Chapter 15. Conflict Avoidance and Resolution Techniques

Data integrity and consistency are perhaps the most significant challenges for the administrator of an advanced replication environment. Since users can perform DML on a given table in multiple Oracle instances, the administrator’s responsibility expands from guaranteeing data integrity locally to ensuring data convergence globally. For example, if two users at two sites update an employee’s salary to two different values, how do we determine which value to accept and how do we ensure that the correct value is propagated to all sites that have the replicated table? It can be done, and Oracle provides a variety of built-in conflict resolution handlers, but to use these techniques successfully, developers and administrators must understand and anticipate all likely scenarios that would result in conflicts. They also must understand how Oracle replicates DML and consider the limitations of the conflict handlers.

Data Integrity Versus Data Convergence

Data integrity refers to data that is consistent with the constraints that are defined for it. These constraints may be referential integrity constraints. For example, the value of the Po_Num field for records in table LINE_ITEMS might be restricted to values of Po_Num that exist in the PURCHASE_ORDERS table. Data also may be restricted to ranges that are independent of other tables; a gender field may be restricted to the values M and F. Other examples of integrity constraints include ...

Get Oracle Distributed Systems now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.