Akasha.
DocsAbout managed KubernetesManaged PostgreSQL (Akasha PSQL)

Managed PostgreSQL (Akasha PSQL)

Create highly available PostgreSQL databases on a cluster, connect with the built-in psql terminal, and schedule S3 backups.

Akasha PSQL runs managed PostgreSQL databases on your cluster, powered by CloudNativePG. It requires a Provisioned cluster with the CloudNativePG add-on installed — the Database bundle at cluster creation sets this up, including the storage add-on it depends on.

Create a database

  1. Go to Kubernetes → Databases and click Create Database.
  2. Select the Cluster and a Namespace (default default).
  3. Enter a Database Name (lowercase letters, numbers, hyphens).
  4. Pick the PostgreSQL Version: 17 (default), 16, or 15.
  5. Set Instances: 1 is a standalone server; 2 or more gives you high availability with streaming replicas.
  6. Set the Storage Size (default 5Gi; presets up to 100Gi, or a custom value) and a Storage Class — the cluster's default class is pre-selected.
  7. Optionally expand Scheduling to add node selector key/value pairs and tolerations — for example, to pin the database to a dedicated, tainted worker pool.
  8. Optionally check Enable Scheduled Backup to configure S3 backups from day one: a cron schedule (default daily), an S3 destination such as s3://bucket/path, and the name of an existing secret in the namespace holding the keys ACCESS_KEY_ID and ACCESS_SECRET_KEY.

The database appears in the list with its status, ready instance count, storage, and current primary.

Connect to a database

The database detail page's Terminal tab opens a psql session in your browser once the database reports a healthy state:

  1. Choose primary (read+write) or replica (read-only).
  2. Optionally set the database name (default postgres). Changing either selector reconnects the session.

To reach the database from outside the cluster — for an application running elsewhere or a desktop client — expose its service with a TCP route on an External IP, using PostgreSQL's port 5432.

Scale and configure

  • Overview — scale Instances up or down (minimum 1), and view the primary, replication mode, and PostgreSQL parameters.
  • Configuration — edit instances, resource requests and limits, synchronous replica bounds, superuser access, and parameters such as shared_buffers and max_connections, either as a form or as full YAML.

Databases deployed through GitOps are read-only in the console — change them through the Git source.

Back up a database

The Backups tab manages scheduled backups to S3-compatible object storage:

  1. Pick a schedule preset and a Retention period (default 7d).
  2. Enter the S3 endpoint, access key, secret key, and bucket. Backups are written under cnpg-backups/<database> in the bucket.
  3. Save. The credentials are stored as a secret inside your cluster.

Use Backup now for an immediate backup; the backup list refreshes automatically. Removing the backup configuration stops the schedule.

Point-in-time recovery and restores are not available in the console — contact your administrator to restore from a backup.

Delete a database

Deleting a database destroys all of its data. On the database detail page, choose delete and type the database name to confirm.