Backup and Recovery

We recommend to use AWS Backup to create periodic backups of Amazon RDS for PostgreSQL and Amazon EFS.

If you do not want to use this service, you may consider the following alternatives:

Backup

Follow the procedures described in Create a Scheduled Backup to create a scheduled backup. Include RDS and EFS as resources. The corresponding identifier can be found in the Output tab of the stack.

Restore

Before you attempt to restore data, you may want to contact the Support in order to determine the best recovery strategy. The main reason is the restore to point in time operation can take several hours to complete depending on the volume of transaction logs to be applied on a given database backup. Also, any restore, be it snapshot or point in time, creates a new RDS DB instance. You may point the application to the new DB instance, but that impacts the integrity of the AWS Cloud​Formation stack.

Apart from that, the restore procedure in general looks like:

  1. Stop the application as described in Stopping Avantra for AWS.

  2. Rename the RDS instance as described in Renaming to Replace an Existing DB Instance. Use e.g. ${NamePrefix}-postgres-multiaz to ${NamePrefix}-postgres-multiaz-renamed.

  3. Restore the database as described in Restoring from a DB Snapshot or Restoring a DB Instance to a Specified Time, and use ${NamePrefix}-postgres-multiaz as new DB instance name.

    This creates second database instance containing the restored database using the original name.

  4. If required, restore files to the EFS as described in Using AWS Backup with Amazon EFS

  5. Start the application as described in Starting Avantra for AWS.

  6. Delete the RDS Database Instance ${NamePrefix}-postgres-multiaz-renamed.