Restore a Two Node Cluster

If one node in a two-node OCSDM cluster becomes non-operational, the non-operational node tries to recover and rejoin the cluster. However, if this node is non-operational for more than 24 hours, use this task to restore the two-node cluster.

Note:

With the introduction of Oracle Communications Session Delivery Manager, Release 8.0, you must select one server to start in the cluster only. Once this server is started and operational, you can start the other server(s) in the cluster. When connectivity between nodes is re-established automatically or manually, the Berkley XML database decides which node becomes the master.

Login to the operational node as the nncentral user.
Change to the bin directory on the operational node. For example:
```
cd /home/nncentral/AcmePacket/NNC<version>/bin/
```
Enter the backupdbhot.sh script to do a hot backup of the application and reporting databases on the operational node.
```
./backupdbhot.sh
```
On the shut down node, login as the nncentral user, navigate to the bin directory, and enter the reinitialize.sh script.
```
./reinitialize.sh
```
Go back to the operational node, and restore the application database and the reporting databases on your OCSDM two-node cluster, enter the restoredb.sh script followed by the directory in which the backup application database is stored. If applicable, you can also specify the Oracle reporting database password and OCSDMDW reporting database password. For example:
```
./restoredb.sh -f/home/nncentral/AcmePacket/NNC<version>/bin/NNC<version>_Backup_2017_04_19.tar.gz -z my_report_db_password -y my_ocsdmdw_db_password
```
On the shut down node, enter the startnnc.sh script. For example:
```
./startnnc.sh
```