In addition to database replication to a standby server, it is also advisable to maintain regular database backups. This backup may be useful in the event that some data on the production database is lost or corrupted, and the changes are replicated to the standby server, causing it to be corrupted as well. See your database vendor’s documentation for further information on how to schedule a database backup task.
Backup jobs must be logged, and the logs must be checked. Automatic alerts must be triggered if the backup does not take place.
Data integrity must be checked to ensure that the archived data is not corrupt. Hash sums of the archive can be used to test this.
The backup must be transferred to a secure isolated location, for example cloud storage, Netapp or tape drives in a separate data center.