Deploy Docker container using Ansible
Deploy Docker container using Ansible
This article will showcase the steps how to deploy docker container using Ansible on a remote Linux machine. This will enable us to automate the process of Docker container deployment, when required, on a multiple remote Linux hosts, with a preset of environment variables and configs for the needed Docker container.
In order to successfully deploy Docker container using Ansible on a remote Linux machine, there are following prerequisites that are required:
- Docker installed and running on the remote host/target – also can be installed with Ansible or run the official Docker install script.
- Here you can check out how to install Docker with Ansible.
- Also some methods and examples we covered with Docker and Ansible: Deploy Nextcloud on Docker with Ansible , Deploy WordPress on Docker with Ansible
- SSH access enabled on the remote hosts with the login parameters preset in the Ansible hosts file
- Ansible installed on client machine(your machine)
- Installed Python docker module for Ansible on target machine
- Installed Python on the target machines and on your local machine
Install Python Docker module for Ansible
Most Linux distros have Python3 preinstalled but for others the Python Docker module that Ansible uses may be missing. You’ll know that it’s if you get an error mentioning that the module is missing or not found. Picture example of the error bellow:
The mentioned module is actually the Docker SDK that Python uses to work with Docker. The easiest way to install the Python Docker module is with the “pip” tool. If the “pip” tool is missing, you can easily install and then with it to install the python docker module:
Debian/Ubuntu sudo apt install python3-pip Fedora sudo dnf install python3-pip CentOS/RedHat sudo yum python3-pip
After you installed the pip, then run the command to install docker module:
pip3 install docker
If by any chance, you encounter an error in Ansible that the it cannot find the Python module, add a python interpreter variable in your hosts file. In most cases it’s located either in “/usr/bin/python3” or “/usr/lib/python3”.
Error looks something like this:
The interpreter variable in the hosts file looks something like this:
Write down the hosts in the hosts file with the login parameters
First step – Adding the necessary parameters in hosts file so that the Ansible can reach, login and interact with our machine:
sudo nano /etc/ansible/hosts
In the hosts file, add the parameters to look something like this:
After the necessary parameters for our remote host are added, save the file and exit.
Ansible playbook for Docker container deployment
Next step – Create the playbook which we’ll run to deploy the Docker container. In the following example, we’ll deploy Nginx web server container, so the playbook will look something like this:
--- - hosts: docker tasks: - name: deploy nginx docker container docker_container: image: nginx:stable name: nginx state: started auto_remove: true ports: - "8080:80"
Playbook file breakdown:
- “hosts: docker” – host target group, in this case the playbook will only target hosts that belong in the group docker
- “name: deploy nginx docker container” – Specified a task that needs to be run at the target group and gave it name
- “docker_container:” – Calling the docker command that manages the docker containers and images
- “image: nginx:stable” – Specified what docker image to pull from the Docker hub
- “name: nginx” – Environment variable to specify the docker container name
- “state: started” – Environment variable to start the docker instantly after the deployment
- “auto_remove: true” – Environment variable that specified the container will be removed automatically when the container stops
- “ports: – “8080:80” – Environment variable to expose container ports
Run the Ansible docker container playbook:
ansible-playbook deploy-docker-container -l docker
Essentially, format and syntax writing for deploying Docker containers with Ansible is same as you write Docker compose file to deploy containers(since both use .yaml format).
Another example we’ll show how to deploy a Docker container with data persistence. For the example, we used Portainer:
--- - hosts: docker tasks: - name: Deploy Portainer docker_container: name: portainer image: portainer/portainer-ce ports: - "9000:9000" - "8000:8000" volumes: - /var/run/docker.sock:/var/run/docker.sock - portainer_data:/data restart_policy: always
We have performed the following steps in this post – run the procedure how to deploy Docker container using Ansible on a remote Linux machine which consists of the following steps:
- Create the hosts file with the remote Linux machines
- Wrote the Ansible Docker container playbook and specified which container to run
- Ran the playbook with the successful result of a deployed Docker container
Thank you for your time…
Great write up, I found an issue with the interpreter and the line you pointed out in the hosts file seemed to resolve the issue.
I also found a problem with the pip install docker and pip install docker-py, the user cannot have both, only one. But we are doing pretty well now. I wanted to say thank you for all of your help.
Thank you for your feedback. I’m glad the article was of use to you.