Redis connection timeout causing n8n Primary to crash loop - data at risk

Question

Hi Railway Support/Community,

My n8n instance has been down since approximately 2:20 AM UTC on May 20, 2026, following what appears to have been an infrastructure incident on Runway end (Postgres logs show database system was interrupted; last known up at 2026-05-20 02:20:39 UTC).

Current situation:

Postgres: Online and recovered

Redis: Shows "Online" but displays "Your application is not running or in an unexpected state" in the Database tab

Primary (n8n): Crash-looping with [Redis client] connect ETIMEDOUT — unable to connect to redis.railway.internal

Worker: Online but useless without Primary

I have no access to the n8n interface and I am very concerned about losing my workflows

What I have tried:

Restarted all services in order: Postgres → Redis → Primary → Worker

Verified Variable References are correctly set: QUEUE_BULL_REDIS_HOST=${{Redis.REDISHOST}}, QUEUE_BULL_REDIS_PORT=${{Redis.REDISPORT}}, QUEUE_BULL_REDIS_PASSWORD=${{Redis.REDIS_PASSWORD}}

Confirmed QUEUE_BULL_REDIS_HOST resolves to redis.railway.internal

On advice from Railway's AI Agent, I also tried:

Ensuring only one region is set (instead of 2) in both Redis and Primary under Scale settings

Checking whether a REDIS_URL variable was needed

Changing Redis / Primary region to Europe

None of these resolved the issue.

My main concerns:

Restoring connectivity between Primary and Redis so n8n comes back online

Ensuring my n8n workflows are not lost — this is critical

This appears to be a network issue on Railway's internal network that did not properly recover after last night's incident.

Thank you