Railway Application Failed to Respond (502)

How to fix it

Try these in order — the most common fix is first.

1. 1. Correct the target port

   Copy steps
   1. Open your service settings and find the public domain
   2. Set the target port to the port your app actually listens on (for example 8080 instead of 3000)
   3. Redeploy and reload the domain
   Did this fix it? Yes No

2. 2. Bind to 0.0.0.0 and the PORT variable

   Copy steps
   1. Railway injects PORT automatically; read it from the environment
   2. Node/Express: app.listen(process.env.PORT || 3000, '0.0.0.0')
   3. Next.js: next start --port ${PORT-3000}
   4. Uvicorn: uvicorn main:app --host 0.0.0.0 --port $PORT
   5. Gunicorn binds to 0.0.0.0 and PORT by default
   Did this fix it? Yes No

3. 3. Check for overload

   Copy steps
   1. Open the Metrics tab of the service and look at CPU
   2. Node.js peaking near 1 vCPU suggests it is saturated
   3. Add replicas in one or more regions to scale horizontally
   Did this fix it? Yes No

Symptoms

Visiting your Railway domain shows 'Application failed to respond' with a 502 Bad Gateway status, even though the deploy shows as successful.

Why it happens

• App not listening on host 0.0.0.0 and the injected PORT variable (most common)
• Target port of the public domain does not match the port the app uses
• App overloaded and unable to answer requests

This guide also covers

• railway application failed to respond
• railway 502 bad gateway
• railway app not working

Other Railway errors

• Railway: No Start Command Could Be Found
• Railway Slow Deployments and Slow App
• Railway Node.js SIGTERM Handler Not Running
• Railway ENOTFOUND redis.railway.internal
• Railway Postgres: Could Not Locate a Valid Checkpoint Record
• Railway SSL Certificate Stuck or Not Working (526, 525, Too Many Redirects)
• Railway 405 Method Not Allowed on POST Requests
• Railway Services Stopped: Trial Credits Ran Out