Restore a MariaDB Cluster from a Backup

With IONOS DBaaS, you may roll back your MariaDB clusters to a previous point in time or use it to create a new cluster. This reliable function prevents data loss and downtime in unexpected situations.

For example, consider that certain values were accidentally deleted from your MariaDB cluster around 11 hours today. The easiest way to recover missing data is to restore a backup before 11 hours.

Result: You can restore only one backup at a time, and you must wait for the restoration process to finish before restoring another backup.

Prerequisites

• The backup must belong to the MariaDB cluster to be restored.

• The cluster must be in the AVAILABLE state.

Restore a Cluster

To restore a cluster, follow these steps:

1. In the DCD, go to Menu > Databases > MariaDB.

2. Select a MariaDB cluster from the list by clicking on its name. Alternatively, click Details in the OPTIONS column.

Select the respective MariaDB cluster from the list

3. The Details window displays the necessary information of the chosen cluster. For more information, see View the list of MariaDB Clusters.

• Point In Time Recovery displays the earliest backup available for restoration. Select Restore backup to choose the date and time for backup restoration.

Point in time recovery option

a\. In the **Restore backup** pop-up window, click within the text box to open the Calendar. 
b\. Backups are available only for those dates that are available for selection. Select a date and click **Ok**.

![Select a date](../../../images/mariadb/mariadb-self-restore-recovery-dates.png)

c\. Select the **Clock** icon to set a time for restoring the appropriate backup from the chosen day and click **Ok** to confirm your selection. 
d\. Select **Restore** to proceed with the restoration. 
e\. Select **Restore** to confirm the restoration. 

![Confirm restoration](../../../images/mariadb/mariadb-self-restore-recovery-confirm-restore.png)

Result: You will receive a confirmation that the cluster will be restored, and the respective cluster's STATE is set to RESTORING.

Restoration is in progress