HTTP Request node and Postgres node execute but show "No output data" - WebSocket disconnecting

Questions
1

Hi,
I’m having an issue with my n8n Cloud instance.
When I execute any HTTP Request node or Postgres node, the button spins and then returns “No output data” with no error message. However, simpler nodes like “Edit Fields” work perfectly fine.
In the browser console I can see:

  • WebSocketClient Connection lost, code=1006
  • WebSocketClient Attempting to reconnect in 1000ms
    I’ve already tried:
  • Different browsers (Chrome, Safari, incognito mode)
  • Different networks (WiFi and mobile data)
  • Updating n8n to latest stable version
  • Both the native Supabase node and HTTP Request node
  • A simple GitHub public API call (also fails)
    The Edit Fields node works fine, so the instance itself is running.
    Any help would be appreciated. Thanks!

Describe the problem/error/question

When I execute any HTTP Request node or Postgres node, the button spins and then returns “No output data” with no error message. Simpler nodes like “Edit Fields” work perfectly fine. In the browser console I see: WebSocketClient Connection lost, code=1006

What is the error message (if any)?

No error message shown. Just “No output data” silently.

Please share your workflow

(Select the nodes on your canvas and use the keyboard shortcuts CMD+C/CTRL+C and CMD+V/CTRL+V to copy and paste the workflow.)

Share the output returned by the last node

No output data

Information on your n8n setup

  1. n8n version: 2.27.6
  2. Database: SQLite
  3. Running n8n via: n8n cloud
  4. Operating system: macOS
2

Hi @Pcoc Welcome!
code 1006 means the WebSocket that pushes execution results to the editor is being dropped, so instant nodes like Edit Fields render but longer ones (HTTP Request, Postgres) finish on the backend without their output reaching the UI. The runs are completing, open the Executions list and you’ll see the output that never made it to the canvas.
On Cloud you can’t switch the push backend yourself, so the drop is happening in your connection path. A VPN, corporate proxy, or antivirus doing HTTPS or traffic inspection stays in that path across both WiFi and mobile data and will time out an idle WebSocket, which is why swapping networks and browsers changed nothing. Turn those off and test again.
If it still drops with all of that disabled, it’s instance-side, contact help@n8n.io so they can check your Cloud instance.