a month ago
Hello Railway team,
I am preparing a Production PostgreSQL environment on the Hobby plan and would like to migrate the existing database service and its attached volume from Amsterdam to Singapore.
Current environment:
Railway CLI version: 5.41.2
PostgreSQL volume: approximately 315 MB used
PITR enabled, with successful full and differential backups
No application business tables or data yet
Existing service deployment region: Amsterdam
Target region: Singapore
On October 9, 2026, at 04:26:14 (UTC+8), I attempted the migration using railway scale, specifying the existing Amsterdam region with zero replicas and Singapore with one replica.
The command exited with code 1. Unfortunately, the original stderr was not preserved.
Subsequent read-only checks confirmed that the database and volume remained in Amsterdam, PITR archiving continued, and no new deployment or committed patch was observed.
Could you please clarify:
Is this volume-backed PostgreSQL region migration supported on Hobby with PITR enabled?
Does the legacy region key ams affect the operation?
Are there any volume migration or PITR prerequisites?
Can Railway investigate the failed attempt using the timestamp?
What is the safest supported procedure for moving the database to Singapore?
I have not retried the operation.
Thank you.
1 Reply
Status changed to Awaiting Railway Response Railway • about 1 month ago
a month ago
The documented way to move a volume-backed service is to change its region in the service's Settings (Deploy region). When a volume is attached, that change migrates the volume to the new region. The service is down while the migration runs, and how long that takes depends on the volume size. Volume backups and PITR archive data follow the volume's region. See Regions.
The ams key is not the cause. The CLI's railway scale accepts friendly keys like ams or eu-west as well as full region IDs such as europe-west4-drams3a and asia-southeast1-eqsg3a. The command doesn't check for attached volumes before sending a multi-region change. Any rejection of a volume-attached service comes back from our API, so the exact reason for your exit code 1 is in the stderr that wasn't saved.
Your Postgres service's latest deployment is still the successful one from October 7, which matches your observation that nothing changed.
For a single-volume database, the region picker in the service settings is the safest path, since it runs the volume migration as one operation rather than as a replica split across two regions.
Status changed to Awaiting User Response chandrika • about 1 month ago
a month ago
This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!
Status changed to Solved Railway • about 1 month ago