Skip to main content
Version: Beta 🚧

Deploy with Docker Swarm

Software​

Install Docker on all servers/VMs: https://docs.docker.com/engine/install.

Create a Swarm​

  1. On one of the servers, initialize the swarm:

    $ docker swarm init

    > Swarm initialized: current node (bvz81updecsj6wjz393c09vti) is now a manager.
    >
    > To add a worker to this swarm, run the following command:
    >
    >     docker swarm join \
    >     --token SWMTKN-1-3pu6hszjas19xyp7ghgosyx9k8atbfcr8p2is99znpy26u2lkl-1awxwuwd3z9j1z3puu7rcgdbx \
    >     172.17.0.2:2377
    >
    > To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.

  2. On the other servers, join the swarm:

    $ docker swarm join \
        --token SWMTKN-1-3pu6hszjas19xyp7ghgosyx9k8atbfcr8p2is99znpy26u2lkl-1awxwuwd3z9j1z3puu7rcgdbx \
        172.17.0.2:2377

  3. Read the administration guide for docker swarm carefully and make the appropriate adjustments for your environment and verify that all nodes are in the swarm:

    $ docker node ls
    ID                                HOSTNAME         STATUS    AVAILABILITY   MANAGER STATUS   ENGINE VERSION
    ko66mqj76xo9ftunxq78luc8p         vm01             Ready     Active         Reachable        23.0.1
    ogziph0qxfeivly5fnekepwx0         vm02             Ready     Active                          23.0.1
    8utl07361ocn5xvzqh27z0c8s *       vm03             Ready     Active         Reachable        23.0.1
    1prl8b1m7xw2ph5b8dnh98glk         vm04             Ready     Active                          23.0.1
    ...

Tag Nodes​

We use labels to tag our nodes and to deploy services to appropriate nodes. You will need the following labels:

  • worker: Deploys Workers and VoD Services (Required, unless you use Runners)
  • runner: Deploys Runners (Required, unless you use Workers)
  • edge: Deploys Edge Servers (Required)
  • voiceservice: Lecture transcription (Optional)

This command adds the label worker to the node vm02, instructing our deployment to deploy Workers on this node:

docker node update --label-add worker=true vm02
This is an example configuration you should aim for:
docker node ls -q | xargs docker node inspect -f '{{ .ID }} [{{ .Description.Hostname }}]: {{ .Spec.Labels }}'

kwgmm6sxb9nqwojoclxuy4mpt [vmgpu01]: map[voiceservice:true] # optional, this is a server with a GPU for transcription
ctfdd9mtkse2yxid8zku2wx1f [vm01]: map[worker:true] # the number of Workers depends on the number of concurrent streams you want to process. 1 worker can process around 5 stream in our environment.
u391iukj6nljosaaygcfkzy2s [vm02]: map[worker:true]
wkxct5tvzclvc4uqm8w573dlf [vm03]: map[worker:true]
72weo6nozra1cdgjs5wghe7gh [vm04]: map[worker:true]
f7ik66qq6tzhsbwphfpdp2vm1 [vm07]: map[worker:true]
i4l8ouumms96qu96evkb6srol [vm08]: map[worker:true]
vq5cw2bgwncenr5cp89xzsi32 [vm09]: map[worker:true]
q4as4i27z2hnwypgzj8ql2dz1 [vm10]: map[worker:true]
lfged5ra1a7z9wlstxa2bml5c [vm11]: map[worker:true]
3wu812ybzynnunrpoqdsay0bf [vm12]: map[worker:true]
itdbo77gempnl251lakioe5y1 [vm13]: map[worker:true]
zcplsihexr88plf0t8q25tdn7 [vm14]: map[worker:true]
fbi92hp7s0u3c2x13tgrb6fd6 [vm15]: map[worker:true]
o6k2egpupik3qjgq2w0azv70o [vm16]: map[worker:true]
th77fn3s91s06sy4ciprita3s [vm17]: map[edge:true] # the number of Edge nodes depends on the number of concurrent viewers you want to support.
wpue8f384h7z71mngov5j72c1 [vm18]: map[edge:true]
urac70xjf1kx5op39kyulykad [vm19]: map[edge:true]
5bqr01nyefxqmkd3luzhh3sne [vm20]: map[edge:true]

Deploy GoCast​

  1. Download the docker-compose.yml file.

  2. Edit the docker-compose.yml file and replace the placeholders with your actual values.

    tip

    To find all placeholders for fields, search for lines containing TODO, while placeholders for paths are marked as /path/to/xyz.

  3. Deploy the stack to the swarm:

    docker stack deploy -c docker-compose.yml gocast

  4. Verify that the services are running:

    docker service ls

    You should see a list of services with their replicas. If the number of replicas for a service is less than the desired number, it means that the service is still starting up or has encountered an error.

  5. To view the logs of a service:

    docker service logs SERVICE_NAME