Oracle7 Server Distributed Systems Volume I: Distributed Data

Failure Resolution Transparency and the RECO Background Process

Network or computer hardware failure is always possible. Oracle's distributed system architecture guarantees that if a network or system failure occurs during the commit of a distributed transaction, the transaction is automatically and transparently resolved globally. When the network or system is restored, either all nodes commit, or all roll back.

The RECO background process automatically recovers pending or in-doubt distributed transactions (transactions not committed or rolled back when a network failure occurs during a two-phase commit process). When a network failure is corrected, the RECO processes of the involved database servers automatically resolve the outcome of any pending distributed transactions. No work is required of the database administrator to resolve pending transactions, although the database administrator may force resolution of in-doubt transactions if he cannot wait for the network failure to be corrected. For information about failure resolution of in-doubt distributed transactions in an Oracle distributed system, see the Oracle7 Server Concepts manual.