Initializing and maintaining the RECON data sets

DBRC records recovery-related information in a pair of key-sequenced data sets (KSDSs) called the RECovery CONtrol (RECON) data set. DBRC uses two RECON data sets to increase availability and recoverability. They contain identical information. The data sets are identified by the DD names RECON1 and RECON2.

If you define only two RECON data sets and an error occurs on one of them during operation, the current jobs continue using the remaining one. New jobs cannot start unless your RECON setting allows a job to start with only one active RECON data set. If you want to continue operations in dual mode, you can define a third RECON data set (RECON3). DBRC does not use this spare data set unless an error occurs on one of the two active RECON data sets. Then, DBRC copies the good RECON to the spare data set (RECON3), which then becomes active (thus maintaining RECON dual-mode operation).

The following figure shows the recommended three RECON data set operating configuration.

Figure 1. DBRC dual RECON set up with a spare
Begin figure description. There are two copies of the RECON data sets (copy1 and copy2) and a spare. All three are viewed as one unit to DBRC. End figure description.

The RECON data sets are critical resources for both DBRC and IMS. If both RECON data sets are lost, DBRC abnormally terminates rather than compromising database integrity. IMS cannot continue processing transactions without a viable RECON data set, so IMS also abnormally terminates.