Restoring from Backup

The options offered when you restore from backup will vary based on the backup provider you chose when deploying your database.

Restoring to all of the nodes in your pgEdge database is generally the safest way to ensure that all nodes are in sync after a restore. If you are not actively performing writes to your database, or if you have data you maintain only in specific regions, you might opt to restore only a single node in your database.

Restoring from a pgBackRest Repository

To restore your database, select Restore from the drop-down Actions menu in the upper-right corner of the Backups dialog.

Use fields on the Restore Database dialog to describe how you would like to restore into database nodes:

• Use the Source Node field to specify the source node from which you wish to restore a database backup.
• Use the Target Nodes field to specify one or more nodes that you wish to restore the database backup into. The target node must be a member of the cluster that the database belongs to.
• Use the Source Repository field to select the repository that you'd like to restore a backup from.
• Select a Restore Method; choose:
  • Restore a specific backup to restore the database to the state it was in when a specific backup was taken.
  • Restore to latest point to restore a database to the latest archived state in the repository.
  • Restore to a specific time to restore a database to the state it was in at a specific time. When the date/time selection fields open, use the calendar and time selector or Relative Time selector to choose the point to which you wish to restore. Note that the point in time you specify must be after the first transaction stored in your repository, but before the latest transaction archived to the repository.

When you've made your selections, type the database name in the confirmation field and click Restore Database to start the restoration process. When the restore completes, replication sets will be created, and replication will be restarted if automatic DDL replication is enabled. If automatic DDL replication is disabled, you will be required to use the Start Replication button (opens in a new tab) to initiate replication of restored tables.

Restoring from a pg_dump Archive

If you are using pg_dump to take backups, you can restore a backup directly into another database node, or download the backup archive (opens in a new tab) for use on another system.

To restore a backup, select the database name in the navigation tree, and choose Restore Database from the Actions drop-down. Then, complete the fields on the Restore Database dialog:

• Use the Backup Name drop-down to select the backup you wish to restore.
• Use the Target Nodes fields to specify one or more nodes on which the backup will be restored. Note that if you have transactions processing,
• Check the box next to Restore schema only? to restore the schema without adding data (recreating empty tables).

To start the restoration process, enter the database name in the confirmation panel and select the Restore Database button. When the restore completes, replication sets will be created, and replication will be restarted if automatic DDL replication is enabled. If automatic DDL replication is disabled, you will be required to use the Start Replication button (opens in a new tab) to initiate replication of restored tables.