Common Issues

This guide covers the most common reasons an n8n workflow misbehaves and how to fix each one: a workflow that never triggers, authentication and credential errors, unreachable webhooks, empty or undefined data between nodes, and execution timeouts. Each section lists the visible symptom and the concrete settings, environment variables, and checks that resolve it.

Why is my workflow not triggering?

Symptom: The workflow exists but never executes on its own.

• Check activation. Verify the toggle in the top bar is set to Active. Inactive workflows only run during manual testing.
• Check the trigger node. The first node must be a trigger (Webhook, Schedule, polling trigger). Action nodes cannot start a workflow automatically.
• Webhook URL mismatch. Confirm the external service calls the production URL (/webhook/...), not the test URL (/webhook-test/...).
• Schedule timezone. Verify GENERIC_TIMEZONE matches your expected timezone. Cron expressions use the instance timezone.

How do you fix authentication errors?

Symptom: A node fails with 401 Unauthorized, 403 Forbidden, or a credential-related error.

• Expired tokens. OAuth2 tokens expire. Open the credential and click Reconnect to re-authorize.
• Insufficient scopes. Check the node reference for required scopes, update them in the OAuth app config, then reconnect.
• Personal access tokens. Tokens can be revoked or expire. Generate a new one if the old token no longer works.
• OAuth redirect mismatch. The callback URL must match what is registered in the provider's app settings. If your n8n domain changed, update both sides.

Why is my webhook not reachable?

Symptom: External services report connection errors or timeouts when calling your webhook.

• WEBHOOK_URL not set. On self-hosted instances, set the WEBHOOK_URL environment variable to your public URL (e.g., https://n8n.example.com/). Without this, n8n generates URLs with localhost.
• Firewall or network rules. Ensure port 5678 (or your configured port) is open and reachable from the internet. If behind a reverse proxy, confirm the proxy passes traffic to n8n.
• HTTPS required. Many services (Slack, GitHub, Stripe) require webhook URLs to use HTTPS. Set up TLS via a reverse proxy.
• Path mismatch. Double-check the full webhook path. A typo in the path segment or an extra trailing slash can cause a 404.

Why is no data coming through?

Symptom: The workflow triggers but downstream nodes receive empty data or undefined values.

• Check node connections. Ensure every node is connected. Look for broken (red) lines in the editor.
• Expression syntax errors. Use the correct syntax: {{ $json.fieldName }}. Common mistakes include missing $json, incorrect nesting, and referencing nonexistent fields.
• Wrong data path. After a Webhook node, body data is under {{ $json.body }}. After most other nodes, data is directly under {{ $json }}. Use the input/output panel to inspect actual structure.
• Empty result set. The upstream node returned zero items. A query with no matches passes nothing to the next node.

How do you fix an execution timeout?

Symptom: The workflow fails with a timeout error before completing.

• Increase the timeout. Set EXECUTIONS_TIMEOUT to a higher value (in seconds). A workflow-level setting may also override the instance default.
• Check external services. A slow API or unresponsive database can stall execution. Test the service independently.
• Optimize large data sets. Use SplitInBatches for smaller chunks, or filter data early to reduce volume.
• Webhook response timeout. If using "When Last Node Finishes" response mode, the caller may time out. Switch to "Immediately" and process asynchronously.

See Also

• Error Handling & Retries
• Error Handling Tips
• Testing & Debugging Tips