Differences

This shows you the differences between two versions of the page.

--- vagrant_kubernetes_install [2020/03/09 19:03] – created andonovj
+++ vagrant_kubernetes_install [2020/03/09 19:42] (current) – andonovj
@@ Line 1: / Line 1: @@
-TODO
+=====Overview=====
+Kubernetes is the next Evolution of Docker swarm, so in order to configure it, we have to firstly configure the Docker (just without the swarm this time)
+We will configure 3 servers again:
+  * One master
+  * Two Workers
+Since I hate seeing only Debian based, I will do it in both: Ubuntu and Centos. Firstly we will be the Ubuntu.
+So let's get going:
+=====Provision the VMs=====
+You can return use the first section of the Vagrant advanced configurations or you can continue here, where we will do pretty much the same. I like wasting space :D
+So let's get going with Vagrant again :)
+<Code:none|Kubernetes Vagrantfile>
+IMAGE_NAME = "bento/ubuntu-16.04"
+N = 2
+Vagrant.configure("2") do |config|
+    config.ssh.insert_key = false
+    config.vm.provider "virtualbox" do |v|
+        v.memory = 1024
+        v.cpus = 2
+    end
+    config.vm.define "k8s-master" do |master|
+        master.vm.box = IMAGE_NAME
+        master.vm.network "private_network", ip: "192.168.50.10"
+        master.vm.hostname = "k8s-master"
+        master.vm.provision "ansible" do |ansible|
+            ansible.playbook = "kubernetes-setup/master-playbook.yml"
+            ansible.extra_vars = {
+                node_ip: "192.168.50.10",
+            }
+        end
+    end
+    (1..N).each do |i|
+        config.vm.define "node-#{i}" do |node|
+            node.vm.box = IMAGE_NAME
+            node.vm.network "private_network", ip: "192.168.50.#{i + 10}"
+            node.vm.hostname = "node-#{i}"
+            node.vm.provision "ansible" do |ansible|
+                ansible.playbook = "kubernetes-setup/node-playbook.yml"
+                ansible.extra_vars = {
+                    node_ip: "192.168.50.#{i + 10}",
+                }
+            end
+        end
+    end
+</Code>
+As you can see, we are refering to two files here:
+  * master-playbook.yml - Ansible Playbook for the Master node
+  * node-playbook.yml - Ansible Playbook for the other Nodes.
+Please create the following files as follows:
+====Configure the Master Playbook====
+Let's start building the master playbook:
+<Code:none|Master Playbook File>
+---
+- hosts: all
+  become: true
+  tasks:
+  - name: Install packages that allow apt to be used over HTTPS
+    apt:
+      name: "{{ packages }}"
+      state: present
+      update_cache: yes
+    vars:
+      packages:
+      - apt-transport-https
+      - ca-certificates
+      - curl
+      - gnupg-agent
+      - software-properties-common
+  - name: Add an apt signing key for Docker
+    apt_key:
+      url: https://download.docker.com/linux/ubuntu/gpg
+      state: present
+  - name: Add apt repository for stable version
+    apt_repository:
+      repo: deb [arch=amd64] https://download.docker.com/linux/ubuntu xenial stable
+      state: present
+  - name: Install docker and its dependecies
+    apt:
+      name: "{{ packages }}"
+      state: present
+      update_cache: yes
+    vars:
+      packages:
+      - docker-ce
+      - docker-ce-cli
+      - containerd.io
+    notify:
+      - docker status
+  - name: Add vagrant user to docker group
+    user:
+      name: vagrant
+      group: docker
+    - name: Remove swapfile from /etc/fstab
+    mount:
+      name: "{{ item }}"
+      fstype: swap
+      state: absent
+    with_items:
+      - swap
+      - none
+</Code>
+===Remove the swap===
+Please bare in mind that the kubelet won't install if the swap is enabled, so we have to add the following:
+<Code:none|Remove the swap>
+  - name: Disable swap
+    command: swapoff -a
+    when: ansible_swaptotal_mb > 0
+</Code>
+===Install Kubelet, kubeadm, kubectl===
+After that, we can add a task to install the: kubelet, kubeadm and kubectl using the below code.
+<Code:none|Add Kubelet, kubeadm and kubectl>
+  - name: Add an apt signing key for Kubernetes
+    apt_key:
+      url: https://packages.cloud.google.com/apt/doc/apt-key.gpg
+      state: present
+  - name: Adding apt repository for Kubernetes
+    apt_repository:
+      repo: deb https://apt.kubernetes.io/ kubernetes-xenial main
+      state: present
+      filename: kubernetes.list
+  - name: Install Kubernetes binaries
+    apt:
+      name: "{{ packages }}"
+      state: present
+      update_cache: yes
+    vars:
+      packages:
+        - kubelet
+        - kubeadm
+        - kubectl
+  - name: Configure node ip
+    lineinfile:
+      path: /etc/default/kubelet
+      line: KUBELET_EXTRA_ARGS=--node-ip={{ node_ip }}
+  - name: Restart kubelet
+    service:
+      name: kubelet
+      daemon_reload: yes
+      state: restarted
+</Code>
+===Initialize the Kubernetes===
+Finally, we can add the Tasks for the Initialization as follows:
+<Code:none|Initialize Kubernetes>
+  - name: Initialize the Kubernetes cluster using kubeadm
+    command: kubeadm init --apiserver-advertise-address="192.168.50.10" --apiserver-cert-extra-sans="192.168.50.10"  --node-name k8s-master --pod-network-cidr=192.168.0.0/16
+</Code>
+===Configure Vagrant User for the Cluster===
+Since we are using vagrant, we can setup vagrant to access the Kubernetes Cluster using the following Task:
+<Code:none|Configure Vagrant user>
+  - name: Setup kubeconfig for vagrant user
+    command: "{{ item }}"
+    with_items:
+     - mkdir -p /home/vagrant/.kube
+     - cp -i /etc/kubernetes/admin.conf /home/vagrant/.kube/config
+     - chown vagrant:vagrant /home/vagrant/.kube/config
+</Code>
+===Configure the Network provider and policy engine===
+<Code:none | Configure network provider and policy Engine>
+  - name: Setup kubeconfig for vagrant user
+    command: "{{ item }}"
+    with_items:
+     - mkdir -p /home/vagrant/.kube
+     - cp -i /etc/kubernetes/admin.conf /home/vagrant/.kube/config
+     - chown vagrant:vagrant /home/vagrant/.kube/config
+</Code>
+====Configure the Node Playbook====
+We will setup a join file which will be used in the playbook for the other nodes:
+<Code:none|Generate Join Command>
+  - name: Generate join command
+    command: kubeadm token create --print-join-command
+    register: join_command
+  - name: Copy join command to local file
+    local_action: copy content="{{ join_command.stdout_lines[0] }}" dest="./join-command"
+</Code>
+The generated, from Kubernetes, join command will be saved in file called: "join-command" which will be executed by the playbook:
+===Configure handlers===
+We have to also setup the handlers for checking the Docker daemon:
+<Code:none|Configure handlers>
+  handlers:
+    - name: docker status
+      service: name=docker state=started
+</Code>
+Finally we can configure the node-playbook.yml
+<Code:none|Node-playbook.yml config>
+  - name: Copy the join command to server location
+    copy: src=join-command dest=/tmp/join-command.sh mode=0777
+  - name: Join the node to cluster
+    command: sh /tmp/join-command.sh
+</Code>