![Variables
When a play is run generally the setup
module is run. We can disable it using
gather_facts: false.
Gather facts gives us access to a lot
of vars on the target machine. Too
many to list here. (ansible all -m setup)
Arguments can be accessed with
bracket or dot notation
Registered vars are where we save the
output from a task
ansible hostname -m setup
- debug: var=ansible_facts
{{ ansible_facts['devices']['xvda']['model'] }}
//can cause conflicts with python
{{ ansible_facts.devices.xvda.model }}
- hosts: web_servers
tasks:
- shell: /usr/bin/foo
register: foo_result
ignore_errors: True
- shell: /usr/bin/bar
when: foo_result.rc == 5](https://image.slidesharecdn.com/ansible1-181125210739/85/Introduction-to-Ansible-Peter-Halligan-15-320.jpg)
Introduction to Ansible - Peter Halligan
- 1. Introduction to Ansible Peter Halligan
- 2. Introduction
- 3. Introduction Ansible is an python based tool for deployment and orchestration of infrastructure It is agentless and communicates with hosts via ssh Push based Easy to read
- 4. overview
- 5. Installation ● Pre-installed on many linux distro ● Use package manager ● Use pip install ● MacOs pip install ○ CC=clang sudo -E pip install pycrypto. ● Not officially supported on windows ○ Cygwin ○ Linux subsytem. We can still manage windows hosts even if Ansible does not support windows as a management machine
- 6. Components - Playbooks - Tasks - Inventory files - Variables - Filters - Roles - Ansible tower - awx Suggested directory structure
- 7. Playbooks, plays & tasks
- 8. Playbook and Plays Playbooks are made up of one or more plays. Indentation is important!!! A play is a set of task which are run on the host pattern outlined by the play. Common keywords used to control the play
- 9. Tasks Tasks are a list of modules to be executed in order. Modules are written to be idempotent. Which means they can be run multiple times and result will be the same each time. Script, command, shell commands may not return the same result each time. Pre-tasks, post-tasks, blocks
- 10. Inventory
- 11. Static Inventory Static defined by user Organised into groups All hosts belong to at least 2 groups Can use patterns to decide what hosts to run plays on, including multiple groups, excluding groups or an intersection of groups We can also store variable data in inventory files
- 12. Dynamic Inventory Python scripts and modules which can generate the inventory file Must enable the plugin Can organise based on tags Difficult set up for gce Plugin not working properly on ansible 2.6 or 2.7
- 13. Variables
- 14. Variables Currently there are over 20 places we can define a variable. Tasks in a role will see their own default vars Tasks defined outside a role will see the last roles role defaults If multiple groups have the same named variable the last to load will be used. Hosts at the same group level load alphabetically ● command line values (eg “-u user”) ● role defaults ● inventory file or script group vars ● inventory group_vars/all ● playbook group_vars/all ● inventory group_vars/* ● playbook group_vars/* ● inventory file or script host vars ● inventory host_vars/* ● playbook host_vars/* ● host facts / cached set_facts ● play vars ● play vars_prompt ● play vars_files ● role vars (defined in role/vars/main.yml) ● block vars (only for tasks in block) ● task vars (only for the task) ● include_vars ● set_facts / registered vars ● role (and include_role) params ● include params ● extra vars (always win precedence) -e “this=that” https://docs.ansible.com/ansible/2.7/user_guide/playbooks_variables.html#variabl e-precedence-where-should-i-put-a-variable
- 15. Variables When a play is run generally the setup module is run. We can disable it using gather_facts: false. Gather facts gives us access to a lot of vars on the target machine. Too many to list here. (ansible all -m setup) Arguments can be accessed with bracket or dot notation Registered vars are where we save the output from a task ansible hostname -m setup - debug: var=ansible_facts {{ ansible_facts['devices']['xvda']['model'] }} //can cause conflicts with python {{ ansible_facts.devices.xvda.model }} - hosts: web_servers tasks: - shell: /usr/bin/foo register: foo_result ignore_errors: True - shell: /usr/bin/bar when: foo_result.rc == 5
- 16. Templates Templates use the jinja2 templating engine The modules takes a source template which contains a placeholder for a var: {{ my_var }} replaces it on the controller machine with the var and then copies it to the destination host
- 17. Roles
- 18. Roles Roles are a way to organise and reuse and share ansible content; tasks, file vars, templates etc Similar to the default structure ansible -galaxy contains many roles Import (static) v include (dynamic) Previously had to use roles, pre-task, post-tasks more difficult to control the flow of tasks
- 19. Rolling updates
- 20. Rolling updates Rolling updates are simple Add the serial keyword to the play Add max_fail_percentage (optional) Run your playbook Example can pass a list of numbers or percentage to a serial and it will run first on 1 host, then on 5, then on 20% of remaining machines until completion Setting max_fail_percentage: would act on all those groups too
- 22. Resources https://docs.ansible.com/ https://www.ansiblefordevops.com/ https://www.safaribooksonline.com/videos/hands-on-infrastructure-automation/9781788991599 Configuration Management 101: Writing Ansible Playbooks https://www.safaribooksonline.com/library/view/learn-ansible/9781788998758/