Docker 1.12 Swarm

https://t.co/RqH1meXdO3

TL;DR. Docker 1.12: Now with Built-in Orchestration. You orchestrate the Docker hosts as a cluster (or swarm), not the containers that run within the Docker host (that’s the task of resource scheduler like Kubernetes)

https://www.youtube.com/watch?v=F7hoq0KwHD4

More at https://www.youtube.com/watch?v=FgXJKw37po8 and https://blog.docker.com/2016/06/dockercon-part-1-top-10-video/

Tested this Swarm cluster using two Turnkey Linux VMs.

Note: More at https://docs.docker.com/engine/swarm/swarm-tutorial

First node (manager)

• docker swarm init
• docker node ls

ID NAME MEMBERSHIP STATUS AVAILABILITY MANAGER STATUS

14kz2qyngq62o754wb5ufa1ul * core Accepted Ready Active Leader
6pkybnsbepnhrz4deoc1aunv9 core Accepted Down Active

• docker node inspect self
• docker swarm update

ID NAME MEMBERSHIP STATUS AVAILABILITY MANAGER STATUS

14kz2qyngq62o754wb5ufa1ul * core Accepted Ready Active Leader
6pkybnsbepnhrz4deoc1aunv9 core Accepted Down Active

• docker node rm 14kz

Error response from daemon: rpc error: code = 9 desc = node

14kz2qyngq62o754wb5ufa1ul is a cluster manager and is part of the quorum. It must be demoted to worker before removal

• docker node rm 6pky

Error response from daemon: rpc error: code = 9 desc = node

6pkybnsbepnhrz4deoc1aunv9 is not down and can’t be removed

• docker node ls

ID NAME MEMBERSHIP STATUS AVAILABILITY MANAGER STATUS

14kz2qyngq62o754wb5ufa1ul * core Accepted Ready Active Leader
6pkybnsbepnhrz4deoc1aunv9 core Accepted Down Active
774skgsihm3w1bkd8dttejud6 core Accepted Ready Active

• docker service create

Example:

docker service create — name vote -p 8080:80 instavote/vote

Output:

7k792z653eycsp7x75hubw5t7

Only the Swarm manager can create a Docker service

• docker service tasks <ID>

Example:

docker service tasks 7k79

• docker service scale vote=6
• docker service update - -image instavote/vote:movies vote

Second node (worker)

• docker swarm join IPaddress:port

where IPaddress of manager, default port is 2377

Example:

docker swarm join 192.168.0.133:2377

• docker node inspect self

Error response from daemon: This node is not participating as a Swarm manager

• docker swarm leave
• docker swarm join 192.168.0.133:2377

Error response from daemon: This node is processing an existing join request that has not succeeded yet. Use “docker swarm leave” to cancel the current request.

• docker swarm update

If a node is already part of a Swarm cluster or if there is a pending request to join a cluster, use docker swarm update to make a node ready.

• docker node ls

Error response from daemon: This node is not participating as a Swarm manager

• docker swarm join 192.168.0.133:2377

This node joined a Swarm as a worker.

Networking

• Docker uses libnetwork, Kubernetes does not

Docker Container Network Model

• by default, libnetwork makes use of Linux-Bridge and IPtables.
• libnetwork provides built-in IP address management, native multi-host networking, and native service discovery
• libnetwork defines the Container Network Model itself
• as of the Docker 1.12 release, Docker now integrates native load balancing, also pulled in from the Linux kernel, by way of the IPVS (IP virtual server)
• Docker 1.12 now also benefits from a routing mesh powered by Gossip protocol.
• macvlan capability enables Docker engine to more easily join an existing physical or virtual network.

Do you want to be a developer, and sustain a fun and meaningful IT Marketplace? Just click HERE.