Job Management Partner 1/Base User's Guide

[Contents][Glossary][Index][Back][Next]


3.5.3 Setup

To operate JP1/Base in a cluster system, you must set up a physical host environment (for primary and secondary nodes) and a logical host environment (for primary and secondary nodes). The setup procedure is shown in the following figure.

Figure 3-12 Setup procedure for a cluster system (in UNIX)

[Figure]

Organization of this subsection
(1) Setup on the primary node
(2) Setup on the secondary node

(1) Setup on the primary node

To define the environment on the primary node:

  1. Set user management for the physical host (when running an authentication server on the physical host).
    Specify this option if you want to run an authentication server on the physical host. For details on user management, see 6.3 User management setup (in UNIX).
  2. Modify the event server settings (conf) for the physical host.
    Modify the settings for the event service communication protocol (ports parameter) and retry limit for forwarding JP1 events (forward-limit parameter).
    For the ports parameter, specify the IP address used for the physical host or the name of the physical host.
    The event service stops during failovers. Use the forward-limit parameter to specify the maximum period to retry forwarding the JP1 events that could not be sent during a failover. By default, the system continues to retry for 3,600 seconds.
    The event server settings file (conf) resides in the following location:
    /etc/opt/jp1base/conf/event/servers/default/
     
    The following shows an example of the parameter settings:
    ports IP-address-of-physical-host jp1imevt jp1imevtapi
    forward-limit 3600
     
    For details on the event server settings file, see Event server settings file in 14. Definition Files.
  3. Set the logical host.
    Execute the command as follows: Create a shared directory and shared file on a shared disk to set up the authentication server.
    jp1base_setup_cluster -h node0 -d /shdsk/node0 -a node0 -s 
     
    For details on the jp1base_setup_cluster command, see jp1base_setup_cluster (UNIX only) in 13. Commands.
  4. Set user management function for the logical host.
    If you have specified a logical host as an authentication server, register the JP1 users, set up user mapping, and set up the operating permissions of the JP1 users as follows:
    • Register the JP1 user in the common definition information (only when using the logical host as the authentication server).
      Make sure that the authentication server is active, and then execute the following command to register a JP1 user:
      jbsadduser -h logical-host-name JP1-user-name
      To check the registered JP1 user, execute the following command:
      jbslistuser -h logical-host-name
    • Register the user mapping information in the common definition information.
      The user mapping definition file (jp1BsUmap.conf) resides in the following location:
      shared-directory/jp1base/conf/user_acl/jp1BsUmap.conf
      After editing the file (jp1BsUmap.conf), execute the following command to register the user mapping definition information:
      jbsmkumap -h logical-host-name
      To check the registered user mapping information, execute the following command:
      jbsgetumap -h logical-host-name
    • Match the common definition information on the physical hosts.
      When you finish these operations, match the information on all the physical hosts, as described in 3.5.5 Modifying common definition information.
    • Set JP1 user operating permissions (only when using the logical host as the authentication server).
      The user permission level file (JP1_UserLevel) is located in the following directory:
      shared-directory/jp1base/conf/user_acl/JP1_UserLevel
      After editing this file (JP1_UserLevel), execute the jbsaclreload command to apply the settings.
    For details on setting user management, see 6.3 User management setup (in UNIX).

    Notes on operating authentication servers in a cluster system:
    The settings files for authentication servers are stored in the following directory.
    shared-directory/jp1base/conf/user_acl/
    If you are using a secondary authentication server, you must copy the settings files from the primary authentication server to the secondary authentication server. Note that the copy destination varies depending on whether you use the secondary authentication server in a cluster system:
    When using a cluster system:
    shared-directory/jp1base/conf/user_acl/
    When not using a cluster system:
    /etc/opt/jp1base/conf/user_acl/
    After copying the settings files, execute the following command to apply the settings. You need to specify the -h option only if you use the secondary authentication server in a cluster system.
    jbs_spmd_reload -h logical-host-name

