Top
PRIMECLUSTER  Installation and Administration Guide 4.4
FUJITSU Software

9.2.1 Changing the IP Address of the Public LAN

The following describes how to change an IP address when the IP address of the public LAN has been changed.

Operation Procedure:

  1. Execute the following command on one of the cluster nodes to stop RMS operation.

    # hvshut -a
  2. Execute the following commands on all the nodes to start the system in single-user mode.

    [For RHEL6]

    # /sbin/shutdown now

    [For RHEL7]

    # /usr/bin/systemctl set-default rescue.target
    # /sbin/shutdown -r now
  3. Mount the local file system on all the nodes.

    (Example)
    # /bin/mount -a -t ext3
  4. Edit the "/etc/hosts" file, and change the IP address on each node.

  5. Change the IP address of the public LAN.

    For details on how to change the IP address, see the Linux documentation.

  6. If the administrative LAN is shared with the public LAN, execute the following command on each node to save rcsd.cfg.

    # mv /etc/opt/SMAW/SMAWsf/rcsd.cfg /etc/opt/SMAW/SMAWsf/rcsd.cfg.work
  7. If an IP address used by Web-Based Admin View also needs to be changed as the IP address of the public LAN is changed, change it on each node.

    See

    For details, see "7.1 Network address," "7.3 Management server," and "7.5 Multi-network between server and client by classified use" in "PRIMECLUSTER Web-Based Admin View Operation Guide."

  8. If a takeover IP address must be changed (when the takeover IP address is changed after installation, or when the takeover IP address is changed due to transfer of the node), correct the IP address being used as the takeover IP address in the "/etc/hosts" file of each node.

    When you have created takeover network resources, and change the subnet mask value due to the change of the public LAN, you also need to edit the /usr/opt/reliant/etc/hvipalias file.

    See

    For information on how to edit the /usr/opt/reliant/etc/hvipalias file, see "6.7.3.6 Setting Up Takeover Network Resources."


  9. If a Gls resource is created, restrict the automatic startup of RMS so that RMS does not start automatically when the node is restarted.

    Check the settings of the current automatic startup of RMS and execute the following command according to the settings.

    # hvsetenv HV_RCSTART
    1 <- Check this value.
    • If "0"is set, the automatic startup of RMS has been restricted. Go to Step 10.

    • If "1" is set, execute the following command to restrict the automatic startup of RMS.

      # hvsetenv HV_RCSTART 0
      # hvsetenv HV_RCSTART
      0 <- Check "0" is output.
  10. Restart OS on all the nodes.

    [For RHEL6]

    # /sbin/shutdown -r now

    [For RHEL7]

    # /usr/bin/systemctl set-default multi-user.target
    # /sbin/shutdown -r now
  11. If the takeover IP address is defined by GLS, see "PRIMECLUSTER Global Link Services Configuration and Administration Guide: Redundant Line Control Function."

  12. If you change the settings that restrict the automatic startup of RMS in Step 9, return the file back to its previous settings.

    # hvsetenv HV_RCSTART 1
    # hvsetenv HV_RCSTART
    1 <- Check "1" is output.

    Note

    If a takeover IP address has been defined by GLS and if the takeover IP address is specified directly with an IP address value, not a host name, in the GLS environment settings and the Gls resource settings, first delete the Gls resource. Then change /etc/hosts and the GLS environment settings, and re-register the Gls resource.

    For details, see "6.2.1 GLS Setup" and "6.7.3.5 Setting Up Gls Resources" in this manual, and "PRIMECLUSTER Global Link Services Configuration and Administration Guide: Redundant Line Control Function."

  13. If rcsd.cfg is saved in Step 6, execute the following command.

    # mv /etc/opt/SMAW/SMAWsf/rcsd.cfg.work /etc/opt/SMAW/SMAWsf/rcsd.cfg

    After that, set the shutdown facility again.