Upgrading BMC ProactiveNet Server in HA mode on Linux/Solaris


This topic provides instructions to upgrade to BMC ProactiveNet Server 9.6.00 in high availability mode on Linux/Solaris operating systems.

The following applications are required to control high-availability for BMC ProactiveNet on Linux/Solaris operating systems:

  • Sun Cluster software for Solaris
  • Red Hat Availability add-on for Linux (with Resilient Storage add-on if GFS2 is required)
Warning

Note

BMC ProactiveNet is not supported on VMware with VMWare Fault Tolerance (FT) as BMC ProactiveNet high availability support.

Before you begin

  • Ensure that the cluster configuration is of node and cluster disk majority type.
  • Ensure that a visible quorum disk is available.
  • Ensure that the HA installer files are not copied to the shared disk.
  • Ensure that Microsoft Clustering is installed on the target computer.
  • Ensure that before you upgrade, you have migrated the customized KBs in your BMC ProactiveNet 9.6 implementation. Follow the procedure for executing the KB Migration Wizard described in Cell configuration migration (BMC ProactiveNet cells. When satisfied that the merged KB is working properly, you can begin the server upgrade procedure. Note the directory path of each migrated KB. The directory path should follow this hierarchy structure: /topLevelDirectory/mergedKB/serve/etc/cellName, where /server/etc/cellName is the standard subdirectory.
    For Windows 2012, ensure that the PowerShell execution policy is running in the unrestricted mode.

    Warning

    Note

    When you upgrade BMC ProactiveNet on a 64-bit operating system, the JServer, Rate, and Agent Controller processes use 64-bit JRE, and the Self-monitoring Service process uses 64-bit JRE on Windows and 32-bit JRE on Solaris. If you have already configured the 64-bit JRE processes in an earlier version of BMC ProactiveNet, they are retained.

Upgrading to BMC ProactiveNet Server in HA mode on Linux/Solaris systems

The following procedure explains how to upgrade the BMC ProactiveNet Server in HA mode on primary and secondary servers:

  1. Archive the old installation files including the user apps, PERL libraries, and any other customizations.
    1. Run the pw system stop command to stop the BMC ProactiveNet Server.
    2. Tar the /usr/pw directory and place the resulting file in a secure location such as on another server, another file partition, or a directory that lies above /usr/pw.
    3. Save the Crontab entries (especially the database archive entry) related to the BMC ProactiveNet Server.

      Warning

      Note

      For more information about archiving your old installation, check with your Solaris administrator.

    4. Restart the BMC ProactiveNet Server by using the pw system start command.
    5. Save any custom Apache settings.
  2. Change the host name of the primary node to the logical host name configured for the cluster by entering hostname logicalHostName.  
  3. On Solaris, change the ip address of the primary node to the logical IP address configured for the cluster by entering the following commands:

    ifconfig <alias interface> plumb

    example: ifconfig hme0:1 plumb

    ifconfig <alias interface> <logical ip> netmask <netmask> up

    example: ifconfig hme0:1 10.129.5.252 netmask 255.255.252.0 up 

  4. Install BMC ProactiveNet Server on the primary node.
    During the installation, choose the cluster file system (/global) that is shared on both nodes as the installation location.
    The installation is completed successfully.
  5. After successful installation of the BMC ProactiveNet Server, change the logical host name back to the host name of the primary node by entering hostname  
    primaryHostName.  
  6. On Solaris, after successful installation of the BMC ProactiveNet Server, change the logical IP address back to the IP address of the primary node by entering the following command from a command line:

    ifconfig <alias interface> down

    example: ifconfig hme0:1 down

    ifconfig <alias interface> unplumb

    example: ifconfig hme0:1 unplumb 

  7. On the primary node, remove the following files so that the BMC ProactiveNet application does not start:
    • /etc/rc3.d/S81BMCProactiveNet
    • /etc/rc0.d/K09BMCProactiveNet
    • /etc/rc1.d/K09BMCProactiveNet
  8. On the standby node, create the required links for the BMC ProactiveNet Server by entering the following commands:
    • ln –s /serverInstallDirectory/pw /usr/pw

    • ln –s /usr/pw/pronto /usr/pronto

    • ln –s /usr/pw/pronto/bin/pw /bin/pw

For information about adding a failover domain and other post-installation tasks, see Setting-up-the-cluster-in-Linux and Setting-up-the-service-to-run-in-high-availability-mode-in-Linux.

Related topics

Upgrading BMC ProactiveNet Server in HA mode on Windows

Configuring-and-using-BMC-ProactiveNet-in-high-availability-mode

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

BMC ProactiveNet 9.6