11 days ago
We are experiencing repeated server connectivity issues on our Railway-hosted application.
Symptoms:
Application is intermittently becoming inaccessible.
Users are unable to log in when the issue occurs.
The issue has occurred multiple times today.
Service is restored temporarily but becomes unavailable again after some time.
Impact:
Client operations are being affected due to inability to access the ERP system.
Request:
Kindly investigate the root cause of the recurring service interruption and advise on the resolution. Please let us know if any additional logs or information are required from our side.
Thank you.
3 Replies
11 days ago
This thread has been opened as a public bounty so the community can help solve it. The thread and any further activity are now visible to everyone.
Status changed to Open Railway • 11 days ago
10 days ago
Most likely causes & fixes:
- Memory leak / OOM crash
Check Railway's Metrics tab for memory spikes. If RAM hits the limit before crashes → scale up memory or fix the leak in your app.
- Missing health check
Railway won't restart your app automatically if it doesn't know it's down. Go to service settings → add a Health Check endpoint (e.g. /health).
- Connection pool exhaustion
If using a database, check if connections are being properly released. Add connection limits/timeouts in your DB config.
Immediate steps:
Go to Railway dashboard → your service → Deployments → check logs during the outage window for OOM errors or crash messages
Check the Metrics tab for memory/CPU spikes
See if the service shows as "up" in Railway but stops responding (= connection pool issue) or actually restarts (= OOM/crash)
Share those findings and the root cause will be clear.
9 days ago
You are not alone, and it is not limited to Railway. Other services that I've tried, that are on top of Amazon AWS, are having the same issues. This is definitely not normal.
Is it because Iran bombed two AWS data centers in the Middle East maybe?
Similar issues reported here: https://station.railway.com/questions/railway-servers-inaccessible-across-regi-41a72291
9 days ago
this is happening to my Railway servers too. I haven’t found any reports about AWS connection issues so it seems more like a recent Railway network change making their whole system unstable.