Administer > Configure NNMi in a High Availability Cluster > Shared NNMi Data in High Availability Environments

Shared NNMi Data in High Availability Environments

This implementation of NNMi running under High Availability (HA) requires the use of a separate disk for sharing files between all NNMi nodes in the HA cluster.

Note NNMi implementations that use Oracle as the primary database also require the use of a separate disk for shared data.

Configuring NNMi for High Availability in an Oracle Environment

  1. If you plan to run Oracle under High Availability (HA), complete that configuration first.
  2. Create an empty Oracle database instance for NNMi.

  3. On the primary NNMi node, install NNMi (including the latest consolidated patch, if any). During installation, do the following:

    1. Select the Oracle database type, and then select Primary Server Installation.
    2. Specify the virtual IP address or hostname for the Oracle HA resource group (if applicable).
  4. On the primary NNMi node, configure NNMi to run under HA as described in Configuring NNMi on the Primary Cluster Node.

  5. Set up the NNMi dependency on the Oracle HA resource group.

    For specific instructions, see the HA product documentation.

  6. On the secondary NNMi node, install NNMi (including the latest consolidated patch, if any). During installation, do the following:

    • Select the Oracle database type, and then select Secondary Server Installation.
    • Specify the virtual IP address or hostname for the Oracle HA resource group (if applicable).
  7. On the secondary NNMi node, configure NNMi to run under HA described in Configuring NNMi on the Secondary Cluster Nodes.
  8. For each additional secondary NNMi node, repeat step 6 and step 7.

Replication of Configuration Files in High Availability Environments

The NNMi High Availability (HA) implementation uses file replication to maintain copies of the NNMi configuration files on all NNMi nodes in the HA cluster.

By default, NNMi manages file replication, copying NNMi configuration files from the active node to a passive node during the failover process. The nnmdatareplicator.conf file specifies the NNMi folders and files included in data replication.

Disabling Data Replication

You can disable data replication as follows:

  1. Edit the following file:

    • Windows: %NnmDataDir%\shared\nnm\conf\ov.conf
    • Linux: $NnmDataDir/shared/nnm/conf/ov.conf

  2. Include the following line:

    DISABLE_REPLICATION=DoNotReplicate

  3. Save your changes.

    Note When you change files (for example, configuration files) on the Active node, these files are automatically replicated to the Standby node on failover.

  4. Restart the NNMi management server:

    Note When making file changes under High Availability (HA), you must make the changes on both nodes in the cluster. If the change requires you to stop and restart the NNMi management server, you must put the nodes in maintenance mode before running the ovstop and ovstart commands. See Maintenance Mode for more information.

    1. Run the ovstop command on the NNMi management server.
    2. Run the ovstart command on the NNMi management server.

Prepare the Shared Disk Manually in High Availability Environments

If the shared disk is of a format that is supported by , the High Availability (HA) configuration script prepares the shared disk, and you can ignore this section. See NNMi High Availability Configuration Information for more information about supported disk formats.

If the shared disk uses a non-tested configuration, such as disk formats supported by the HA product, you must prepare the disk manually. Enter the value none for the file system type during HA configuration, and then configure the shared disk and the NNMi HA resource group’s use of the shared disk.

Tip You can configure the disk before or after configuring the NNMi HA resource group.

To prepare the shared disk manually, follow these steps:

  1. Configure the shared disk as described in Configuring a SAN or a Physically Connected Disk.

  2. Configure the NNMi HA resource group to recognize the disk by completing both of the following procedures:

Configuring a SAN or a Physically Connected Disk

Connecting and formatting a disk that disk into a vxfs or ext3 file system. To configure a SAN or a physically-connected disk, follow these steps:

  1. Verify that the shared disk is not configured to be mounted at system boot time.

    The resource group is responsible for mounting the shared disk.

  2. Connect the device:

    • For a SAN disk, add the SAN device to the network.

      The logical volume on the SAN disk should be in exclusive mode, if that mode is available.

    • For a physically-connected disk, attach the disk using a Y cable.

  3. Add operating system entries to all cluster nodes (disk group, logical volume, volume group, and disk):

    • For a SAN disk, the entries reference the SAN.
    • For a physically-connected disk, the entries reference the disk hardware.
  4. Format the disk using a supported disk format. See NNMi High Availability Configuration Information for more information.

  5. Ensure that the SAN mounts.

    Tip For Linux systems, a reference web site is: http://www.unixguide.net/unixguide.shtml

  6. Unmount and deport the disk.
  7. To test the configuration, add the disk to a resource group and initiate failover.

Setting the High Availability Variables in the ov.conf File

The NNMi High Availability (HA) resource group uses the following variables to access the shared disk:

  • HA_POSTGRES_DIR=<HA_mount_point>/NNM/dataDir/shared/nnm/databases/Postgres
  • HA_EVENTDB_DIR=<HA_mount_point>/NNM/dataDir/shared/nnm/eventdb
  • HA_NNM_LOG_DIR=<HA_mount_point>/NNM/dataDir/log
  • HA_JBOSS_DATA_DIR=<HA_mount_point>/NNM/dataDir/nmsas/NNM/data
  • HA_MOUNT_POINT=<HA_mount_point>
  • HA_CUSTOMPOLLER_DIR=<HA_mount_point>/NNM/dataDir/shared/nnm/databases/custompoller

Tip If you plan to run any NNM iSPIs in the NNMi HA resource group, also set the ov.conf variables for each of those NNM iSPIs. For more information, see the documentation for the appropriate NNM iSPI.

To set the product variables for accessing the shared disk in the ov.conf file, run the following command for each of the preceding variables:

  • Windows:

    %NnmInstallDir%\misc\nnm\ha\nnmhaclusterinfo.ovpl –config NNM –set <variable> <value>

  • Linux:

    $NnmInstallDir/misc/nnm/ha/nnmhaclusterinfo.ovpl –config NNM –set <variable> <value>

Note When making file changes under High Availability (HA), you must make the changes on both nodes in the cluster. If the change requires you to stop and restart the NNMi management server, you must put the nodes in maintenance mode before running the ovstop and ovstart commands. See Maintenance Mode for more information.

Moving the Shared Disk into the NNMiHA Resource Group

Modify the disk configuration file according to the product documentation to move the shared disk into the NNMi HA resource group. For example:

Tip You can also use this process to add other resources, such as a NIC card or a backup disk to the NNMi HA resource group.

  • WSFC: Use Failover Management to add resources to the resource group.
  • VCS: Add disk entries and links to the HA configuration file by using the
    /opt/VRTSvcs/bin/hares command. For example:

  • RHCS:

    /etc/cluster/cluster.conf

A Note about Shared Disk Configuration on Windows Server

Note According to Microsoft Knowledge Base article 237853, dynamic disks are not supported for clustering with Windows Server.

To ensure the correct disk configuration, review the information located on the following web sites:

Send Help Center feedback