Differences

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

--- lab:stack_of_nucs:ansible_playbook_-_install_kubernetes [2023/05/09 20:32] – [Kubernetes Dashboard] user
+++ lab:stack_of_nucs:ansible_playbook_-_install_kubernetes [2024/05/06 02:11] (current) – removed user
@@ Line 1: / Line 1: @@
-====== Ansible Playbook - Install Kubernetes ======
-In our previous step we [[Ansible Playbook - FAH Removal|cleaned up the FAH installation]] on our [[start|Stack of NUCs]].
-Now we are going to install Kubernetes with:
-  * one NUC master node this we can call [[NUC 3]]
-  * remaining NUCs "workers"
-Purpose:
-  * Demonstrate a running a complex workload of web applications on Kubernetes
-References
-  * [[https://www.goodreads.com/book/show/26759355-kubernetes|Kubernetes: Up & Running]] by O'Reilly
-  * [[https://github.com/brettplarson/nuctestlab]]
-  * [[https://www.linuxtechi.com/install-kubernetes-on-ubuntu-22-04/]]
-  * [[https://github.com/torgeirl/kubernetes-playbooks]]
-====== Step 1 - Set Up for Kubernetes ======
-  - From [[NUC 1]], log in to the Ansible control node, [[NUC 2]].
-  - Create the k8s folder on the Ansible control node, [[NUC 2]]
-    * ''mkdir /home/ansible/my-project/k8s''
-  - Create inventory file
-    * Choose one IP address to be the Kubernetes master, add it to ''[master]''
-    * The rest of the IP address should be added to ''[workers]''
-    * <file yaml inventory>
-[master]
-[workers]
-[all:vars]
-ansible_python_interpreter=/usr/bin/python3
-ansible_user='ansible'
-ansible_become=true
-ansible_become_method=sudo
-</file>
-  - Set up ansible.cfg file to tell Ansible to use the inventory file
-    * <file yaml ansible.cfg>
-[defaults]
-inventory         = inventory
-</file>
-  - Set up the host files for name resolution
-    * Playbook will update the '/etc/hosts' files on all the nodes to allow them to resolve each other by name, but without DNS
-    * Also adds this information to the Ansible control node (see 'localhost' in the script)
-    * <file yaml updatehostsfile.yml>
----
-- name: Update etc/hosts file
-  hosts: all, localhost
-  gather_facts: true
-  tasks:
-    - name: Populate all /etc/hosts files
-      tags: etchostsupdate
-      become: true
-      become_user: root
-      lineinfile:
-        path: "/etc/hosts"
-        regexp: '.*{{ item }}$'
-        line: "{{ hostvars[item]['ansible_default_ipv4'].address }}\t{{ hostvars[item]['ansible_hostname'] }}\t{{ hostvars[item]['ansible_hostname'] }}"
-        state: present
-      with_items: '{{ groups.all }}'
-</file>
-    * ''ansible-playbook updatehostsfile.yml''
-====== Step 2 - Set Up Kubernetes Using Ansible======
-Check [[https://github.com/torgeirl/kubernetes-playbooks]] for updates to the playbooks below
-  - Install some prerequisites on ALL the Kubernetes nodes
-    * <file yaml kube-dependencies.yml>
----
-- hosts: all
-  become: true
-  tasks:
-    - fail:
-        msg: "OS should be Ubuntu 22.04, not {{ ansible_distribution }} {{ ansible_distribution_version }}"
-      when: ansible_distribution != 'Ubuntu' or ansible_distribution_version != '22.04'
-    - name: Update APT packages
-      apt:
-        update_cache: true
-    - name: Reboot and wait for reboot to complete
-      reboot:
-    - name: Disable SWAP (Kubeadm requirement)
-      shell: |
-        swapoff -a
-    - name: Disable SWAP in fstab (Kubeadm requirement)
-      replace:
-        path: /etc/fstab
-        regexp: '^([^#].*?\sswap\s+sw\s+.*)$'
-        replace: '# \1'
-    - name: Create an empty file for the Containerd module
-      copy:
-        content: ""
-        dest: /etc/modules-load.d/containerd.conf
-        force: false
-    - name: Configure modules for Containerd
-      blockinfile:
-        path: /etc/modules-load.d/containerd.conf
-        block: |
-             overlay
-             br_netfilter
-    - name: Create an empty file for Kubernetes sysctl params
-      copy:
-        content: ""
-        dest: /etc/sysctl.d/99-kubernetes-cri.conf
-        force: false
-    - 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: Install APT Transport HTTPS
