Customers have this question about the feasibility of running a peer to peer (master-master) replication without having to deploy the conflict resolution routines that require PL/SQL stored procedures to be configured, and thus need expertise in that area, or do not have time to configure such procedures.
General information.
If not intending to deploy conflict resolution routines in a peer to peer replication environment, then the only way to be able to do so without incurring any out of sync is to point applications (that make data changes) to only one of the nodes at any given time. This will keep all the nodes in sync at all times unless changes happen simultaneously at multiple nodes and may result in conflicts. Plus any of the nodes will be ready to act as source at any time so long as there are no data in the Post queue on the node that is to assume the role of primary.
Many customers configure their peer to peer replication this way but are aware of the caveat involving conflicts.
