19.4.2 Configuring NNMi for HA

This subsection explains the procedure for configuring NNMi for HA and the settings that must be determined at the evaluation stage.

The two distinct phases of configuring NNMi for HA are as follows:

Copy the NNMi data files to the shared disk.

Perform this task on the primary cluster node.
Configure NNMi to run under HA.
- Perform this task on the primary cluster node.
- Perform this task on the secondary cluster node.

The following shows the configuration procedure.

[Figure]

Designate one HA cluster node as the primary NNMi management server. This is the node you expect to be active most of the time. First, configure the primary cluster node, then configure the other node in the HA cluster as secondary cluster nodes.

Important: If you use HTTPS communication to access NNMi servers, you must set up to use a certificate before you verify startup of the primary cluster node. For details, see 10.3.6 Working with Certificates in High-Availability Environments.

Note

You cannot configure NNMi for HA simultaneously on multiple cluster nodes. In a cluster environment, configure nodes for HA one by one, so that the HA setting process on the primary cluster node finishes before the HA setting process on the secondary cluster node starts.
For HA Monitor, you perform the configuration procedure without using nnmhaconfigure.ovpl. For details about the configuration procedure, see the Release Notes.

During failover, the NNMi console is unresponsive. After failover completes, NNMi users must sign in to continue their NNMi console sessions.

For important information about NNMi's TrapReceiver process and how it relates to failovers, 21.16 NNMi NmsTrapReceiver process.

Organization of this subsection

(1) NNMi HA configuration information

(1) NNMi HA configuration information

The HA configuration script (nnmhaconfigure.ovpl) collects information about the NNMi HA resource group. The table below lists and describes the information that you will need for configuring the primary cluster node. Gather this information before you begin the configuration procedure.

You will enter this information interactively during the configuration process by running an HA configuration script (nnmhaconfigure.ovpl). The information requested depends on the OS in use, the type of HA product, and the system configuration. Enter the requested information by following the instructions.

Table 19‒3: NNMi HA primary cluster node configuration information
HA configuration item	Description
HA resource group	The name of the resource group for the HA cluster that contains NNMi. This name must be unique, specific to NNMi, and not currently in use. Example: `nnmtest1` Note: The HA resource group name cannot contain any space characters. Note: The character set and the number of characters permitted for the name depend on the cluster software specifications. For details, see the information provided following this table. Note: Make sure that the HA resource group name is unique among all other resource names and resource group names (including any substring of a resource name or a resource group name). For example, if there is a resource group named `testA`, a name containing any substring of `testA`, such as `test` or `est`, cannot be used. Note: Once the HA resource group has been created, it cannot be renamed. To rename the HA resource group, you must cancel HA configuration for NNMi, and then perform HA configuration again using a new name.
Virtual host name	The name of the virtual host. This is the short name, not the FQDN name containing the domain name. This host name must map to the virtual IP address for the HA resource group. The virtual host short name and the virtual IP address must be resolvable. Note: The virtual host cannot be renamed after the HA configuration has been completed. If you want to rename the virtual host, cancel HA configuration for NNMi, and then perform HA configuration again a new name. Note: If NNMi is unable to resolve the virtual host short name or the virtual host IP address, the HA configuration script could leave the system in an unstable state. Therefore, we recommend that you implement a secondary naming strategy (such as entering the information in the `%SystemRoot%\system32\drivers\etc\hosts` file on the Windows operating system or `/etc/hosts` file on Linux operating systems) in case DNS is not available during NNMi HA configuration.
Virtual host netmask	The subnet mask that is used with the virtual host IP address, which must be an IPv4 address.
Virtual host network interface	The network interface on which the virtual host IP address is running. Examples: Windows: Local Area Connection Linux: `eth0`
Shared file system type	The type of shared disk configuration being used for the HA resource group. The following are the possible values: `disk`: The shared disk is a physically attached disk that uses a standard file system type. The HA configuration script can configure the shared disk. For details, see the File system type entry in this table. `none`: The shared disk uses a configuration other than that described for the disk option, such as SAN or NFS. The HA configuration script configures the shared disk. Note: Because JP1/NNMi does not support the operation when `none` is specified, you must specify `disk`.
File system type	(Linux only) The file system type of the shared disk (when the shared file system type is `disk`). The HA configuration scripts pass this value to the HA product so that it can determine how to validate the disk. The following shared disk formats have been tested: `ext2`, `ext3`, and `vxfs` for VCS or SCS
Disk group	(Linux only) The name of the disk group for the NNMi shared file system. Example: `shdg01`
Volume group	(Linux only) The name of the volume group for the NNMi shared file system. Example: `vg03`
Mount directory (mount point)	The directory location for mounting the NNMi shared disk. This mount point must be consistent among systems (that is, all nodes must use the same name for the mount point). In Windows, specify `drive-letter` or `drive-letter:\any-directory`. The directory name cannot contain any spaces. Example: Windows: `Y:` or `Y:\JP1` Linux: `/nnmmount` Note: NNMi's shared data is stored in the directory named `NNM` that is created immediately below the directory specified here (path of the storage directory is shown below). The storage directory name (`NNM`) is fixed. Windows: `drive-letter:\NNM` or `drive-letter:\any-directory\NNM` Linux: `mount-point/NNM` Note: You will not be able to change the mount point after HA configuration for the relevant NNMi instance is complete. In a Windows environment, you will also not be able to change the drive letter of the mount point after HA configuration for the relevant NNMi instance is complete. If it is necessary to change the mount point, perform the procedure in 19.7 Unconfiguring NNMi from an HA cluster to undo the HA configuration, and then perform the procedure in 19.4 Configuring HA to configure HA again.

The character set and the number of characters permitted for the name of the HA resource group for NNMi must comply with the cluster specifications. The following characters are available to name your HA resource group for NNMi:

In Windows WSFC
- Characters:
  
  Alphabetic characters (a to z, A to Z), numeric characters (0 to 9), hyphen (-), underscore (_), period (.)
- Number of characters: Maximum of 247 characters including the path name %NnmDataDir%hacluster\resource-group
In Linux VCS or Linux SCS
- Characters:
  
  Alphabetic characters (a to z, A to Z), numeric characters (0 to 9), hyphen (-), underscore (_)
  
  The name must begin with an alphabetic character.
- Number of characters: Maximum of 255 characters
In Linux HA Monitor
- Characters:
  
  Alphabetic characters (a to z, A to Z), numeric characters (0 to 9)
  
  The name must begin with an alphabetic character.
- Number of characters: Maximum of 8 characters

To Page Top