Both volumes in my project disappeared while all four services stayed intact. I am

trying to find out what happened and whether the Postgres storage can be recovered.

Identifiers

Workspace: 021eb9ec-57c2-44b4-81cd-29915712906c

Project: eac841c8-5c5b-4c1a-9f78-d6d4a717e6ba ("n8n", created 2025-02-28)

Environment: aae9afb8-ef2e-4e2a-8641-49446fed9b54 ("production")

Postgres svc: 78ae614d-d20f-4b0a-8038-9ccc464ca1f0

Redis svc: 9b108414-d929-4e14-8621-79ffef16830f

What I observe

1. All four services (Primary, Worker, Postgres, Redis) still exist, created

   2025-02-28, deletedAt = null.

2. Postgres deployments from 2025-02-28 and 2025-05-15 both recorded

   volumeMounts: ["/var/lib/postgresql/data"].

3. The deployment on 2026-08-26T23:27:14Z records volumeMounts: [] and fails with:

   "This service requires a volume to be mounted at /var/lib/postgresql/data.

   Please attach a volume to the service and try again."

4. Redis fails identically for /bitnami. Both volumes went at the same time.

5. project.volumes -> empty. environment.volumeInstances -> empty, including

   entries with deletedAt set. No soft-deleted records remain.

6. I have no record of deleting these volumes and did not do so intentionally.

   The project was idle from roughly 2026-02 until I returned to it on 2026-08-26.

Questions

1. When were the volume instances for services 78ae614d-... and 9b108414-...

   removed, and by what process (reclamation, inactivity, billing)?

2. Is the underlying Postgres storage recoverable in any form, even a stale

   snapshot? I understand my plan reports volumes.maxBackupsCount = 0.

3. If it is not recoverable, please say so plainly so I can stop looking.

The volume held the only copy of a self-hosted n8n database (workflow definitions),

with no external replica. If any retention window is still open, I would be grateful

if it could be checked before it closes.