Hello Railway Support,

I’m running an n8n custom deployment on Railway with a connected Railway Postgres database. Everything works fine immediately after deployment, but the service always degrades within ~24 hours and stops functioning until I redeploy. I've read similar threads but no solution help.

Environment

• App:n8n-railway-custom

• n8n version:1.109.2

• DB: Railway Postgres (same project, Docker image, restarted recently)

• Region: [insert region if you know it, e.g. us-west2]

Problem Description

• After redeploy, the service starts normally:

  • n8n initializes, workflows activate, HTTP requests return 200/304 as expected.

• Within ~24 hours, the service degrades:

  • Deploy logs show:

    503 Database is not ready!
    timeout exceeded when trying to connect
    Failed to hard-delete executions
    

  • HTTP logs:

    • /rest/push continues working (status 0).

    • Other endpoints (e.g. /home/workflows, /cta/become-creator) return 503.

• Redeploying the app restores service, but the cycle repeats daily.

What I’ve Checked

• DB credentials are correct.

• No disk/memory/resource alerts on Railway dashboard.

• Build and deploy succeed cleanly.

• Issue persists regardless of redeploy.

Question

• what is the cause?

Any help or guidance to resolve this would be greatly appreciated