: ReferenceError: X is not defined

Cause: JavaScript syntax error in transform step

Fix:

1. Check variable spelling and scope
2. Use const or let to declare variables
3. Test transform logic in isolation

Debugging workflow

1. Identify failing step: Check logs to see which step failed
2. Review input: Inspect data passed to that step
3. Test in isolation: Run Flow with test data that triggers error
4. Add logging: Use transform steps with console.log() to inspect values
5. Fix and verify: Update Flow, test, publish

Preventing failures

Input validation

Add validation at Flow start:

1
if (!input.email || !input.email.includes('@')) {
2
  throw new Error('Invalid email provided');
3
}

Test Flows with edge cases: empty inputs, null values, invalid data, API failures. Proactive testing prevents production failures.

When errors are expected

Some failures are normal and should be handled, not prevented:

• User provides invalid input → Return clear error message
• External API temporarily down → Retry or queue for later
• Record doesn’t exist → Create new Record or use defaults

Design Flows to handle these scenarios gracefully.

Next steps

• Debugging flows for detailed debugging techniques
• Debugging failed executions for production troubleshooting
• Viewing execution logs to access error details
• Common errors and solutions for quick reference