3 months ago
I attempted a region change on my Postgres service via API. The operation FAILED - the service is still in US-West (California). However, the volume was completely wiped as a side effect.
The region never changed (still shows US-West)
But the volume was reinitialized (PostgreSQL ran initdb on empty volume)
191 production users lost all their data
I'm a Pro customer
This seems like a bug: a failed region change should not have destructive side effects on the volume. If the region can't be changed (e.g., because the service has a volume), the API should reject the request without wiping data.
Railway support bot says they don't restore data for "user-initiated actions", but this was a FAILED action that shouldn't have had any effect.
Project ID: 1a5b24df-f1ae-4fdf-9d5b-ad3b1316088f
Volume ID: 53601a0c-5cd4-4345-8b47-10a7561d69bd
Has anyone experienced something similar? Is there any way to get human support to review this?
2 Replies
Status changed to Awaiting Railway Response Railway • 3 months ago
3 months ago
The current PostgreSQL service appears to have been initialized from scratch on 2026-05-29 at 21:46 UTC. The logs show initdb, CREATE DATABASE, and later errors like relation "public.User" does not exist. I cannot find any backups in the dashboard. Can you check whether there is a previous volume, deleted database service, snapshot, or internal backup available before that timestamp?
Status changed to Awaiting User Response Railway • 3 months ago
Status changed to Closed brody • 3 months ago