To recover a cluster that has Persistence enabled after a cluster-wide restart, you can force a cluster to start without some members when one or more of them fail.
Before you Begin
You should trigger a partial-start only if your cluster has Persistence enabled and it is configured to use one of the PARTIAL policies. See Choosing a Cluster Recovery Policy.
Using Management Center
See the Management Center docs.
Using the REST API
See the REST API docs.
Using Cluster Management Scripts
See the Cluster management scripts.