Can you share the error that is causing n8n to crash?

The 502 is just the public domain saying n8n didn't answer in time. The real issue is in the Railway canvas screenshot: Postgres and Redis are online, but both Primary and Worker are crashed.

So I’d start with the Primary service, not the domain.

Click Primary -> latest failed deployment -> logs, then look for the first real error before it exits. Do the same for Worker after that.

For this n8n template, the usual causes are:

• missing/changed N8N_ENCRYPTION_KEY
• Postgres env vars not pointing to the Postgres service anymore
• Redis/queue variables different between Primary and Worker
• n8n version migration failing during startup

Don’t delete the Postgres/Redis volumes if you have real workflows there. Paste the first 30-50 lines around the crash from the Primary logs and it should be much easier to tell the exact fix.

You can use this template as an alternative: https://railway.com/deploy/n8n-pro-stack

You need to deploy your Redis service(s). It's offline.