Overview

Adabas Audit Data Retrieval archives lists and their indexes in archive datasets. You can define whether you want to archive data in single or multiple archives. You can also define on which media data is to be archived (archive media) and how long archived data is to be kept (archive retention period).

Adabas Audit Data Retrieval compresses archived data, which reduces archive storage space by up to 70 percent. At the same time, compression also protects the archived data against unauthorized access.

Note: Archiving to disk requires SMS.

Attributes of archive datasets

Archive datasets are always allocated with the following attributes (irrespective of the device used):

• RECFM: V
• LRECL: 32756
• BLOCKSIZE: 32760

Important: Adabas Audit Data Retrieval uses hardware pointers for efficient positioning during reload. The pointers that are stored in the database must be updated when creating copies of archive datasets. To copy archive datasets, therefore always use the utility BST08OCP.

What is required for a list to be archived?

The Adabas Audit Data Retrieval archive batch utility will archive a list if the following is true:

• The list has been marked for archiving. Lists can be marked for archiving in the following ways:
  • Automatically, when reading in the list, which requires that the Adabas Audit Data Retrieval list definition specifies Yes in the Automatic Archive field
  • Manually, using the line command A

  Note: The list generation record contains the information required for archiving (archive medium, archive retention period, and owner).

• A valid Adabas Audit Data Retrieval archive pool exists, whose definition matches the archive medium, the archive retention period, and the owner of the list.

Single or multiple archiving

Each archive pool has one or several archive subpools. The number of archive subpools determines whether a single or multiple archive is created.

Each list will be archived in all archive subpools of the corresponding archive pool.

The archive subpools of an archive pool can use the same or different archive media and they can have the same or different archive retention periods.

When does archiving take place?

The Adabas Audit Data Retrieval archive batch utility archives all lists that have been marked for archiving. Archiving takes place at the next run of the archive batch utility and is independent of the remaining online retention period of the list.

A list to be archived has the archive status Pend. After the list has been archived successfully, the archive status of this list is changed to Yes.

The archive batch utility should run as part of the regular Adabas Audit Data Retrieval daily maintenance. The job should run before the Adabas Audit Data Retrieval cleanup jobs.

Grouping lists for archiving

Each list is assigned to an archive pool at the run time of the archive batch utility and is then archived in all subpools pertaining to this archive pool. Assigning lists to archive pools is based on the following three criteria:

• Archive retention period
• Archive media
• Owner

Matching archive pool

In order for the archive pool to come into question for the assignment, it must fulfill the following conditions:

• The archive pool definition has the same archive medium as the list.
• The archive pool definition has the same owner as the list.
• The archive pool definition has the same or a higher archive retention period as the list.

If several archive pool definitions fulfill these conditions, the list is assigned to the archive pool whose archive retention period matches best.

The list will be archived in all archive subpools pertaining to the best matching archive pool.

Assigning lists to archive pools

The archive batch utility assigns a list having a given archive retention period and archive medium to an archive pool as follows:

• First, it tries to assign the list to an archive pool having the same archive retention period, the same archive medium, and the same owner.
• If this is not possible, it tries to assign the list to the archive pool having the same archive medium, the same owner, and the next higher archive retention period.
• If this is not possible, the list is not assigned to any archive pool and the archive status of the list remains Pend (Archive pending). The archive batch utility sets RC=8, and includes the name of the list in the error log.

Archive expiration date

The archive expiration date results from:

• The date of the archive run
• The archive retention period of the list
• The start date entered in the archive subpool (optional)

Examples:

• A list to be archived, which has an archive retention period of 1095 days, is archived in an archive subpool on 26.9.2018. The list receives the archive expiration date 25.9.2021.
• A list to be archived, which has an archive retention period of 1095 days, is archived in an archive subpool on 26.9.2018. It is determined in the archive pool that the calculation is made on the basis of the start date 01.01. The list receives the archive expiration date 31.12.2021.

Valid archive pool and subpool definitions

After program start, the archive batch utility checks the validity of all archive pool definitions.

The following must be true for an archive pool definition to be valid:

• The archive pool has at least one subpool whose archive retention period and archive medium is identical to the archive retention period and archive medium of the archive pool.
• If the archive pool has additional archive subpools, the archive retention period of these archive subpools may not be longer than the archive retention period of the archive pool.

If one of these conditions is not true, the corresponding archive pool definition is invalid and the archive batch utility does the following:

• If no lists are assigned to this invalid archive pool, the archive batch utility sets RC=4 and logs the invalid archive pool.
• If any lists are assigned to this invalid archive pool, the archive batch utility sets RC=8. The lists assigned to this archive pool are not archived during this run of the archive batch utility and they keep the archive status Pend (archive pending). The archive batch utility logs the invalid archive pool and the names of the lists.

Rules for archiving to tape and optical disk

The following rules will ensure that the data of different customers will be archived on separate media and will prevent the fragmentation of archive media after individual archive datasets have expired.

The following applies when archiving to tape and optical disk:

• Only archive datasets of one subpool will be written to one volume.

In addition, the following applies when archiving to optical disk:

• Only archive datasets of the same owner will be written to one platter.