Cluster configurations and cluster-wide transactions are currently not supported in the RavenDB persistence.
RavenDB version 4 and higher supports multi-node cluster configurations. The NServiceBus.RavenDB persister will guard against cluster configurations, as they can lead to data loss if not used correctly (with cluster-wide transactions).
Database nodes in a database group, distributed across multiple cluster nodes operate in master mode, which means that any database node in the cluster can accept writes. Due to missing optimistic concurrency control for cluster-wide transactions, data loss may happen on concurrent writes to different database nodes.
If the database is configured as a multi-node cluster, remove all nodes but one.
Removing a node that contains a database will remove it. Make sure the node is empty before deleting it.