Clone a PostgreSQL Cluster
You can create a new PostgreSQL cluster from a backup of an existing one. Unlike v1, the PostgreSQL v2 does not have a dedicated clone operation; cloning is performed through the standard create cluster flow with a backup restore payload restoreFromBackup. The new cluster is pre-filled with the source cluster's configuration, which you can adjust before saving.
Note: If you set a Recover target time, the backup is replayed up to that point in time (PITR). If left empty, the full backup is applied.
To clone a cluster, follow these steps:
Navigate to the cluster
1. In the DCD, go to Menu > Databases > PostgreSQL.
A list of all PostgreSQL clusters is displayed.
2. Select a cluster from the list by clicking on its name. Alternatively, click Details in the OPTIONS column.
3. In the cluster details page, click Clone.
Define clone properties
In the Properties section, specify the following:
1. Cluster Name: Enter a name for the cloned cluster. By default, the name is set to name-of-the-original-cluster-clone. The name can contain a maximum of 63 characters, including letters, numbers, spaces, slashes, dashes, underscores, and dots.
2. Location: Select the region where the cloned cluster will be deployed. In v2, each cluster is bound to a region-specific API endpoint, so the location determines which endpoint manages the cluster, and affects latency for all cluster operations.
Configure the cluster
In the Cluster configuration section, review and adjust the inherited settings as needed:
1. PostgreSQL Version: Select a PostgreSQL version from the drop-down list. You can only upgrade to a higher version; downgrades are not supported.
2. Instances: Enter the number of instances for the cluster. We recommend a minimum of 3 instances for production usage.
3. Synchronization mode: Select the replication mode from the drop-down list:
Asynchronous: Transactions are committed without waiting for replicas to confirm. Offers better performance but may result in a small amount of data loss during failover.
Synchronous: Each transaction waits for confirmation from at least one replica before committing. Ensures no data loss on failover at the cost of some latency.
4. Connection pooler: Select the connection pooler mode from the drop-down list to manage client connections to the database. Select Disabled if connection pooling is not required.
5. Backup location: Select the region where cluster backups will be stored. We recommend selecting a backup location different from the primary location to enhance data safety.
6. Backup retention (days): Enter the number of days backups are retained. The valid range is 1–365 days.
Configure instances
In the Instance configuration section, adjust the resources allocated per instance:
1. Number of CPUs (per instance): Use the slider to set the number of CPU cores per instance.
2. RAM Size (per instance): Use the slider to set the amount of RAM per instance.
3. Storage Size: Enter the storage size in GB. For optimal performance, we recommend setting the storage size to 100 GB.
Note: The Current availability panel on the right displays the total resources currently in use and the limits available in your contract for CPU, RAM, and storage.
Note: PostgreSQL v2 clusters use SSD Premium storage exclusively. HDD and SSD Standard storage types available in v1 are not supported in v2.
Configure the network
In the Network configuration section, provide the network details for the cloned cluster:
1. Virtual Data Center: Select a datacenter from the available list.
2. VDC LAN: Select a LAN for your datacenter.
3. IP/Subnet: Enter the private IP address and subnet for the cluster.
Note: Make sure the selected private IP address is not already in use within the DCD.
Set the maintenance window
In the Maintenance period (optional) section, you can set a maintenance day and time:
1. Day: Select a day from the drop-down list. 2. Start Time (UTC): Enter a time using the pre-defined format (hh:mm). The maintenance window covers a 4-hour period from the selected start time.
Note: If no maintenance period is specified, maintenance can occur any time from Monday to Friday, 7 AM–5 PM.
Configure Observability
In the Observability section, you can allow the following:
1. Enable logs: Turn-on to forward cluster logs to your logging endpoint.
2. Enable metrics: Turn-on to forward cluster metrics to your monitoring endpoint.
Note: Enabling observability (logs and metrics) may incur additional costs on your bill. Observability services must be activated for your contract before logs and metrics can be collected. To activate the Observability service, see Central Logging and Central Monitoring.
Create the database user
In the User Creation section, provide the following details:
1. Username: Enter a username for the PostgreSQL cluster. The username must be between 1 and 63 characters and contain only lowercase letters, numbers, or hyphens. It must start and end with a lowercase letter.
2. Password: Enter a password for the respective user.
3. Database Name: Enter a name for the initial database to be created in the cluster.
Important: If the user already exists in the backup, its credentials will be overwritten.
Result: The cloned cluster is created. Its STATE is set to PROVISIONING while the cloning process is in progress. Once complete, the cluster returns to AVAILABLE.
Next steps
Restore from Backup: Restore the cloned cluster to a previous state using an available backup.
View Cluster Metrics: Monitor the performance of your cloned cluster.
Last updated
Was this helpful?