After the recent Railway major outage, my n8n services started entering a continuous crash loop. The containers start normally, but fail shortly after because they cannot establish a connection to the internal Redis service, returning repeated ETIMEDOUT errors. Redis appears as online in the dashboard and has an attached persistent volume, but the application cannot reach it over the internal network. As a result, n8n exits after retrying for 10 seconds and restarts indefinitely, making the service unavailable. Redeploying and restarting did not resolve the issue.