Skip to content

Managed PostgreSQL Service

The Managed PostgreSQL Service lets you deploy and run production-ready PostgreSQL databases on your registered infrastructure. The platform handles provisioning, configuration and day-2 operations (resizing, version upgrades, high availability), so you can focus on your data instead of operating a database.

PostgreSQL databases are backed by the CloudNativePG operator running on Kubernetes. The platform manages that layer for you — you work only with the database instance.

PostgreSQL Key Features

  • On-Demand Provisioning: Create a PostgreSQL instance by picking a size, version and storage.
  • High Availability: Optionally run the instance as a 3-instance high-availability cluster for resilience.
  • Vertical Scaling: Change the instance size as your workload grows.
  • Storage Growth: Increase storage on demand. Storage can only grow, never shrink.
  • Version Upgrades: Move an instance to a newer supported PostgreSQL version.
  • On-Demand Credentials: Read live connection details when you need them; passwords are never stored by the platform.
  • Built-in Metrics: Enable per-instance monitoring and view time-series charts (availability, connections, database size, transactions, replication lag) right from the instance's detail page.

Prerequisites

Managed PostgreSQL is offered through a Service Config that an organization administrator prepares on an infrastructure (see Infrastructure, Service Configs and Service Platforms). Before you can create a database, the target infrastructure's Service Config must have:

  1. An MKS network platform (defines the external network), and
  2. A PostgreSQL platform that is enabled.

When you create an instance you select the project and the Service Config; the platform deploys onto that config's PostgreSQL platform.

PostgreSQL Instances

A PostgreSQL Instance is the core resource representing a managed database. You first choose where the database runs — narrowing by Region, Zone and Infrastructure — and then select the Service Config. Only Service Configs that have an enabled PostgreSQL platform are selectable. You then configure:

  • Project: The project that owns the instance.
  • Service Config: The infrastructure Service Config (and therefore the PostgreSQL platform) the instance runs on.
  • Size: A t-shirt size that maps to CPU and memory for the instance.
  • Version: The PostgreSQL major version. Available versions come from the platform.
  • Storage: The amount of persistent storage, in GiB. Defaults to 20 GiB (minimum 1 GiB). Storage can be increased later but never reduced.
  • High Availability: Whether the instance runs as a 3-instance high-availability cluster. This is a simple on/off switch — the replica count is fixed at 3, not configurable.
  • Monitoring: Whether metrics collection is enabled for the instance. On by default; you can turn it off at creation or toggle it later. See Monitoring & Metrics.
  • Database (optional): The name of the initial database to create. Defaults to app. Set once at creation — it cannot be changed later.
  • Username (optional): The name of the initial application user. Defaults to admin. Set once at creation — it cannot be changed later.

Connecting to your database

Connection details are read on demand from the running instance and are never stored by the platform. From the instance's detail page you can retrieve:

  • Host and Port
  • Database name
  • Username
  • Password
  • A ready-to-use connection string (postgresql://…)

You can reveal and copy each value, and refresh the credentials when you need to. Because credentials are fetched live rather than persisted, the platform can surface them without ever holding your password at rest.

Managing an Instance

After creation you can:

  • Resize: Change the instance size to give the database more (or less) CPU and memory.
  • Grow storage: Increase the storage allocation. This is one-way — storage cannot be shrunk.
  • Upgrade the version: Move to a newer supported PostgreSQL version. PostgreSQL cannot be downgraded, so only newer versions are offered.
  • Toggle high availability: Switch the 3-instance HA cluster on or off.
  • Toggle monitoring: Enable or disable metrics collection for the instance (see Monitoring & Metrics).

Spec changes (version, size, storage, high availability) are applied to the backing cluster by the platform.

What you cannot change

The Service Config, project and infrastructure are fixed once the instance is created, along with the database name and username.

Monitoring & Metrics

Each instance can report metrics to the platform so you can watch its health and load over time. Monitoring is a simple on/off property of the instance:

  • It is enabled by default when you create an instance.
  • You can turn it off at creation, or toggle it at any time from the instance's Edit dialog.
  • Turning it off stops metrics collection for that instance; turning it back on resumes it. No data is deleted from your database either way — this only controls the metrics pipeline.

When monitoring is enabled, open the Metrics tab on the instance's detail page to see time-series charts over a selectable window (last 1 hour, 6 hours, 24 hours or 7 days). The v1 dashboard covers:

  • Availability — whether the database is up.
  • Connections — active client connections.
  • Database size — on-disk size of the database.
  • Transactions — commit/rollback throughput.
  • Replication lag — for high-availability instances, how far replicas trail the primary.

Charts refresh automatically and show gaps (rather than interpolating) when a data point is missing.

When metrics are unavailable

The Metrics tab shows an empty state when monitoring is disabled for the instance (enable it from Edit), or when the instance runs on a Service Config where the platform only records the database and does not manage its cluster. Metrics are always scoped to your own instance — you never see another tenant's data.

Lifecycle States

Managed PostgreSQL instances follow this lifecycle:

  1. Pending: The instance is being prepared for creation.
  2. Creating: The platform is provisioning the database.
  3. Active: The instance is running and available.
  4. Updating: A change (resize, storage growth, version upgrade, HA toggle) is being applied.
  5. Deleting: The instance and its resources are being removed.
  6. Error: An operation failed and the instance requires attention.

Getting Started with Managed PostgreSQL

To deploy your first database, navigate to the Managed Services > PostgreSQL section in the platform dashboard. Pick the region, zone and infrastructure, select a Service Config and a project, then define the instance's size, version and storage.

If the PostgreSQL option is unavailable, ask your organization administrator to add and enable a PostgreSQL platform on the infrastructure's Service Config — see Infrastructure, Service Configs and Service Platforms.