Container Linux: Difference between revisions

Revision as of 07:57, 3 April 2014

Revision as of 08:05, 3 April 2014

Line 108:

}}</ref>

CoreOS also provides '''<tt>fleet</tt>''', a cluster manager daemon that controls CoreOS' separate systemd instances at the cluster level. By using <tt>fleet</tt>, CoreOS creates a distributed [[init]] system that ties together separate systemd instances and a cluster-wide <tt>etcd</tt> deployment. Using <tt>fleet</tt> allows single or multiple [[Software container|containers]] to be deployed cluster-wide, with more advanced options including [[Redundancy (engineering)|redundancy]], [[failover]], locking oftying containers to specificcluster machinesmembers, dependencies between containers, and grouped deployment of containers. Command-line utility called '''<tt>fleetctl</tt>''' is used to configure and monitor this distributed init system by communicating with the <tt>etcd</tt> cluster. When used locally on a cluster member, <tt>fleetctl</tt> communicates with the local <tt>etcd</tt> instance over the [[loopback interface]]; when used from an external host, [[SSH tunnel]]ing is used with authentication provided through [[public SSH key]]s. Internally, <tt>fleet</tt> daemon communicates with systemd over [[D-Bus]].<ref>{{Cite web

| url = https://coreos.com/docs/launching-containers/launching/launching-containers-fleet/

| title = Launching containers with fleet | work = CoreOS documentation