18.7.2 Starting, stopping, and restarting NNMi
- Organization of this subsection
(1) Starting and stopping NNMi
When you use the ovstop or ovstart command to stop or start NNMi configured for application failover, NNMi actually runs nnmcluster commands. For details about the actual commands that run, see 18.4.3 ovstart and ovstop commands used on NNMi management servers configured for application failover.
With NNMi configured for application failover, whether a server becomes the active or standby server is adjusted automatically during the startup. The server on which the nnmcluster command runs first becomes the active server. For details about the behavior during the startup, see 18.4 Using the application failover feature.
When the startup process is completed, the servers go into the states described below for normal operation. To check the server state, run the nnmcluster command without any option (in the interactive mode) or run the nnmcluster -display command, and check the display in the State column.
- States during normal operation
Active server: ACTIVE_NNM_RUNNING
Standby server: STANDBY_READY
When you start NNMi, confirm that the servers go into these states for normal operation. The following table describes the major server states:
State displayed |
Role |
Explanation |
---|---|---|
ACTIVE_NNM_STARTING |
Active |
NNMi is starting. |
ACTIVE_DB_BACKUP |
Active |
NNMi database is being backed up. |
ACTIVE_NNM_RUNNING |
Active |
NNMi is running. |
STANDBY_RECV_DBZIP |
Standby |
Database is being transferred from the NNMi of the active server. |
STANDBY_READY |
Standby |
Server with NNMi is ready as the standby server. |
Before performing an application failover, for example switching between the active and standby servers, make sure that the servers are in the states for normal operation. If you perform a failover before the servers go into these states, the servers cannot be correctly synchronized, and as a result NNMi might fail to start and will go into the ACTIVE_NNM_FAILED state. In this case, stop both servers and restart NNMi. If a database problem prevents NNMi from starting, reset the database, restore the backup data, and then restart NNMi.
(2) Restarting NNMi
You can restart the standby NNMi management server at any time with no special instructions. If you are restarting both the standby and active NNMi management servers, restart the active NNMi management server first.
To restart either the active or the standby NNMi management server:
Run the nnmcluster -disable command on the NNMi management server to disable the application failover feature.
Restart the NNMi management server by executing the following commands:
ovstop ovstart
Run the nnmcluster -enable command on the NNMi management server to enable the application failover feature.
- Important
For important information about NNMi's TrapReceiver process and how it relates to failovers, see 21.16 NNMi NmsTrapReceiver process.
- Application failover control after a communication failure
Once a communication failure between the two cluster nodes has been resolved, the NNMi management server that had been running the longest before the communication failure occurred (that is, the previous active NNMi management server) is designated as the active server.
(3) NNMi failover
When a system configured for application failover fails, a failover automatically occurs according to a scenario described in 18.4.2 Application failover scenarios and NNMi starts on the active server.
To manually initiate a failover:
Run the ovstop -failover command on the active server.
After NNMi stops, the cluster manager (nnmcluster) stops and the standby server becomes the new active server, and then NNMi starts.
When step 1 is performed, the original active server is removed as a cluster configuration member. To bring back this server into the cluster as a standby server, run the ovstart command.
The ovstart command is replaced with the nnmcluster -daemon command before being executed.