Following are the anisble playbook used to deploy a 1 master 2 worker k8s cluster.
Environment
- Ubuntu VM's running on Vmware
- K8s with Docker Runtime
User Creation
- Asks for the User Name which has to be created
- Create's the user
- Adds a dedicated Sudo entry
- Setting up Password less sudo for user
- Copy the local uses ssh key to server for password less auth
- Print the details
- Updates the System
- hosts: all
become: yes
vars_prompt:
- name: "new_user"
prompt: "Account need to be create in remote server."
private: no
tasks:
- name: creating the user {{ new_user }}.
user:
name: "{{ new_user }}"
createhome: yes
shell: /bin/bash
append: yes
state: present
- name: Create a dedicated sudo entry file for the user.
file:
path: "/etc/sudoers.d/{{ new_user }}"
state: touch
mode: '0600'
- name: "Setting up Sudo without Password for user {{ new_user }}."
lineinfile:
dest: "/etc/sudoers.d/{{ new_user }}"
line: '{{ new_user }} ALL=(ALL) NOPASSWD: ALL'
validate: 'visudo -cf %s'
- name: Set authorized key for user copying it from current {{ new_user }} user.
authorized_key:
user: "{{ new_user }}"
state: present
key: "{{ lookup('file', lookup('env','HOME') + '/.ssh/id_rsa.pub') }}"
- name: Print the created user.
shell: id "{{ new_user }}"
register: new_user_created
- debug:
msg: "{{ new_user_created.stdout_lines[0] }}"
- name: "Update cache & Full system update"
apt:
update_cache: true
upgrade: dist
cache_valid_time: 3600
force_apt_get: true
Package Installation in Master and Worker Nodes
- Copy the local host files to all the server for name resolution
- update the hostnames of the machines based on the names in host file
- Temporary Swap off
- Swap off in fstab
- Create a empty file for containerd module.
- Configure module for containerd.
- Create a empty file for kubernetes sysctl params.
- Configure sysctl params for Kubernetes.
- Apply sysctl params without reboot
- Installing Prerequisites for Kubernetes
- Add Docker’s official GPG key
- Add Docker Repository
- Install containerd.
- Configure containerd.
- Configure containerd.
- Creating containerd Config file
- Enable containerd service, and start it.
- Add Google official GPG key
- Add Kubernetes Repository
- Installing Kubernetes Cluster Packages.
- Enable service kubelet, and enable persistently
- Reboot all the Kubernetes nodes.
- hosts: "master, workers"
remote_user: ansible
become: yes
become_method: sudo
become_user: root
gather_facts: yes
connection: ssh
tasks:
- name: Copying the host file
copy:
src: /etc/hosts
dest: /etc/hosts
owner: root
group: root
- name: "Updating hostnames"
hostname:
name: "{{ new_hostname }}"
- name: Make the Swap inactive
command: swapoff -a
- name: Remove Swap entry from /etc/fstab.
lineinfile:
dest: /etc/fstab
regexp: swap
state: absent
- name: Create a empty file for containerd module.
copy:
content: ""
dest: /etc/modules-load.d/containerd.conf
force: no
- name: Configure module for containerd.
blockinfile:
path: /etc/modules-load.d/containerd.conf
block: |
overlay
br_netfilter
- name: Create a empty file for kubernetes sysctl params.
copy:
content: ""
dest: /etc/sysctl.d/99-kubernetes-cri.conf
force: no
- name: Configure sysctl params for Kubernetes.
lineinfile:
path: /etc/sysctl.d/99-kubernetes-cri.conf
line: "{{ item }}"
with_items:
- 'net.bridge.bridge-nf-call-iptables = 1'
- 'net.ipv4.ip_forward = 1'
- 'net.bridge.bridge-nf-call-ip6tables = 1'
- name: Apply sysctl params without reboot.
command: sysctl --system
- name: Installing Prerequisites for Kubernetes
apt:
name:
- apt-transport-https
- ca-certificates
- curl
- gnupg-agent
- vim
- software-properties-common
state: present
- name: Add Docker’s official GPG key
apt_key:
url: https://download.docker.com/linux/ubuntu/gpg
state: present
- name: Add Docker Repository
apt_repository:
repo: deb [arch=amd64] https://download.docker.com/linux/ubuntu xenial stable
state: present
filename: docker
update_cache: yes
- name: Install containerd.
apt:
name:
- containerd.io
state: present
- name: Configure containerd.
file:
path: /etc/containerd
state: directory
- name: Configure containerd.
