The volume attached to profound-miracle is in a healthy state: it is mounted at /node/home/.n8n, currently using about 83 MB, and is not pending deletion. There is no scheduled deletion on this volume, so no restoration action is needed.

You checked the currently attached replacement volume, not the detached original volume I am trying to recover.

The attached volume contains the new n8n database created on 2026-08-30 and returns “No workflows found” when running n8n export:workflow --all.

Railway previously confirmed that a separate original volume remains detached, contains approximately 4 MB including my previous database.sqlite, and is scheduled for permanent deletion on 2026-08-31 at 22:08 UTC.

Please query all volumes in the project, including detached volumes with no serviceId and volumes where isPendingDeletion=true. Do not limit the query to the volume currently attached to profound-miracle. Please provide the detached volume ID and restoration link or cancel its pending deletion server-side.

Also, the verified mount path is /home/node/.n8n, not /node/home/.n8n. Please resolve this contradiction before the deletion deadline.

To correct our earlier reply: the detached volume you are looking for is not in the project you referenced, it is in a different project in your workspace. That volume (n8n-volume, ~4 MB, mounted at /home/node/.n8n) is still intact and its scheduled deletion is 2026-08-31 at 22:08 UTC, so you have a narrow window to restore it. Open that project's Activity feed, find the volume deletion entry, and click Restore to cancel the pending deletion and recover the volume with its data.