CTTRCV – Recover the Media database using trace file

The CTTRCV utility recovers damaged Media database (MDB) contents according to trace data recorded during MDB update operations. The utility can perform either a physical or a logical recovery.

• Physical Recovery (Roll Forward)

  Physical recovery assumes that the last available backed up MDB was restored to disk, as described in the topic CTTTRB – Synchronize Trace file and Media database backup. From this point forward, recovery is performed according to trace data.

  Note: By default, the Control-M/Tape daily job uses the DFSMSdss to back up the MDB. In this case, you must use the DFSMSdss RESTORE command to restore the backed up MDB.

  Trace records are read starting with the first record after the Media database backup record. Recovery continues until the last trace record is reached. When the process is finished, the MDB returns to the condition it was in before it was damaged.

  In situations where no MDB backup has been performed, the recovery can be started from the first trace record.

  To prevent unpredictable results, the Control-M/Tape environment must be inactive while the physical recovery process is running.

• Logical Recovery (Roll Back)

  Logical recovery rolls back selected changes from the current MDB. This type of recovery can be performed when a user modifies the status of one or more MDB records and then discovers that these modifications must be reversed. A good example of this scenario is when a job is run that scratches a series of active volumes or moves a series of volumes to a vault incorrectly.

  When this type of recovery is used, the trace records are read backward (that is, starting from the last record towards the first record). The improper modifications are reversed in a way that restores the MDB to the condition it was in prior to the improper modifications.

  To maintain the logical consistency of the MDB, the CTTRCV utility verifies that the current contents of the MDB match the tracked information in the Trace file.

  For example, if the status of a volume was changed from Active to Scratch by TSO user ABC and a logical recovery is performed on all operations performed by user ABC, the utility verifies that the current volume status in the MDB is Scratch before the recovery reverses the operation.

  Note: You can bypass this check by setting FORCE to either YES or LOGICAL. However, this may cause unpredictable results.