17.7.1 Procedures for recovering from serious failures such as disk failures
Use the following procedure to restore Performance Management when there is a serious error related to a disk failure.
- Organization of this subsection
(1) Uninstallation
Uninstall the Performance Management program.
For details on how to uninstall the Performance Management program, see the chapter that describes installation and setup in the JP1/Performance Management Planning and Configuration Guide.
(2) Reinstallation
Reinstall the Performance Management program.
For details on how to reinstall the Performance Management program, see the chapter that describes installation and setup in the JP1/Performance Management Planning and Configuration Guide.
(3) Setup
Set up the Performance Management program again. During setup, use the instance name and physical host name used in the environment before the error occurred.
For details on how to set up the Performance Management program, see the chapter that describes installation and setup in the JP1/Performance Management Planning and Configuration Guide.
(4) Restoring definition information
Restore the backed-up service definition information.
For details on the service definition information to be restored, see 9.2 Backing up and restoring definition information.
(5) Restoring performance data and event data
Restore the backed-up Store database that contains performance data and event data.
For details on restoring the Store database, see 9.3 Backing up and restoring operation monitoring data.
(6) Starting services
Start the Performance Management program services and confirm that the services start normally.
For details on starting the services, see 1.2 Starting services. Use the jpctool service list command to check the service status. For details about checking the service status, see 1.6 Checking the status of services.
(7) Restoring the report and alarm table definition information
Restore the backed-up report and alarm table definition information.
For details about restoring the report definition information, see 9.2.8 Backing up and restoring a report definition. For details about restoring the alarm definition information, see 9.2.9 Backing up and restoring an alarm definition.
(8) Checking operations
Lastly, make sure that the trouble has been resolved. Check whether the following items are normal:
-
Check whether performance data can be collected
Run Performance Management for more than twice the length of the collection interval for performance data to confirm that performance data can be collected without a problem.
For details on the collection interval of performance data, see 4.1.1 Modifying the recording options for performance data.
-
Check whether there is a problem with the data in the Store database
Export the data of the Store database to a text file and check whether there is a problem with the data. Use the jpctool db dump command to export the data in the Store database to a text file.
For details on the jpctool db dump command, see the chapter that describes commands in the manual JP1/Performance Management Reference.
-
Check the report and alarm definitions
Check whether there is a problem with the report and alarm definitions. Use PFM - Web Console to check the report and alarm definitions.
For details on the report definition, see 5. Creation of Reports for Operation Analysis. For details about the alarm definition, see 6. Monitoring Operations with Alarms.
-
Check the binding of the alarm table
Check the binding of the alarm table and bind it as necessary.
For details on binding the alarm table, see 6.6.1 Changing the association between an alarm table and a monitoring agent.