3.9 KiB

Raw Blame History

page_title: Docker Swarm page_description: Swarm: a Docker-native clustering system page_keywords: docker, swarm, clustering

Docker Swarm: a Docker-native clustering system

Docker swarm helps you control a cluster of Docker hosts (known as nodes) and expose them as a single "virtual" host.

The Docker swarm manager can be interacted with using the Docker API, which means any tool which can communicate with a Docker Daemon using that API, can control a Docker swarm transparently: dokku, fig, krane, flynn, deis, docker-ui, shipyard, drone.io, Jenkins... and of course the Docker client itself.

Like the other Docker projects, swarm follows the "batteries included but removable" principle. It ships with a simple scheduling backend out of the box, and as initial development settles, an API will develop to enable pluggable backends. The goal is to provide a smooth out-of-box experience for simple use cases, and allow swapping in more powerful backends, like Mesos, for large scale production deployments.

Installation

Note: The only requirement for Swarm nodes is they all run the same release Docker daemon (version 1.4.0 and later), configured to listen to a tcp port that the Swarm manager can access.

Docker swarm is currently only available as a single go binary on Linux. Download it from the latest release page on GitHub.

For example:

	$ wget -O swarm https://github.com/docker/swarm/releases/download/v0.1.0-rc1/swarm-Linux-x86_64
	# OR
	$ curl -SsL https://github.com/docker/swarm/releases/download/v0.1.0-rc1/swarm-Linux-x86_64 > swarm
	$ chmod 755 swarm
	$ sudo cp swarm /usr/local/bin

Nodes setup

Each swarm node will run a swarm node agent which will register the referenced Docker daemon, and will then monitor it, updating the discovery backend to its status.

The following example uses the Docker Hub based token discovery service:

# create a cluster
$ swarm create
6856663cdefdec325839a4b7e1de38e8 # <- this is your unique <cluster_id>

# For each of your nodes, start a swarm agent
#  the Docker daemon <node_ip> doesn't have to be public (eg. 192.168.0.X),
#  as long as the manager and the docker cli can reach it, it is fine.
$ swarm join --addr=<node_ip:2375> --discovery token://<cluster_id>

# start the manager on any machine or your laptop
$ swarm manage -H tcp://<swarm_ip:swarm_port> --discovery token://<cluster_id>

# use the regular docker cli
$ docker -H tcp://<swarm_ip:swarm_port> info
$ docker -H tcp://<swarm_ip:swarm_port> run ...
$ docker -H tcp://<swarm_ip:swarm_port> ps
$ docker -H tcp://<swarm_ip:swarm_port> logs ...
...

# list nodes in your cluster
$ swarm list --discovery token://<cluster_id>
<node_ip:2375>

Note: In order for the Swarm manager to be able to communicate with the node agent on each node, they must listen to a common network interface. This can be achieved by starting with the -H flag (e.g. -H tcp://0.0.0.0:2375).

TLS

Swarm supports TLS authentication between the CLI and Swarm but also between Swarm and the Docker nodes. However, all the Docker daemon certificates and client certificates must be signed using the same CA-certificate.

In order to enable TLS for both client and server, the same command line options as Docker can be specified:

swarm manage --tlsverify --tlscacert=<CACERT> --tlscert=<CERT> --tlskey=<KEY> [...]

Please refer to the Docker documentation for more information on how to set up TLS authentication on Docker and generating the certificates.

Note: Swarm certificates must be generated withextendedKeyUsage = clientAuth,serverAuth.

Discovery services

See the Discovery service document for more information.

Advanced Scheduling

See filters and strategies to learn more about advanced scheduling.

3.9 KiB Raw Blame History