The main components of a Mirror Activator system are:
Primary database
Disk replication system
Mirror log devices
Mirror Replication Agent
Replication Server
Standby database
The primary database is the source of transactions that are replicated to the standby database. The primary data server maintains the primary database transaction log, and it controls the primary database log devices.
The disk replication system replicates data at the device level. It may include a storage area network (SAN), network attached storage (NAS), or disk mirroring/synchronization mechanism, and it may incorporate both hardware and proprietary system software.
The mirror log devices are off-site copies of the primary database transaction log devices. They are managed by the disk replication system, which requires exclusive write control of the mirror devices, so they are accessible on a read-only basis.
Mirror Replication Agent reads primary database transactions from mirror log devices, and then sends those transactions to Replication Server for distribution to the standby database. The Mirror Replication Agent requires only read access to mirror log devices.
Replication Server receives the replicated transactions from Mirror Replication Agent. Replication Server processes the transactions and converts them into SQL, which it sends to the standby database for processing. When the replicated transactions are processed successfully in the standby database, the standby database is synchronized with the primary database.
The standby database is the destination of transactions that are replicated from the primary database. During normal Mirror Activator system operation, the standby database is always online, processing the replicated transactions it receives from the Replication Server.
During normal system operation, the Replication Server is the only client allowed to send data-changing transactions to the standby database. All other clients are restricted to read-only database access.
