Hi Railway team,

I need urgent help recovering data from a Postgres service that was shut down today when my trial expired. The container itself was healthy and serving production traffic right up until the shutdown — this was not a crash, and I'm hoping the underlying disk/filesystem might still be recoverable even though no persistent volume was attached.

Project: diplomatic-art

Project ID: 15998120-0eed-46eb-ab59-37824c92f29e

Service: postgres

Service ID: dee91cdc-2dcd-4bb0-a8d3-7c2eaf48fcf5

Account email: e.balbi00@gmail.com

Affected deployment IDs (the long-running instance with real data):

• 267eebd2-1aaf-47ab-aad3-1604b7c1f031 (created 2026-06-30T14:40:08.741Z)
• 6e92d8b1-a157-4676-849a-decdaae5ebbc (created 2026-06-30T14:40:19.646Z, the one that stayed active)

Per the deployment logs, this instance ran continuously and healthily from 2026-06-30 until:

2026-07-21 16:05:38 UTC — received a "fast shutdown request" / "terminating connection due to administrator command", then shut down cleanly. This timing lines up with my trial expiring.

Because no volume was attached to this service, I believe the container's disk was destroyed on shutdown. I only realized this after the fact, and in the process of restoring the app I redeployed a fresh (empty) Postgres:16 image to this same service today (new deployment created 2026-07-21T17:34–17:35 UTC), so the service is now running a new, empty database.

Is there ANY way to recover the filesystem/disk of the container from the deployment that shut down at 16:05:38 UTC — a snapshot, a backup, or a grace-period copy before it's permanently purged? This database held real production data for a live client (peñas, artistas, eventos, chat messages, comments) that only existed in that database, with no external backup.

I understand this may not be possible, but given how recent the shutdown was (about 1-2 hours ago as of writing), I wanted to ask before that window closes. Any help — even partial — would be hugely appreciated.

Thank you,

Enzo