Become Single Cluster operation

This operation assumes that this instance (either primary or mirror) is the only instance that is to be left functional in the environment.

Before you begin

Important: If IBM Storage Scale pattern type V1.2.16.0 or later is installed on your system, the Become Single Cluster operation is not available for use on the Manage > Instance Console > Operations > GPFS-Manager.GPFS_Manager > Failover active-active page of the Cloud Pak System Software console.

If you use IBM Storage Scale V5.x, complete these steps on the node on which this operation must be done.

Shut down GPFS daemons on all gpfs nodes by running this command.
```
su - gpfsprod -c 'sudo /usr/lpp/mmfs/bin/mmshutdown -a’
```

Disable CCR by running this command.

su - gpfsprod -c 'sudo /usr/lpp/mmfs/bin/mmchcluster --ccr-disable'

Start the Cluster by running this command.

su - gpfsprod -c 'sudo /usr/lpp/mmfs/bin/mmstartup -a'

About this task

As part of this process all nodes in the cluster are shut down and clients (local and remote) will need to remount the file systems on the cluster.

Procedure

Expand the Failover Active-Active section of the GPFS_Manager operations.
Expand the Become Single Cluster node.
Click Submit.
Click Confirm when asked to validate your request.

Results

This operation assumes that the tiebreaker and the opposite primary or mirror instance are temporarily down. The operation assigns the IBM Storage Scale primary and secondary servers to the remaining instance in the environment. It will also move the cluster manager node and file systems manager to the remaining instance. It excludes the non-running nodes from the environment and the disks associated with those nodes. As part of this process all nodes in the cluster are shut down and clients might need to remount the file systems on the cluster.

When the instance that is down can be attached back to the cluster, run the Become Replicated Cluster operation on that instance.