shell: /usr/bin/containerd config default > /etc/containerd/config.toml
- name: Creating containerd Config file
copy:
dest: "/etc/crictl.yaml"
content: |
runtime-endpoint: unix:///run/containerd/containerd.sock
image-endpoint: unix:///run/containerd/containerd.sock
timeout: 2
debug: false
- name: Enable containerd service, and start it.
systemd:
name: containerd
state: restarted
enabled: yes
daemon-reload: yes
- name: Add Google official GPG key
apt_key:
url: https://packages.cloud.google.com/apt/doc/apt-key.gpg
state: present
- name: Add Kubernetes Repository
apt_repository:
repo: deb http://apt.kubernetes.io/ kubernetes-xenial main
state: present
filename: kubernetes
mode: 0600
- name: Installing Kubernetes Cluster Packages.
apt:
name:
- kubeadm
- kubectl
- kubelet
state: present
- name: Enable service kubelet, and enable persistently
service:
name: kubelet
enabled: yes
- name: Reboot all the kubernetes nodes.
reboot:
msg: "Reboot initiated by Ansible"
connect_timeout: 5
reboot_timeout: 3600
pre_reboot_delay: 0
post_reboot_delay: 30
test_command: whoami
Master Configuration
- Pulls all needed images
- Reset Kubeadm if its already configured
- Initialize K8s cluster
- Create Directory for Kube config file in master
- Create a local kube config file in master
- Copy the kube config file to ansible local server
- Genarates the Kube toke for workers and stores it
- Copy the token to master's tmp directory
- Copy the toke to ansible local tmp direcotry
- Initialize the pod network with fannel
- Copy the output to mater file
- Copy the output to ansible local server
- hosts: master
remote_user: ansible
become: yes
become_method: sudo
become_user: root
gather_facts: yes
connection: ssh
tasks:
- name: Pulling images required for setting up a Kubernetes cluster
shell: kubeadm config images pull
- name: Resetting kubeadm
shell: kubeadm reset -f
register: output
- name: Initializing Kubernetes cluster
shell: kubeadm init --apiserver-advertise-address=$(ip a |grep ens160| grep 'inet ' | awk '{print $2}' | cut -f1 -d'/') --pod-network-cidr 10.244.0.0/16 --v=5
register: myshell_output
- debug: msg="{{ myshell_output.stdout }}"
- name: Create .kube to home directory of master server
file:
path: $HOME/.kube
state: directory
mode: 0755
- name: Copy admin.conf to user's kube config to master server
copy:
src: /etc/kubernetes/admin.conf
dest: $HOME/.kube/config
remote_src: yes
- name: Copy admin.conf to user's kube config to ansible local server
become: yes
become_method: sudo
become_user: root
fetch:
src: /etc/kubernetes/admin.conf
dest: /Users/rahulraj/.kube/config
flat: yes
- name: Get the token for joining the nodes with Kuberentes master.
shell: kubeadm token create --print-join-command
register: kubernetes_join_command
- debug:
msg: "{{ kubernetes_join_command.stdout_lines }}"
- name: Copy K8s Join command to file in master
copy:
content: "{{ kubernetes_join_command.stdout_lines[0] }}"
dest: "/tmp/kubernetes_join_command"
- name: Copy join command from master to local ansible server
fetch:
src: "/tmp/kubernetes_join_command"
dest: "/tmp/kubernetes_join_command"
flat: yes
- name: Install Pod network
shell: kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
register: myshell_output
- name: Copy the output to master file
copy:
content: "{{ myshell_output.stdout }}"
dest: "/tmp/pod_network_setup.txt"
- name: Copy network output from master to local ansible server
fetch:
src: "/tmp/pod_network_setup.txt"
dest: "/tmp/pod_network_setup.txt"
flat: yes
Worker Configuration
- Copy the token from ansible local file to worker nodes
- Reset the kubeadm
- Join the Worker node to Master by running the command
- hosts: workers
remote_user: ansible
become: yes
become_method: sudo
become_user: root
gather_facts: yes
connection: ssh
tasks:
- name: Copy token to worker nodes.
become: yes
become_method: sudo
become_user: root
copy:
src: /tmp/kubernetes_join_command
dest: /tmp/kubernetes_join_command
mode: 0777
- name: Resetting kubeadm
shell: kubeadm reset -f
register: output
- name: Join the Worker nodes with the master.
become: yes
become_method: sudo
become_user: root
command: sh /tmp/kubernetes_join_command
register: joined_or_not
- debug:
msg: "{{ joined_or_not.stdout }}"
K8s should be up with the worker nodes now.
