Overview of Recovery Sensor
You can use Stand-alone Recovery Sensor to collect sensor data from the content of DBRC RECON data sets.
Recovery Sensor runs as a batch job step of z/OS®.
Support for recovery autonomics
The Recovery Sensor is a critical component in supporting passive recovery autonomics for IMS database backup and recovery preparedness provided in IMS Recovery Solution Pack V2.1 and later.
The data collected by Recover Sensor is used by Autonomics Director to monitor and maintain the image copy, change accumulation, and recoverability of the database.
Basic function
Recovery Sensor scans the content of DBRC RECON data sets and collects database statistics including data about individual databases, HALDB partitions, FP Areas, and Change Accumulation Groups. IMS Recovery Expert System Level Backups (SLB) can also used to discover available image copies.
Then it organizes the statistics into sensor data, and stores the sensor data in the IMS Tools Knowledge Base Sensor Data repository in the form of records that are comprised of data elements. When Recover Sensor completes storing the sensor data in the repository, it notifies the Autonomics Director server that the sensor data is available so that the new data can be immediately analyzed and evaluated if necessary.
Policy, rule, and repository support for the recovery domain used by recovery autonomics is provided by the Policy Services and IMS Tools Knowledge Base components of IBM® IMS Tools Base 1.6 and later.
For a list of the data elements that are collected by Recovery Sensor, see the topics about data elements in the IMS Tools Base Policy Services User’s Guide.
During the job, Recovery Sensor prints a runtime summary report to the IROSPRT DD. The report contains the extracted sensor values in table format and is designed for technical analysis only.
Basic characteristics
Recovery Sensor has the following characteristics:
- Recovery Sensor can process online IMS databases
- One Recovery Sensor job can process multiple databases and partitions.
- Recovery Sensor can process both offline DEDB areas and online DEDB areas
- One Recovery Sensor job can process multiple DEDB area data sets
The maximum number of area data sets is 2048
- Recovery Sensor uses high performance I/O techniques to reduce the amount of time that is required to scan area data sets
- For HALDB, Recovery Sensor processes only the active side (A or M side) of data sets.
How sensor data is used in the autonomics process
Sensor data is the information collected by a sensor utility when, at an instance in time, it scans one or more IMS database environments and measures the specified conditions (or states) occurring in those environments.
Fast Path and Full-Function Sensor utilities function in an IMS Tools space management and reorganization domain and capture statistics about the characteristics and organization of data in each database. Information is also collected from the system catalog, VSAM catalog, and Volume Table of Contents (VTOC).
The Recovery Sensor utility functions in an IMS Tools recovery domain and captures information from the content of DBRC RECON data sets. Data is gathered about individual databases, HALDB partitions, Fast Path areas, and change accumulation groups.
Sensor data is stored in the IMS Tools Knowledge Base Sensor Data repository in the form of records made up of data element values. The data record is stored in a well-understood and flexible format that allows its use for years and multiple product releases later in time. The data and its format is understandable between products and releases to ensure reliable functionality. The data can then be used for later database analysis and tuning purposes.
Autonomics Director jobs can use sensor data to monitor and maintain the health, performance, and recoverability of the database. In the Autonomics Director jobs, Policy Services evaluation is internally called to evaluate the database state based on the stored sensor data.
When the jobs end, you can use the Management Console to view graphical visualization and charting of the sensor data, the exceptions that were detected by the policy evaluations, and the recommendations for resolving the exceptions.