The software described in this documentation is either no longer supported or is in extended support.
Oracle recommends that you upgrade to a current supported release.

6.1 Swarm Management

A swarm consists of one or more manager nodes and several worker nodes. The manager node is used to dispatch tasks to worker nodes. The manager also performs all of the orchestration and cluster management functions to maintain the state of the swarm. A single manager node is elected as the leader manager node and other manager nodes remain on standby so that they are ready to take on the role of leader at any point in time. Manager nodes are elected to the leader role through node consensus. Although it possible to run an environment with a single manager node, ideally three or more nodes should run as manager nodes.

Worker nodes receive tasks from manager nodes and execute required actions for the swarm, such as starting or stopping a container. By default, manager nodes also behave as worker nodes, but this behavior is configurable. The leader manager node tracks the state of the cluster and in the event that a worker node becomes unavailable, the manager ensures that any containers that were running on the unavailable worker node are started on an alternative worker node.

Tasks that affect the swarm can be executed via any manager node regardless of whether it is the leader. At any point, a manager node can be used to promote a worker node to management status. See Section 6.1.4, “Promoting nodes to manager status” for more information.

In this section, the steps to set up the nodes in a swarm are discussed in some detail. Instructions are also provided for some basic swarm management.