    Notes when not operating authentication servers in a cluster system:
    If you omit the -s option when executing the jp1base_setup_cluster command, the authentication server process will not start for the JP1/Base instance running on that logical host.
    By changing the configuration settings, you can start the authentication server process after executing the jp1base_setup_cluster command.
    Follow these steps:
    1. Stop JP1/Base.
    Stop the logical host whose configuration you are changing and all programs dependent on JP1/Base on that logical host.
    2. Modify the definition file.
    Execute the following command to change the JP1/Base process definition file:
    cd /shared-directory/jp1base/conf
    cp -p jp1bs_spmd.conf.session.model jp1bs_spmd.conf
    3. Restart JP1/Base.
    Restart the logical host whose configuration you changed and the programs dependent on JP1/Base on that logical host.
    The changed definition takes effect when you restart JP1/Base.
  5. Modify the event server settings (conf) for the logical host.
    Modify the settings for the event service communication protocol (ports parameter) and retry limit for forwarding JP1 events (forward-limit parameter).
    For the ports parameter, specify the IP address used for the logical host or the name of the logical host.
    The event service stops during failovers. Use the forward-limit parameter to specify the maximum period to retry forwarding the JP1 events that could not be sent during a failover. By default, the system continues to retry for 3,600 seconds.
    The event server settings file (conf) resides in the following location:
    shared-directory/event/
     
    The following shows an example of the parameter settings:
    ports IP-address-of-logical-host jp1imevt jp1imevtapi
    forward-limit 3600
     
    For details on the event server settings file, see Event server settings file in 14. Definition Files.

This completes JP1/Base setup on the primary node.

If any of the programs that require JP1/Base (JP1/IM, JP1/AJS, or JP1/Power Monitor) is installed, you must complete the failover settings for these programs. For details, see the Job Management Partner 1/Integrated Management - Manager Configuration Guide, Job Management Partner 1/Integrated Management - Manager Administration Guide, Job Management Partner 1/Automatic Job Management System 2 Planning and Administration Guide, Job Management Partner 1/Automatic Job Management System 3 System Design (Configuration) Guide, Job Management Partner 1/Automatic Job Management System 3 System Design (Work Tasks) Guide, Job Management Partner 1/Automatic Job Management System 3 Administration Guide, and the Job Management Partner 1/Power Monitor Description, User's Guide and Reference.

(2) Setup on the secondary node

Before you start to set up on the secondary node, make sure that you complete the setup tasks for JP1/Base, JP1/IM, JP1/AJS, and JP1/Power Monitor on the primary node. To set up on the secondary node, perform the following procedure:

  1. Set user management for the physical host (when running an authentication server on the physical host).
    Specify this option if you want to run an authentication server on the physical host. For details on user management, see 6.3 User management setup (in UNIX).
  2. Modify the event server settings (conf) for the physical host.
    Modify the settings for the event service communication protocol (ports parameter) and retry limit for forwarding JP1 events (forward-limit parameter).
    For the ports parameter, specify the IP address used for the physical host or the name of the physical host.
    The event service stops during failovers. Use the forward-limit parameter to specify the maximum period to retry forwarding the JP1 events that could not be sent during a failover. By default, the system continues to retry for 3,600 seconds.
    The event server settings file (conf) resides in the following location:
    /etc/opt/jp1base/conf/event/servers/default/
     
    The following shows an example of the parameter settings:
    ports IP-address-of-physical-host jp1imevt jp1imevtapi
    forward-limit 3600
     
    For details on the event server settings file, see Event server settings file in 14. Definition Files.
  3. On the primary node, execute the jbsgetcnf command.
    Execute the following command on the primary node. This command saves the common definition information to the backup file:
    jbsgetcnf -h logical-host-name > backup-file-name
  4. Copy the backup file to the secondary node.
  5. On the secondary node, execute the jbssetcnf command:
    Execute the following command on the secondary node. In backup-file-name, specify the backup file created by the jbsgetcnf command:
    jbssetcnf backup-file-name
  6. Set the logical host.
    Execute the command as follows:
    jp1base_setup_cluster -h node0
    For details on the jp1base_setup_cluster command, see jp1base_setup_cluster (UNIX only) in 13. Commands.

This completes JP1/Base setup on the secondary node.

[Contents][Back][Next]


[Trademarks]

All Rights Reserved. Copyright (C) 2009, Hitachi, Ltd.