Akasha.
DocsAbout managed KubernetesEtcd backups

Etcd backups

Schedule encrypted etcd snapshot backups of a cluster's state to S3, trigger manual backups, and browse backup history.

Etcd stores your cluster's entire state — every object you have created with kubectl, the console, or GitOps. The cluster's Etcd tab schedules etcd snapshot backups to S3-compatible object storage so that state can be recovered after a disaster.

Configure scheduled backups

  1. Open the cluster and select the Etcd tab.
  2. Pick a Schedule: every 6 hours (default), every 12 hours, daily, weekly, or a custom cron expression.
  3. Enter the S3 details: Endpoint, Bucket, Access Key, Secret Key, and optionally Region and an S3 Prefix (by default, objects are prefixed with the cluster name).
  4. Optionally paste an Age Public Key (age1...) to encrypt each snapshot with public-key encryption. Without a key, snapshots are stored unencrypted.
  5. Optionally enable zstd compression.
  6. Save. The schedule runs as a job inside your cluster; the status banner shows the schedule and the last successful backup time.

You can use any S3-compatible store, including Akasha buckets — the S3 endpoint address for buckets is specific to your deployment, so check the console or ask your administrator.

To stop scheduled backups, remove the configuration from the same tab; this deletes the schedule and the stored credentials from the cluster.

Trigger a manual backup

Click Backup now to run an immediate snapshot outside the schedule. The new object appears in the backup history once it completes.

Browse and delete backups

The backup history lists each snapshot with its filename, size, and date. Listing requires the S3 credentials to be entered in your browser: they are kept in your browser's local storage only — Akasha uses them to query the bucket but does not store them server-side.

To delete a backup, use the delete action on its row and type the filename to confirm.

Restore

There is no restore flow in the console. Restoring a cluster from an etcd snapshot is a disruptive, cluster-wide operation — contact your administrator if you need a restore.

Etcd backups protect the cluster's control-plane state, not your application data. Use volume snapshots, database backups, and VM backups for data-level protection.