-      apt:
-        name: apt-transport-https
-        state: present
-    - name: Add Docker apt-key
-      apt_key:
-        url: https://download.docker.com/linux/ubuntu/gpg
-        state: present
-    - name: Add Docker's APT repository
-      apt_repository:
-        repo: "deb [arch=amd64] https://download.docker.com/linux/ubuntu {{ ansible_distribution_release }} stable"
-        filename: "docker-{{ ansible_distribution_release }}"
-    - name: Add Kubernetes apt-key
-      apt_key:
-        url: https://packages.cloud.google.com/apt/doc/apt-key.gpg
-        state: present
-    - name: Add Kubernetes' APT repository
-      apt_repository:
-        repo: deb https://apt.kubernetes.io/ kubernetes-xenial main
-        state: present
-        filename: 'kubernetes'
-    - name: Install Containerd
-      apt:
-        name: containerd.io
-        state: present
-    - name: Create Containerd directory
-      file:
-        path: /etc/containerd
-        state: directory
-    - name: Add Containerd configuration
-      shell: /usr/bin/containerd config default > /etc/containerd/config.toml
-    - name: Configuring the systemd cgroup driver for Containerd
-      lineinfile:
-        path: /etc/containerd/config.toml
-        regexp: '            SystemdCgroup = false'
-        line: '            SystemdCgroup = true'
-    - name: Enable the Containerd service and start it
-      systemd:
-        name: containerd
-        state: restarted
-        enabled: true
-        daemon-reload: true
-    - name: Install Kubelet
-      apt:
-        name: kubelet=1.26.*
-        state: present
-        update_cache: true
-    - name: Install Kubeadm
-      apt:
-        name: kubeadm=1.26.*
-        state: present
-    - name: Enable the Kubelet service, and enable it persistently
-      service:
-        name: kubelet
-        enabled: true
-    - name: Load br_netfilter kernel module
-      modprobe:
-        name: br_netfilter
-        state: present
-    - name: Set bridge-nf-call-iptables
-      sysctl:
-        name: net.bridge.bridge-nf-call-iptables
-        value: 1
-    - name: Set ip_forward
-      sysctl:
-        name: net.ipv4.ip_forward
-        value: 1
-    - name: Check Kubelet args in Kubelet config
-      shell: grep "^Environment=\"KUBELET_EXTRA_ARGS=" /etc/systemd/system/kubelet.service.d/10-kubeadm.conf || true
-      register: check_args
-    - name: Add runtime args in Kubelet config
-      lineinfile:
-        dest: "/etc/systemd/system/kubelet.service.d/10-kubeadm.conf"
-        line: "Environment=\"KUBELET_EXTRA_ARGS= --runtime-cgroups=/system.slice/containerd.service --container-runtime-endpoint=unix:///run/containerd/containerd.sock\""
-        insertafter: '\[Service\]'
-      when: check_args.stdout == ""
-    - name: Reboot and wait for reboot to complete
-      reboot:
-- hosts: master
-  become: true
-  tasks:
-    - name: Install Kubectl
-      apt:
-        name: kubectl=1.26.*
-        state: present
-        force: true  # allow downgrades
-</file>
-    * ''ansible-playbook kube-dependencies.yml''
-  - Configure kubernetes cluster on master node
-    * <file yaml master.yml>
----
-- hosts: master
-  become: true
-  tasks:
-    - name: Create an empty file for Kubeadm configuring
-      copy:
-        content: ""
-        dest: /etc/kubernetes/kubeadm-config.yaml
-        force: false
-    - name: Configuring the container runtime including its cgroup driver
-      blockinfile:
-        path: /etc/kubernetes/kubeadm-config.yaml
-        block: |
-             kind: ClusterConfiguration
-             apiVersion: kubeadm.k8s.io/v1beta3
-             networking:
-               podSubnet: "10.244.0.0/16"
-             ---
-             kind: KubeletConfiguration
-             apiVersion: kubelet.config.k8s.io/v1beta1
-             runtimeRequestTimeout: "15m"
-             cgroupDriver: "systemd"
-             systemReserved:
-               cpu: 100m
-               memory: 350M
-             kubeReserved:
-               cpu: 100m
-               memory: 50M
-             enforceNodeAllocatable:
