The information you have supplied is not particularly clear.

When patching clustered SQL instances the process you should follow is:

1. Patch the passive nodes of the cluster

2. Failover the active node of the cluster to a passive node, this will begin the upgrade of the system databases etc.

3. Once what was the former passive node is now active patch what was previously the active node.

4. Once this is complete failback to ensure all nodes are running the correct version of SQL.