Advanced Assignment

This assignment takes just about all the knowledge you have gained so far and combines it into a single assignment.

Assignment 0

We need to create some infrastructure first. In the explanation below we’ll use Docker and docker-compose, but you can use any infrastructure you like. Eventually you’ll need 4 hosts, 2 webservers and 2 databaseservers:

(The hostnames can vary a bit, the hostnames you will be using should end up in the Ansible inventory.)

Hostname	Group
node-1	webservers
node-2	webservers
node-3	databaseservers
node-4	databaseservers

Feel free to use the infrastructure directory as a starting point.

All the assignments and examples will be based on this Docker setup.

Assignment 1

Let’s first setup Ansible without too much Ansible code, just the inventory and it’s variable.

1. Create a directory for this assignement, for example advanced_assignment.
2. Create an ansible.cfg file in this directory.
3. Configure ansible.cfg to use the inventory directory as inventory.
4. Create an inventory directory.
5. Create an inventory file hosts in the inventory directory.
6. Create a group webservers in the inventory file.
7. Add 2 hosts to the webservers group: node-1 and node-2, both using ansible_connection=docker.
8. Create a group databaseservers in the inventory file.
9. Add 2 hosts to the databaseservers group: node-3 and node-4, both using ansible_connection=local.
10. Create a group production in the inventory file, add node-1 and node-3 to it.
11. Create a group development in the inventory file, add node-2 and node-4 to it.
12. Create a directory for the group_vars.
13. Create a file group_vars/all/ntp.yml in the group_vars directory.
14. Set the ntp_server variable to 0.pool.ntp.org for all hosts.

Verification of assignment 1

Once the above steps are completed, run these commands to verify if all is working as expected:

ansible -m ping all

The value all used in the above command can be changed for:

• webservers
• databaseservers
• production
• development

Group	Expected output
all	4 nodes: node-1, node-2, node-3 and node-4.
webservers	2 nodes: node-1 and node-2.
databaseservers	2 nodes: node-3 and node-4.
production	2 nodes: node-1 and node-3.
development	2 nodes: node-2 and node-4.

Running this command:

ansible -m ansible.builtin.debug -a "msg=" all

Should return 0.pool.ntp.org for all nodes.

Assignment 2

You should have assignment 1 ready before starting this assignment.

1. Create a playbook playbook.yml in the advanced_assignment directory.
2. Install these tools on all nodes: sudo, tcpdump, vim, wget.
3. Install Apache HTTPD (the package differs for each distribution) on all nodes in the webservers group.
4. Start and enable Apache HTTPD on all nodes in the webservers group.
5. Install MariaDB (the package differs for each distribution) on all nodes in the databaseservers group.
6. Start and enable MariaDB on all nodes in the databaseservers group.

To prevent a high load, set the serial parameter to 50% in the playbook.

Verification of assignment 2

Once the above steps are completed, run these commands to verify if all is working as expected:

ansible-playbook playbook.yml

You should see:

• All nodes being managed.
• The installation of the packages being done.
• The starting of the services being done.

As a bonus; try to keep your playbooks or roles idempotent.

Assignment 3

We’re going to create a play for the NTP setup.

Note: The name ntp is being used, but the package name, configuration file and service name change over different distributions.

Distribution	Package name	Configuration file	Service name
Ubuntu	ntp	/etc/ntp.conf	ntp
CentOS	chrony	/etc/chrony.conf	chronyd

1. Add a task in the playbook.yml for all nodes to install the package for NTP.
2. Add a task to configure the NTP server for all nodes in the webservers group to 0.ch.pool.ntp.org. Make sure to restart NTP on changes made to the configuration file.
3. Start the NTP service.

Some hints on the configuration file: You can login to the nodes and cat a default NTP configuration, and place a modified version in the templates directory of your role.

Verification of assignment 3

Run this Ansible command and check that nothing has changed.

Note, please remember that different names of the package, configuration file and service.

ansible -m lineinfile -a "path=/etc/chrony.conf line='pool 0.pool.ntp.org iburst'" all --check

Assignment 4

Modify the playbook to:

1. Place an index.html on the webservers.
2. Make sure the content of index.html read something like this: “Hello from MY_HOST_NAME.” (Where MY_HOST_NAME is the variable hostname of the node.)

Verification of assignment 4

Run this command: curl http://localhost:80 and make sure “Hello from node-1.” is returned. Run this command: curl http://localhost:81 and make sure “Hello from node-2.” is returned.

Assignment 5

We’re going to play with error handling.

Append to the playbook, a task that:

• Displays a message, something like “Hello.”
• Randomly fails. You’ll need to use the random filter.
• In case the job (randomly) failed, display a message, something like “Oh no, it failed.”.

Verification of assignment 5

When running the playbook a few times, you should see the play summary, where rescued is 1:

node-1       : ok=1    changed=0    unreachable=0    failed=0    skipped=0    rescued=1    ignored=0   
node-2       : ok=1    changed=0    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
node-3       : ok=1    changed=0    unreachable=0    failed=0    skipped=0    rescued=0    ignored=0   
node-4       : ok=1    changed=0    unreachable=0    failed=0    skipped=0    rescued=1    ignored=0