About Oracle discovery
The NetBackup Discovery Service (nbdisco) discovers Oracle database instances throughout the NetBackup environment. The discovery service reports to the primary server when it finds instances and databases to help you build an Oracle Intelligent Policy. The service polls the clients upon NetBackup installation and periodically after installation (every 4 hours). Instance management collects the discovered instances in an instance repository. The user can access this repository on the NetBackup web UI or by using the nboraadm command.
The NetBackup Discovery Service searches for instances and databases in different areas where Oracle is installed. The following areas are where the Discovery Service searches:
Non-RAC Single instances are discovered by searching the
oratabfile on UNIX and from the registry on Windows.NetBackup looks for the Oracle health check files that are found in the Oracle home. These are not cleaned up when a database is deleted. You may need to delete them manually otherwise NetBackup can continue to find the databases that are deleted.
Oracle RAC databases are discovered when NetBackup queries the Oracle Cluster Ready Services (CRS) using the Oracle Clusterware high availability API.
Oracle RAC in the web UI does not support upgrades from legacy script-based policies. Also, there is no web UI support for the configurations that are created using Appendix A or Appendix B in the NetBackup for Oracle Administrator's Guide.
To allow the NetBackup web UI to discover a RAC instance or cluster:
Remove the Oracle RAC from any configuration that is setup using Appendix A or Appendix B in the NetBackup for Oracle Administrator's Guide.
Remove any Oracle RAC from existing OIP policies in the current NetBackup Administrator's Console.
Note:
When an Oracle RAC database is discovered, that database does not have a Database ID. A Database ID is required to manually add additional RAC instances to the database. You must register the RAC database and provide a Database ID before adding additional instances.
By default, this service is enabled to report instances. However, you can use the REPORT_CLIENT_DISCOVERIES client configuration entry to shut down or restart the service on a particular client. By default, REPORT_CLIENT_DISCOVERIES is not present in the Windows registry or the UNIX bp.conf file.
To change the default setting, use bpsetconfig to add or change the entry:
In the Windows registry.
In the
/usr/openv/netbackup/bp.conffile on UNIX.
Use the following format: REPORT_CLIENT_DISCOVERIES = TRUE | FALSE
Set REPORT_CLIENT_DISCOVERIES to FALSE to shut down the discovery service. The service shuts down within 10 minutes and remains down on the client. To turn on the discovery service on that client, set REPORT_CLIENT_DISCOVERIES to TRUE or remove the entire entry. Then run bp.start_all on the client to restart the service.
To set this value on a client remotely, run the following command from the primary server:
echo REPORT_CLIENT_DISCOVERIES=FALSE | bpsetconfig -h clientname