-             - pods
-    - name: Initialize the cluster (this could take some time)
-      shell: kubeadm init --config /etc/kubernetes/kubeadm-config.yaml >> cluster_initialized.log
-      args:
-        chdir: /home/ansible
-        creates: cluster_initialized.log
-    - name: Create .kube directory
-      become: true
-      become_user: ansible
-      file:
-        path: $HOME/.kube
-        state: directory
-        mode: 0755
-    - name: Copy admin.conf to user's kube config
-      copy:
-        src: /etc/kubernetes/admin.conf
-        dest: /home/ansible/.kube/config
-        remote_src: true
-        owner: ansible
-    - name: Install Pod network
-      become: true
-      become_user: ansible
-      shell: kubectl apply -f https://raw.githubusercontent.com/flannel-io/flannel/master/Documentation/kube-flannel.yml >> pod_network_setup.log
-      args:
-        chdir: $HOME
-        creates: pod_network_setup.log
-</file>
-    * ''ansible-playbook master.yml''
-  - SSH to the master and verify the master nodes gets status ''Ready''
-    * ''ssh MASTER_IP kubectl get nodes''
-  - Initialize workers
-    * Modify the file to replace MASTERIP with the IP address of your master node in two (2) places
-    * <file yaml workers.yml>
----
-- hosts: master
-  become: true
-  # gather_facts: false
-  tasks:
-    - name: Get join command
-      shell: kubeadm token create --print-join-command
-      register: join_command_raw
-    - name: Set join command
-      set_fact:
-        join_command: "{{ join_command_raw.stdout_lines[0] }}"
-- hosts: workers
-  become: true
-  tasks:
-    - name: TCP port 6443 on master is reachable from worker
-      wait_for: "host={{ 'MASTERIP' }} port=6443 timeout=1"
-    - name: Join cluster
-      shell: "{{ hostvars['MASTERIP'].join_command }} >> node_joined.log"
-      args:
-        chdir: /home/ansible
-        creates: node_joined.log
-</file>
-    * ''ansible-playbook workers.yml''
-  - SSH to the master and verify all the nodes return status ''Ready''
-    * ''ssh MASTER_IP kubectl get nodes''
-  - Install kubectl on [[NUC 2]] for automation with Kubernetes
-    * create file ''/home/ansible/my-project/k8s/kubectlcontrolnode.yml''
-    * <file yaml kubectlcontrolnode.yml>
----
-- hosts: localhost
-  become: true
-  gather_facts: false
-  tasks:
-    - name: Update APT packages
-      apt:
-        update_cache: true
-    - name: Add Kubernetes apt-key
-      apt_key:
-        url: https://packages.cloud.google.com/apt/doc/apt-key.gpg
-        state: present
-    - name: Add Kubernetes' APT repository
-      apt_repository:
-        repo: deb https://apt.kubernetes.io/ kubernetes-xenial main
-        state: present
-        filename: 'kubernetes'
-    - name: Install Kubectl
-      apt:
-        name: kubectl=1.26.*
-        state: present
-        force: true  # allow downgrades
-</file>
-    * ''ansible-playbook kubectlcontrolnode.yml''
-    * Running ''kubctl version'' will fail at this point because you don't have credentials
-    * Copy credentials
-      * ''scp -r IPMASTER:/home/ansible/.kube ~/''
-    * Confirm it's working by running ''kubectl version''
-  - Install kubectl on [[NUC 1]] for remote testing
-    * <code>
-sudo apt update
-sudo apt install -y ca-certificates curl
-sudo curl -fsSLo /etc/apt/keyrings/kubernetes-archive-keyring.gpg https://packages.cloud.google.com/apt/doc/apt-key.gpg
-echo "deb [signed-by=/etc/apt/keyrings/kubernetes-archive-keyring.gpg] https://apt.kubernetes.io/ kubernetes-xenial main" | sudo tee /etc/apt/sources.list.d/kubernetes.list
-sudo apt update
-sudo apt install -y kubectl
-</code>
-    * Running ''kubctl version'' will fail at this point because you don't have credentials
-    * Copy credentials
-      * ''scp -r tux@IPMASTER:/home/ansible/.kube ~/''
-    * Confirm it's working by running
-      * ''kubectl version''
-      * ''kubectl get nodes''
-====== Next Step ======
-You have successful installed Kubernetes on your [[start|Stack of NUCs]]! Next we will [[demonstrate_app_on_k8s|demonstrate running a web app on Kubernetes]].
