18.7.4 Modifying the NNMi settings
- Organization of this subsection
(1) Modifying the settings
This subsection explains how to modify the NNMi settings in the application failover configuration.
(a) NNMi settings files
When modifying the NNMi settings files, make the same modification on both the active and standby servers so that the files are identical in content.
- Important
If a setting modification will result in a restart of NNMi, stop both servers before modifying the setting.
To modify settings without stopping NNMi operation, follow the procedure described below. In each step, run the nnmcluster -display command to verify that the processing has been completed before proceeding to the next step.
Run the ovstop -failover command on server A.
Server A stops and server B becomes active.
Modify the settings on server A.
Run the ovstart command on server A to start it as the standby server.
Run the ovstop -failover command on server B.
Server B stops and server A becomes active.
Modify the settings on server B.
Run the ovstart command on server B to start it as the standby server.
(b) NNMi database
Because the active and standby servers automatically maintain synchronization with each other on the NNMi database, no operation by the system administrator is required.
For details about this behavior, see 18.4 Using the application failover feature.
(2) Database reset
To perform the database reset explained in 4.8 Resetting the NNMi configuration and database, you must cancel the application failover configuration temporarily by doing the following:
(Optional) If you wish to save the current NNMi settings, do the following on the active server:
Use the nnmconfigexport.ovpl command to output the NNMi settings to an XML file.
Use the nnmtrimincidents.ovpl command to archive the NNMi incidents.
Run the nnmcluster -halt command on the active server.
NNMi stops on both the active and standby servers.
To verify that the NNMi service has stopped, run the nnmcluster command with no options specified (interactive mode) or the nnmcluster -display command.
Reset the NNMi database on the active server.
a. Temporarily cancel the application failover configuration settings.
Modify the following file to comment out the cluster name com.hp.ov.nms.cluster.name (by adding the characters #! at the beginning of the line):
- Windows:
%NnmDataDir%shared\nnm\conf\props\nms-cluster.properties
- Linux:
$NnmDataDir/shared/nnm/conf/props/nms-cluster.properties
b. (Optional) Before the database data is deleted, use the following command to back up the existing database as needed:
nnmbackup.ovpl -type offline -target backup-directory
c. Delete the NNMi database and then re-create it:
nnmresetembdb.ovpl -nostart
d. Run the ovstop command on the active NNMi management server. The service that was started in step c stops.
e. Reconfigure application failover.
Modify the following file to uncomment the cluster name com.hp.ov.nms.cluster.name (by deleting the #! characters added in step a):
- Windows:
%NnmDataDir%shared\nnm\conf\props\nms-cluster.properties
- Linux:
$NnmDataDir/shared/nnm/conf/props/nms-cluster.properties
Delete or move the following directories on both the active and standby servers:
Windows:
%NnmDataDir%shared\nnm\databases\Postgres_standby
%NnmDataDir%shared\nnm\databases\Postgres.OLD
Linux:
$NnmDataDir/shared/nnm/databases/Postgres_standby
$NnmDataDir/shared/nnm/databases/Postgres.OLD
Run the ovstart command on the active NNMi management server.
To verify that this step is complete, run the nnmcluster -display command and look for an ACTIVE_NNM_RUNNING message.
Run the ovstart command on the standby NNMi management server.
To verify that this step is complete, run the nnmcluster -display command and look for a STANDBY_READY message.
This leaves the NNMi database in the default settings only.
Set up NNMi on the active server. To import the NNMi settings saved in step 1, use the nnmconfigimport.ovpl command.