Skip to main content
Perform manual backups and restores of ClickHouse data using SQL commands and S3 storage. For full syntax details, see the upstream ClickHouse backup documentation. For API-driven backup operations, see Manage Backups via API.

Prerequisites

  • An S3 bucket in the same region as your cluster
  • The IAM role tied to the Kubernetes service account must have read, write, and list access to the bucket

S3 Path Format

All backup commands use the following S3 path:

Perform a Full Backup

From one of the server pods, run:
The backup runs asynchronously. Check progress via the system.backups table:

Incremental Backups

See Take an incremental backup in the upstream docs.

Restore from a Backup

Run on the cluster you want to restore into:
$RESTORE_ID is any unique identifier you assign to the restore operation.
Last modified on August 7, 2026