-====== Optional ======
-Do some tests with Kubernetes.
-===== Manually Create Pod =====
-  - <code bash>kubectl run speedtester --image=docker.io/doritoes/speedtester:latest</code>
-  - ''kubectl get pods''
-    * ''kubectl get pods -w'' will "watch" the process of the container being created and changing status to Running; press control-c to exit
-  - ''kubectl describe pod speedtester''
-  - ''kubectl delete pods/speedtester''
-    * this demonstrates another way to specify the pod, same as ''kubectl delete pod speedtester''
-===== Create Pod using Pod Manifest =====
-  - Create yaml file
-    * <file yaml speedtest-pod.yml>
-apiVersion: v1
-kind: Pod
-metadata:
-  name: speedtester
-spec:
-  containers:
-    - image: docker.io/doritoes/speedtester:latest
-      name: speedtester
-      ports:
-        - containerPort: 8080
-          name: http
-          protocol: TCP
-</file>
-  - Build the pod
-    * ''kubectl apply -f speedtest-pod.yml''
-  - Examine the pod
-    * ''kubectl get pods''
-    * ''kubectl describe pods speedtester''
-  - Test connecting to the speedtest application on the pod
-    * from [[NUC 1]]
-      * ''kubectl port-forward speedtester 8080:80''
-      * browse to [[http://localhost:8080]]
-      * NOTE in testing, the port forwarding tended to break when it the upload test started ("error creating error stream for port 8080 -> 80: Timeout occurred")
-        * the worker node NUC doesn't seem to be loaded at all
-          * ''kubectl exec --stdin --tty speedtester -- /bin/bash''
-          * ''htop''
-        * one workaround is to keep re-launching the port forward command, but that doesn't solve the upload issue
-        * later on, we will demonstrate it is stable when using a better forwarding mechanism
-        * ''while true; do kubectl port-forward speedtester 8080:80; done''
-  - Manage the pod
-    * <code bash>
-kubectl logs speedtester
-kubectl exec speedtester -- date
-kubectl exec speedtester -- uname -a
-</code>
-  - Delete the pod
-    * <code bash>
-kubectl delete -f speedtest-pod.yaml
-kubectl get pods
-</code>
-===== k9s =====
-You can experiment running k9s on your Kubernetes master node
-  * [[https://github.com/derailed/k9s/releases]]
-  * <code bash>
-mkdir k9s
-cd k9s
-wget https://github.com/derailed/k9s/releases/download/v0.27.3/k9s_Linux_amd64.tar.gz
-tar xzvf k9s_Linux_amd64.tar.gz
-./k9s
-</code>
-===== Kubernetes Dashboard =====
-References:
-  * [[https://kubernetes.io/docs/tasks/access-application-cluster/web-ui-dashboard/]]
-  - Step 1 - Install
-    * Install the dashboard from [[NUC 1]] or [[NUC 2]]
-      * ''kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.7.0/aio/deploy/recommended.yaml''
-  - Step 2 - Create a service account and bind cluster-admin role to it
-    * <code bash>
-kubectl create serviceaccount dashboard -n kubernetes-dashboard''
-kubectl create clusterrolebinding dashboard-admin -n kubernetes-dashboard  --clusterrole=cluster-admin  --serviceaccount=kubernetes-dashboard:dashboard
-</code>
-  - Step 3 - Get a Bearer Token
-    * ''kubectl -n kubernetes-dashboard create token dashboard''
-  - Step 4 - Access
-    * From [[NUC 1]] run
-      * ''kubectl proxy''
-      * UI can __only be accessed from the machine where the command is executed__
-      * See ''kubectl proxy --help'' for more options.
-    * From a web browser on [[NUC 1]] open:
-      * [[http://localhost:8001/api/v1/namespaces/kubernetes-dashboard/services/https:kubernetes-dashboard:/proxy/]]
-    * Authenticate with "Token" and paste in the token from Step 3 above
-  - Step - 5 Deploy containerized applications
-    * [[https://kubernetes.io/docs/tasks/access-application-cluster/web-ui-dashboard/]]