Recovery of data sets with RUV

Recovery processing after a data loss can mean the difference between operating at maximum efficiency and not operating at all. For users of CICS, VSAM data loss creates a need to recover large amounts of mission-critical data. The process of recovering CICS VSAM data is complex, and few qualified products exist to accomplish the task.

The RUV product provides quick and easy VSAM data recovery processing. RUV allows you to focus on the process of recovering data rather than the program that recovers the data.

You can use RUV to recover VSAM data sets in the following situations:

• Physical VSAM data is lost or damaged, which may occur for the following reasons:
  • A hardware failure occurred.
  • A part of a DASD is not readable.
  • The data was deleted.
  • A natural disaster, such as fire or storm, occurred.
• Logical data integrity is lost, which may occur for the following reasons:
  • An application error occurred.
  • A failure occurred during CICS transaction backout.

Loss of logical data integrity occurs when a CICS application incorrectly updates your VSAM spheres. CICS transaction backout maintains logical data integrity by backing out partially completed transactions from VSAM data sets. Transaction backouts can fail under some conditions. For example, an I/O error can occur when CICS accesses the VSAM data set. In this type of transaction backout failure, RUV recovers your damaged VSAM data sets and all updates that were made before the problem occurred.

The recovery process is a sequence of tasks that begins with the last usable backup, and may include various logs, journals, and archival data sets. The recovery process is complete when data is recovered and processing capability is restored. The following figure shows how the recovery process ensures business continuity of the data processing environment:

As shown in the previous figure, actual loss of data may have occurred during a very short span of time. The time needed to completely recover data sets, however, which includes the elapsed disruption time, is a long span of time.

RUV recovers VSAM data sets in a forward mode or a backout mode.

Data set structures

RUV processes the following VSAM data set structures and organizations:

• VSAM key-sequenced data sets (KSDS)
• VSAM entry-sequenced data sets (ESDS)
• VSAM fixed and variable length relative-record data sets (RRDS)

The various data set structures are generically referred to as VSAM spheres. A VSAM sphere consists of the base cluster and any alternate index paths that are associated with the base cluster. RUV identifies a VSAM sphere by the name of the base cluster.

Recovery process

Recovery of VSAM data sets is an automated process within the RUV environment. RUV collects and records recovery information; however, it is your responsibility to determine when recovery processing is needed and which data sets require recovery.

As shown in the following figure, RUV can gather and coordinate information from many sources, including z/OS log streams, CICS journals, RPCV journals, RUV batch journals, and RUV backups.

The recovery information is recorded in the RUV repository. With this recorded information, recovery is simply a matter of selecting the VSAM data sets to be recovered. RUV automatically selects which backups and archives are needed and constructs the correct recovery JCL.