16.2.1 Prerequisites for setting up application failover
For application failover to work successfully, the NNMi management servers must meet the following requirements:
-
Only a configuration in which NNMi is used alone is supported.
A configurations in which NNMi is linked to another related program product, such as JP1, is not supported. In such a case, use an HA configuration using cluster software.
-
It must be possible to resolve the host name and IP address of the active server and of the standby server at both NNMi management servers.
-
Both NNMi management servers must be running the same type of operating system. For example, if the active server is running the HP-UX 11iV3 operating system, the standby server must also be running the HP-UX 11iV3 operating system.
-
Both NNMi management servers must be running the same NNMi version. The NNMi patch levels must also be the same on both servers. For example, if NNMi 10-10-01 is running on the active server, the identical NNMi version, NNMi 10-10-01, must be on the standby server.
-
The system password must be the same on both NNMi management servers.
-
For an installation on Windows, NNMi must be installed in the same directory in both NNMi management servers and the %NnmDataDir% and %NnmInstallDir% system variables must be set to identical values on both servers.
-
Both NNMi management servers must have identical licensing attributes (number of nodes managed, and whether NNMi or NNMi Advanced is used). For example, the node counts and licensed features must be identical.
- Important note
-
The standby server must also have the identical license.
-
Do not enable application failover until NNMi is in an advanced stage of initial discovery. For details, see 4.4 Evaluating discovery.
-
For application failover to function correctly, the active and standby servers must have unrestricted network access to each other. Any software that locks files or restricts network access can cause NNMi communication problems. Configure these applications to ignore the files and ports used by NNMi.
-
We do not recommend that you set up a firewall between the active and standby servers. When setting up a firewall, configure the systems such that both servers can use all ports for communication.
-
To run a firewall in an NNMi management server, permit communication between processes in the local server and communication with remote servers at all ports. Application failover performs communication dynamically using any available port.
-
If you are using a firewall that permits communication on a process-by-process basis (such as Windows Firewall), permit communication by the cluster manager (nnmcluster.exe).
-
If you are using a firewall that permits communication on a port-by-port basis, permit the following types of communication:
IP addresses: All IP addresses assigned to the local server and remote servers
Ports: All ports
-
-
If you plan to integrate NNMi with a directory service through LDAP, and both NNMi management servers have had their passwords encrypted prior to configuring application failover, make sure that the ldap.properties file has been copied to the secondary NNMi management server.
This copying is normally performed after the secondary NNMi management server joins the cluster for the first time.
To determine that the copying was successful, enable the cluster and wait five minutes; then check the ldap.properties file to make sure it was copied into the secondary NNMi management server. For details, see 16.3 Configuring NNMi for application failover.
-
The same password must be set for the NNMi databases on both the active and standby servers.
If you have changed the password for an NNMi database, set the same password on all servers before you start configuring application failover.
If all these conditions are satisfied, complete the steps shown in 16.3 Configuring NNMi for application failover. For details, see C. List of Ports Used by NNMi.