Skip to content

day18.md

Latest commit

 

History

History
47 lines (44 loc) · 4.22 KB

day18.md

File metadata and controls

47 lines (44 loc) · 4.22 KB
Raw
title DevOps: Day 18
tags
DevOps

DevOps: Day 18

  • to set up ansible, it is recommended to run it on a linux machine/instance.
  • next, you will want to run sudo apt update before installing ansible with sudo apt install ansible.
  • set up two different servers, and then configured passwordless connection between them using keypair (private and public) generated using ssh-keygen.
  • after doing this, to access one on another it's just simply ssh <private-ip-of-server>.
  • this passwordless authentication is also one of the requirements of ansible.
  • download and install ansible and setup a passwordless authentication.
  • ansible playbook is the name of ansible files. So simply put, ansible files are called ansible playbooks.
  • ansible commands can be run directly from the terminal too, so not always will one have to write an ansible playbook. these commands that can run directly are called ansible adhoc commands.
  • before running the ansible adhoc commands, make sure you have the inventory file in the same location you're running the ansible adhoc commands.
  • this inventory file is simply just a file that you will put the ip address (private ip address) of all of your host computers that you are working with.
  • in the same location where the inventory file is located, the first command I run is the ansible -i inventory all -m "shell" -a "touch devopsclass"
    • in the above command, -m stands for module, -a stands for argument.
  • it is advisable to use the ansible adhoc commands when you want to run just a single command. While use the ansible playbook if you want to run a series of ansible commands.
  • simply put, ansible adhoc commands for one or two tasks, while ansible playbook for multiple commands.
  • in a situation where one has different servers for different things, let's say webservers and dbservers, then one can create a distiction between these two or more division of servers using the square bracket. 
    [dbservers]
172.31.62.28
[webservers]
172.31.62.111
    so for executing an ansible adhoc command for just the webservers, one just needs to make use of: ansible -i inventory webservers -m "shell" -a "df" as against using all in same command ansible -i inventory all -m "shell" -a "df".
  • Task: we want to install and restart nginx, hence since we are executing multiple tasks, we will need to write an ansible playbook. An ansible playbook is written in yaml/yml file.
    • in this yaml file, the first line should have three dash lines (---) which shows it is a yaml file. next comes the name: what the playbook does, then the hosts: all/grouped server(s), become: the user it is executing as e.g. root etc.
    • next, we start writing tasks. name: Install nginx, shell: apt install nginx or apt: name: nginx, state: present. Or shell: apt install nginx
  • it is always advisable to go with the package manager installation in the ansible playbook whenever it is available.
  • to execute an ansible playbook, we make use of the command: ansible-playbook, while for the ansible adhoc command, we make use of ansible.
  • running ansible-playbook command with the -vvv enables debug mode making it verbose.
  • one can create services using Ansible, but the idle tool for that is the Terraform. Ansible is more idle for configuration. Terraform is an IaC (Infrastructure as Code) tool. Thus, anything that has to do with creating infrastructure, the best tool for the job is Terraform.
  • Ansible roles is efficient way of writing ansible playbooks that will only enable one's efficiency of writing complex playbooks. This solves the issue that involves writing many tasks (e.g. 50 - 60 tasks) in a single ansible playbook.
  • to start using ansible roles:
    • ansible-galaxy role init kubernetes
    • a directory kubernetes will be created with a README.md file and the following sub-directories: templates, files, handlers, defaults, vars, meta, tests and tasks.
    • inside the tasks, we write all the tasks.
    • meta is used to write meta data information and also the licensing information.
    • var folder can be used to set variables.
    • tests folder is used to add tests, unit test
    • handlers is used to take care of exception.
    • files contains like certificate, index.html etc.