Procedure
Step 1 
Perform an upgrade to the new Cisco Unified Communications Manager version on your discontinued first node (publisher) server. Refer to the preceding sections in this chapter that describe the kind of upgrade you want to do. Observe the warning on the console that tells you that the only thing you can do with the new Cisco Unified Communications Manager version is to perform a DRS backup ("This hardware has limited functionality. Backup and Restore is the only supported functionality.").
Step 2 Perform an upgrade to the new Cisco Unified Communications Manager version on your subsequent node (subscriber) servers. Refer to the preceding sections in this chapter that describe the kind of upgrade you want to do.
Step 3 Verify database synchronization between all nodes. You can use the CLI commands utils dbreplication runtime state and utils dbreplication status. For more information, refer to the Command Line Interface Reference Guide for Cisco Unifed Communications Solutions.
Step 4 Using the new Cisco Unified Communications Manager version on your discontinued first node server, perform a DRS backup. The DRS backups are encrypted using the cluster security password provided at install time. You must remember this security password as the "old" password, because you may be prompted to enter this "old" password at the time of restore. Refer to the Disaster Recovery System Administration Guide.
Step 5 Disconnect your discontinued server from the network.
Step 6 Install the new Cisco Unified Communications Manager version on your new supported first node server. You must obtain and install a new license on this server. Refer to the guide Installing Cisco Unified Communications Manager. You will be prompted to enter a "new" security password, a password that is different from the "old" password you noted in Step 4. The guide Installing Cisco Unified Communications Manager describes the requirements of a "new" security password that Cisco Unified Communications Manager will accept. You must remember this "new" security password.
Step 7 Using the new Cisco Unified Communications Manager version on your new supported first node server, perform the Disaster Recovery System Administration Guide procedure "Restoring the First Node only (Rebuilding the Publisher Alone)". First, select only select the first node for restore. You can only select the subsequent nodes for restore after the completion of first node restore. Use the discontinued server's backup file that you created in Step 4. You will be prompted for the "old" security password that you noted in Step 4. For further details, refer to the Disaster Recovery System Administration Guide.
Step 8 On your new supported first node server, reactivate all services that used to be active on your discontinued first node server before the bridge upgrade. Refer to the Administration Guide for Cisco Unity Connection Serviceability.
Step 9 Verify database synchronization between all nodes. You can use the CLI commands utils dbreplication runtime state and utils dbreplication status. For more information, refer to the Command Line Interface Reference Guide for Cisco Unifed Communications Solutions.
So, you can expect an outage until you perform the bridge upgrade on the whole cluster. Backup and Restore is the only supported functionality of the bridge upgrade, so no you can not leave the old hardware in place. You should migrate the cluster to the UCS asap and at one time to have a shorter down time.
Regards,
Anthony
