Cluster Properties

The spectrum-container.properties file contains the properties that configure a cluster. This table describes how to set the cluster properties for a cluster where the server and configuration database are installed together on each node, and for nodes in the server cluster in a separated configuration database installation. If you are setting up a configuration database cluster, see Cluster Properties for a Configuration Database Cluster.

Property Description
spectrum.cluster.enabled Set this to true.
spectrum.cluster.name If this is the first node in the cluster, set this to the name you want to give to the cluster.

If this is not the first node in the cluster, set this to the name of the cluster that you want this node to join. If you are unsure of the cluster name, check the spectrum.cluster.name property on another node in the cluster.

spectrum.cluster.address Specify the node's own IP address.
spectrum.cluster.seeds If you are setting up the first node in the cluster, or if this is the first node you are starting after having stopped the cluster, specify only the host name or IP address of this server. This effectively points this node to itself as a seed node, which is necessary in order for it to be able to start as the first node in the cluster. Do not specify 127.0.0.1.
If this is not the first node in the cluster, specify the host name or IP address of each of the other nodes in the cluster, including nodes that you have not yet installed. Separate each seed node with a comma. When starting up, the node will attempt to access each seed node in the order listed. If none of the seed nodes can be reached, the node will not join the cluster.
Tip: We recommend that you specify multiple seed nodes so that if one seed node is down others can be used to allow nodes to join the cluster. We also recommend that you specify the same seed nodes in the same order on all nodes in the cluster.
spectrum.cluster.nodeID Set this to an integer value to use as the ID for this node. The ID must be an integer and must be unique among all nodes in the cluster.
Tip: We recommend that the first node you install have a node ID of 1, the second one 2, and so on.
spectrum.repository.server.cluster.nodeCount This property sets the number of repository nodes in the cluster. This property must be set on nodes that run a Neo4j instance. The way in which you specify this property depends upon your cluster configuration:
  • For cluster configurations with separate servers and repositories: Set this flag in the spectrum-container.properties file on each repository node configuration, and not on the server node configurations.
  • For combined server-repository configurations: Set the property in the spectrum-container.properties file on each node of the cluster.
spectrum.runtime.hostname Un-comment this property and set this property to the load balancer host name.
spectrum.runtime.port Un-comment this property and set this property to the port on which you have configured the load balancer to listen.
spectrum.repository.addresses Set this to the same IP address specified in spectrum.cluster.address. The default value localhost cannot be used in a clustered environment.

For clustered standard installations, in which the repository is embedded on the server machine, define this property with each node's IP address. Separate each IP address with a comma. For example, in a three-node cluster, you would specify: spectrum.repository.addresses=node1-IPaddress,node2-IPaddress,node3-IPaddress

In distributed clustered configurations, with separate clusters of servers and repositories, on different machines, specify the IP addresses of each node in the configuration database cluster. Separate each IP address with a comma. For example, in a three-node database cluster, you would specify: spectrum.repository.addresses=dbNode1-IPaddress,dbNode2-IPaddress,dbNode3-IPaddress