There is little point running clusters with even numbers of nodes. To see why this is imagine you have one cluster of 3 nodes, and a second cluter of 4 nodes. As usual a majority of nodes are required for each cluster to reach consensus on any change.
A majority is defined as `N/2 + 1` where `N` is the number of nodes in the cluster. For a 3-node a majority is 2; for a 4-node cluster a majority is 3. Therefore a 3-node cluster can tolerate the failure of a single node, and 4-node cluster can also only tolerate a failure of a single node.
So a 4-node cluster is no more fault-tolerant than a 3-node cluster, so running a 4-node cluster provides no advantage over a 3-node cluster. Only a 5-node cluster can tolerate the failure of 2 nodes. An analogous argument applies to 5-node vs. 6-node clusters, and so on.
With this command a single node is started, listening for API requests on port 4001 and listening on port 4002 for intra-cluster communication and cluster-join requests from other nodes. This node stores its state at `~/node`.
Once executed you now have a cluster of two nodes. Of course, for fault-tolerance you need a 3-node cluster, so launch a third node like so on _host3_:
It is the nature of clustered systems, nodes can fail at anytime. Depending on the size of your cluster, it will tolerate various amounts of failure. With a 3-node cluster, it can tolerate the failure of a single node, including the leader.
If an rqlite process crashes, it is safe to simply to restart it. The node will pick up any changes that happened on the cluster while it was down.
You can grow a cluster simply by joining a new node to an existing cluster, via requests to the leader, at anytime. The new node will pick up changes that have occurred on the cluster since the cluster first started.
If a node fails completely and is not coming back -- a complete loss -- then it must be first removed from the cluster. To remove a node from a cluster, execute the following command:
assuming `localhost` is the address of the cluster leader. Removing a node does not change the number of nodes required to reach quorum, so you must add a new node to cluster. To do so, simply follow the instructions for _Growing a cluster_.