Your Postgres service is crash-looping due to a stale container image. Your data is safe, as it lives on the volume which is intact. To fix this, open the service in your project, press Cmd/Ctrl+K to open the command palette, and select "Redeploy source image" to pull a fresh image. A normal redeploy from the 3-dot menu will not work, as it does not re-pull the image.

OK, redeploy helped.

I would expect Railway to notify me about such known issues, not my clients to do that.

Glad the redeploy resolved it. That's fair feedback, and we appreciate you sharing it. So sorry for the impact this had on you.