Hitachi

JP1 Version 12 JP1/Performance Management User's Guide


9.3.2 Backing up and restoring the event data

Organization of this subsection

(1) Backing up the event data

  1. Log on to the PFM - Manager host.

  2. Execute the jpctool service list command, and make sure that the services have started.

    Make sure that the Name Server, Master Manager, and Master Store services are running.

    For further details on the jpctool service list command, see the chapter that describes the command in the manual JP1/Performance Management Reference.

  3. Execute the jpctool db backup command.

    When Running on a Non-Cluster System:

    For example, to back up the event data for the Master Store service with the service ID PS1001 in the PFM - Manager Store database, execute the command as follows:

    jpctool db backup -id PS1001

    When Running on a Logical Host:

    For example, to back up the event data for the Master Store service with the service ID PS1001 in the PFM - Manager Store database on a logical host, execute the command as follows:

    jpctool db backup -id PS1001 -lhost logical-host-name

    By default, executing this command creates the backup file named PA.DB in the backup directory.

    Table 9‒33: Event data backup files

    Conditions

    Files to be backed up

    Non-cluster system

    PFM - Manager host

    In Windows

    installation-folder\mgr\store\backup\generation-number#\PA.DB

    In UNIX

    /opt/jp1pc/mgr/store/backup/generation-number#/PA.DB

    Cluster system

    Shared disk

    In Windows

    environment-directory\jp1pc\mgr\store\backup\generation-number#\PA.DB

    In UNIX

    environment-directory/jp1pc/mgr/store/backup/generation-number#/PA.DB

    #

    The generation number is assigned in ascending order from 01. The maximum generation number is the value specified in the Backup Save in the jpcsto.ini file. By default, the maximum generation number is 05.

    For further details on the jpctool db backup command, see the chapter that describes the command in the manual JP1/Performance Management Reference. For details on service IDs, see the section that describes service naming rules in an appendix of the JP1/Performance Management Planning and Configuration Guide.

(2) Restoring event data

  1. Log on to the PFM - Manager host.

  2. Execute the jpcspm stop command to stop the PFM - Manager service.

    For details on how to stop the PFM - Manager service, see 1.3 Stopping services. For further details on the jpcspm stop command, see the chapter that describes the command in the manual JP1/Performance Management Reference.

  3. Check the storage location for the backup file.

    For details on the default backup destinations for event data, see Table 9-33 Event data backup files.

  4. Execute the jpctool db restore command.

    When Running on a Non-Cluster System:

    For example, execute the following command to restore the event data in installation-folder\mgr\store\backup\01. (Manager indicates the service key for PFM - Manager):

    jpctool db restore -key Manager -d "installation-folder\mgr\store\backup\01"

    When Running on a Cluster System:

    For example, execute the following command to restore the event data in environment-directory\jp1pc\mgr\store\backup\01 on the logical host jp1-ha1. (Manager indicates the service key for PFM - Manager):

    jpctool db restore -key Manager -d "environment-directory\jp1pc\mgr\store\backup\01" -lhost jp1-ha1

    For further details on the jpctool db restore command, see the chapter that describes the command in the manual JP1/Performance Management Reference.

  5. Execute the jpcspm start command to start the PFM - Manager services.

    Start the PFM - Manager services.

    For details on how to start the PFM - Manager services, see 1.2 Starting services. For further details on the jpcspm start command, see the chapter that describes the command in the manual JP1/Performance Management Reference.

Note:

Restoring event data sometimes causes a delay in restarting the services, because the index of the database has to be rebuilt when the services start up again.