What it does
Under /admin/databases an administrator can request a managed PostgreSQL database from DigitalOcean. A reconciler drives the request from requested to ready, and a database counts as ready after, and not before, an authenticated TLS connection has reached the intended database. It is PostgreSQL, not KynetraDB. It runs on a single local primary, and pause, resume and on-demand backup return typed unsupported answers.
How it works
The create call stores the intent (tenant, provider account, name, engine, version, region, plan and node count) in a protected record, under an operation id, before any provider call, so a retry with the same operation id replays the first result. A background reconciler makes the provider calls; the create call does not. If a create has an unknown outcome, the reconciler looks for the resource by a provider-side tag instead of repeating the create. When ownership or uniqueness cannot be shown, the record becomes reconciliation_required.
The states are requested, provisioning, then ready, failed or reconciliation_required; and from ready, deleting, then deleted or cleanup_required. Ready needs the provider to report ready and a certificate- and name-verified TLS probe that signs in as the intended user to the intended database. Delete needs tenant ownership, the cloud account identity and the resource identity, and ends with a confirmed removal or an authoritative not-found. Anything ambiguous becomes cleanup_required and never deleted.
Credentials live in the sealed envelope of the provider vault. List, get, audit and log output carry none. GET /admin/databases/{id}/credentials returns them to a platform administrator or an organisation owner. Every state change is written to the tamper-evident admin audit log. Platform administrators can act on any tenant, and organisation owners and administrators on their own.
Registrations from an earlier implementation that kept metadata rows in a JSON file are listed read-only as unverified registrations. They never authorize a deletion.
Where you reach it
- HTTP
GET|POST /admin/databases - HTTP
GET|DELETE /admin/databases/:id, GET /admin/databases/:id/credentials - HTTP
POST /admin/databases/:id/reconcile - HTTP
POST /admin/databases/:id/pause, /resume, /backup (typed 501 unsupported)
Limits
- PostgreSQL, not KynetraDB. This creates a database for your applications, or one you can point the PostgreSQL storage backend of KynetraDB at. It is not a managed KynetraDB, and managed native provisioning returns a typed unsupported answer. The storage backend is chosen with KYNETRA_STORAGE_BACKEND=postgres (or pg) and KYNETRA_PG_URL; any other value starts the server on the file log without an error, so check the spelling.
- DigitalOcean is the provider with a managed database adapter in this edition.
- It needs a standalone primary with local storage and strict durability (KYNETRA_SYNC=true). Replicas, clusters with failover, Raft, replication quorum and external storage writers are refused with a 503.
- The tests drive a scriptable fake cloud. The opt-in live DigitalOcean test covers App Platform apps, not managed databases.
- Pause, resume and on-demand backup are not supported. A backup request with auto-backup set to true at creation is refused.
Evidence
- On this siteDocumentation hub
- TestsCovered in the engine repository by about 90 tests in the managed-database crate (create reconciliation, deletion ownership, TLS probe, log redaction, credential snapshots, reconciler runtime), 27 more in the provider crate for the DigitalOcean managed-database adapter against a fake cloud, and mounted-router tests in the server
Test coverage is listed as present in the engine repository at this edition. This page does not report test results.
KynetraDB is in private preview
Request a preview to talk through the limits above against your own workload.