Our databases are unmanaged, so automatic snapshots, WAL recovery, and PITR aren't available to restore your data unless you've enabled them on the service. Volume backups and point-in-time recovery are both native Railway features on Pro. We don't recover data lost to user-initiated actions. Going forward, enable volume backups on your stateful services so you can self-restore if this happens again.

so whats the way forward

To correct our earlier message, your service does have a volume attached at the database path, and it survived the redeploy with ~60 MB of data on it. The data loss was in-place, meaning something in the application overwrote or re-initialized the database file during the redeploy rather than the storage being discarded. We don't recover data lost to user-initiated actions, so restoring the previous database contents is not available. The way forward is to re-enter the lost records and, to protect against this in the future, upgrade to Pro and enable native volume backups on the service so you can self-restore from a snapshot if it happens again.