[n8n Cloud] Instance persistently unresponsive/hanging after large Excel generation - Restart not fixing it

Questions
1

Description of the issue: Hi everyone,

I am experiencing critical performance issues on my n8n Cloud (Starter Plan) instance. The instance is “stuck” on “Connection Lost“, and standard troubleshooting hasn’t resolved it.

The Workflow & Suspected Cause: I have an AI Agent workflow that queries a database and generates Excel reports.

  • Normal Load: ~10 columns x 300 rows (Works fine).

  • The Event: I recently attempted to generate a heavier report (40 columns x 10,000 rows).

  • The Result: I suspect this caused a massive memory spike or a process that is refusing to die. Since that execution, the instance has been unstable.

What I have tried so far (and failed):

  1. Restarted the Workspace: I used the dashboard to restart, but the sluggishness persists.

  2. Updated Version: I upgraded to the latest stable version (1.121.3) hoping it would clear the cache/processes.

  3. Wait Time: I stopped all interactions for 24 hours to let any queues drain, but the issue remains.

  4. Support: I have reached out to official support but haven’t heard back yet.

  5. No process running: In executions there is no proccess on Running or Queued.

My Configuration (Screenshots attached):

  • Hosting: n8n Cloud (Starter Monthly)

  • Version: 1.121.3

  • Execution Data: I currently have “Save Successful Executions” turned ON (I realize now this might have contributed to bloating the database during the large file generation and now it´s turned OFF).

Question: Since I am on n8n Cloud, I don’t have CLI access to kill specific Node processes or flush the Redis/Postgres buffer manually. Is there a way to perform a “Hard Reset” of the instance memory or clear the execution backlog without waiting for a ticket response? Has anyone else experienced a “zombie” process surviving a workspace restart on Cloud?

Thanks in advance!

2

The issue you’re experiencing is likely due to resource exhaustion from the large Excel generation, especially on the Starter Plan which has limited resources. Try this:

1. Export your workflows and credentials

2. Delete and recreate your workspace in n8n Cloud

3. Import your workflows back in

This will give you a fresh instance without any lingering processes or memory issues. For future large-scale operations, consider upgrading to a higher plan with more resources or splitting the task into smaller chunks.