This lab will help us understand how Ansible works with managed hosts and deploys software onto them.

Create an instance using GitHub Codespaces. Choose repository denoslab/ensf400-lab5-ansible. Install Ansible.

$ pipx install ansible-core
$ ansible-galaxy collection install ansible.posix
$ ansible-galaxy collection install community.general

Before creating managed hosts, we need to prepare an SSH key pair for authentication purposes by Ansible.

$ mkdir secrets
$ ssh-keygen -t rsa -N "" -C "[email protected]" -f secrets/id_rsa
$ chmod 400 secrets/id_rsa
$ cp secrets/id_rsa.pub secrets/id_rsa_container.pub

After this step, you will have an SSH key pair with one additional file (public key) for mounting purposes.

• id_rsa: This is the private key Ansible Controller will hold as the credential to access the managed hosts.
• id_rsa.pub: This is the public key that will be distributed to the managed hosts to authorize the Ansible Controller to login.
• id_rsa_container.pub: This is the public key copy to be mounted to the containers.

This step will create three slightly modified Alpine Linux containers with SSH access. Note that after these containers start, there is no additional applications installed. For the Web application to be deployed, we will use Ansible to set it up.

Now, we would like to create three hosts as containers to be managed by Ansible. Under the root directory of the repo, create \texttt{docker-compose.yml} with the following content.

version: "3.7"

services:
  managedhost-app-1:
    build: 
      context: managed-host/alpine/app1
    restart: unless-stopped
    network_mode: "bridge"
    environment:
      - SSH_ENABLE_ROOT=true
  managedhost-app-2:
    build: 
      context: managed-host/alpine/app2
    restart: unless-stopped
    network_mode: "bridge"
    environment:
      - SSH_ENABLE_ROOT=true
  managedhost-app-3:
    build: 
      context: managed-host/alpine/app3
    restart: unless-stopped
    network_mode: "bridge"
    environment:
      - SSH_ENABLE_ROOT=true

Edit docker-compose.yml to complete the following tasks to allow these three instances to be managed by Ansible:

1. Mount the public key from the host secrets/id_rsa_container.pub to the container at the path /root/.ssh/authorized_keys. This will authorize the Ansible controller to use its private SSH key to access the managed instances.
2. For managedhost-app-1, map Container port 3000 to Host port 3000 (Web). Map Container port 2223 to Host port 2223 (SSH).
3. For managedhost-app-2, map Container port 3000 to Host port 3001 (Web). Map Container port 2224 to Host port 2224 (SSH).
4. For managedhost-app-3, map Container port 3000 to Host port 3002 (Web). Map Container port 2225 to Host port 2225 (SSH).

Under the root directory of the repo, run the following command to start the managed host containers.

$ docker compose up -d

Use docker compose ps to check if the containers are up and running as expected.

$ docker compose ps
NAME                                      IMAGE                                   COMMAND                  SERVICE             CREATED             STATUS              PORTS
ensf400-ansible-lab-managedhost-app-1-1   ensf400-ansible-lab-managedhost-app-1   "/entry.sh /usr/sbin…"   managedhost-app-1   7 seconds ago       Up 6 seconds        0.0.0.0:2223->2223/tcp, 0.0.0.0:3000->3000/tcp
ensf400-ansible-lab-managedhost-app-2-1   ensf400-ansible-lab-managedhost-app-2   "/entry.sh /usr/sbin…"   managedhost-app-2   7 seconds ago       Up 6 seconds        0.0.0.0:2224->2224/tcp, 0.0.0.0:3001->3000/tcp
ensf400-ansible-lab-managedhost-app-3-1   ensf400-ansible-lab-managedhost-app-3   "/entry.sh /usr/sbin…"   managedhost-app-3   7 seconds ago       Up 6 seconds        0.0.0.0:2225->2225/tcp, 0.0.0.0:3002->3000/tcp

Under the root directory of the repo, create an inventory file called hosts.yml using the following template. Note that since we are using a bridged network, the IP address of all managed hosts should be 0.0.0.0. The following properties shall be set for each managed host:

• There are 3 hosts to manage: managedhost-app-1, managedhost-app-2, and managedhost-app-3.
• ansible_host is 0.0.0.0
• ansible_connection is using ssh
• ansible_port should be consistent with the configuration in docker-compose.yml.
• ansible_user is root.
• The host group app_group shall include all 3 managed hosts.

ungrouped:
  hosts:
    # TODO
app_group:
  hosts:
    # TODO

Under the root directory of the repo, create a file named ansible.cfg using the following template. This file will configure Ansible. Complete the following tasks:

• Set the inventory property to the inventory file you created earlier.
• Set the private_key_file property to the private RSA key file created.

[defaults]

inventory = TODO
private_key_file = TODO
host_key_checking = False
action_warnings = False

Use the following command to check if all hosts are added to the inventory. Note that we will include localhost as we take it as the Ansible Controller.

$ export ANSIBLE_CONFIG=$(pwd)/ansible.cfg
$ ansible all:localhost --list-hosts
  hosts (4):
    managedhost-app-1
    managedhost-app-2
    managedhost-app-3
    localhost

After verifying that all hosts have been added, ping the hosts to see if they can be successfully connected.

$ ansible all:localhost -m ping
localhost | SUCCESS => {
    "changed": false,
    "ping": "pong"
}
managedhost-app-3 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/bin/python"
    },
    "changed": false,
    "ping": "pong"
}
managedhost-app-1 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/bin/python"
    },
    "changed": false,
    "ping": "pong"
}
managedhost-app-2 | SUCCESS => {
    "ansible_facts": {
        "discovered_interpreter_python": "/usr/bin/python"
    },
    "changed": false,
    "ping": "pong"
}

The playbook sample is named \texttt{hello.yml}. It consists of 2 plays:

• First play: target all managed hosts to create a simple file.
• Second play: targets the alpine node to run NodeJS app (running inside the container on port 3000, running externally on port 3000-3002 - check docker-compose).

# run the playbook sample:
$ ansible-playbook hello.yml

After that, check if the playbook runs successfully. Use port forwarding to expose Ports 3000, 3001, and 3002 of the GitHub Codespaces instance. GitHub Codespaces will automatically create the port forwarding. In the PORTS tab, use the "Forwarded Address" URLs to access Ports 3000, 3001, and 3002 to verify the services.

The TA will check the completion of the following tasks:

• Ansible connectivity to all hosts by ping.
• Successful deployment of the NodeJS app on all three managed hosts by accessing the web pages.

$ docker compose down