There are two options for implementing OpenEdge Replication—single-target replication and replication set.

Single-target replication consists of a Source database on the primary machine and a Target database on the secondary machine. The replication server that runs on the primary machine continuously sends database updates to the Target database via the replication agent running on the secondary machine. When the Source fails, the Target transitions to a new Source and replication stops. When a database is restored on the primary machine, it can be synchronized with the temporary Source on the secondary machine. Then, database activity can be failed back to the original configuration.

With a replication set, there are two Target databases. A replication set consists of a Source database on the primary machine and a Target database on each of two secondary machines. One of the Targets is configured as the primary Target, the other as the secondary Target. A replication server runs on the primary machine and a replication agent runs on each secondary machine. The replication server sends Source database updates to the replication agents, which in turn apply the updates to their Targets.

The advantage of a replication set is that if there is a loss of the Source or if the Source is brought down for maintenance, the primary Target becomes the new Source and continues replicating to the other Target. In this way, with a replication set, even if the Source database is down, replication continues.

The overall steps in the OpenEdge Replication process for both option are similar up to step six. At step six for a replication set, the procedure ends. The single-target replication requires additional steps.

  1. During primary (normal) replication, the primary database has the role of source database and the secondary database has the role of target database. The Replication server exists on the source database and the Replication agent exists on the target database.
  2. If there is a failure on the machine hosting the primary database, the Replication agent on the target database loses communication contact with the source database's Replication server.
  3. The Replication agent on the target database enters pretransition.
  4. Transition then occurs (automatically or manually, depending on how you set the transition properties), making the secondary database (formerly a target) a source database.
  5. All database activity is moved to the secondary database, which is now functioning as a source database. The secondary database now becomes the production database.
  6. For the Replication Set option the second target transitions to be an agent of the new secondary database.
    For single-target replication after the machine hosting the primary database is fixed, the replication process continues as follows:
    1. The primary database must be restored from a backup of the secondary database and then enabled as a target database before secondary replication can begin.
    2. The new target can receive replicated data from the current source.
    3. At a convenient time, you can perform the failback process to return the databases to the roles originally established, with the primary database as the source and the secondary database as the target. You can do this by transitioning the secondary database using the failover command.