Merge pull request #1 from DmitriySafronov/k3s-ha

node

```

If multiple hosts are in the master group, the playbook will automatically setup k3s in HA mode with etcd.

https://rancher.com/docs/k3s/latest/en/installation/ha-embedded/

This requires at least k3s version 1.19.1

If needed, you can also edit `inventory/my-cluster/group_vars/all.yml` to match your environment.

Start provisioning of the cluster using the following command:

k3s_version: v1.22.3+k3s1

ansible_user: debian

systemd_dir: /etc/systemd/system

master_ip: "{{ hostvars[groups['master'][0]]['ansible_host'] | default(groups['master'][0]) }}"

# If you define multiple masters you should be providing a loadbalanced

# apiserver endpoint to all masters here. This default value is only suitable

# for a non-HA setup, if used in a HA setup, it will not protect you if the

# first node fails.

# Also you should define k3s_token so that masters can talk together securely

apiserver_endpoint: "{{ hostvars[groups['master'][0]]['ansible_host'] | default(groups['master'][0]) }}"

# k3s_token: "mysupersecuretoken"

extra_server_args: ""

extra_agent_args: ""

---

k3s_server_location: /var/lib/rancher/k3s

server_init_args: >-

  {% if groups['master'] | length > 1 %}

    {% if ansible_host == hostvars[groups['master'][0]]['ansible_host'] | default(groups['master'][0]) %}

      --cluster-init --tls-san {{ apiserver_endpoint }}

    {% else %}

      --server https://{{ hostvars[groups['master'][0]]['ansible_host'] | default(groups['master'][0]) }}:6443

    {% endif %}

    --token {{ k3s_token }}

  {% endif %}

  {{ extra_server_args | default('') }}

---

- name: Clean previous runs of k3s-init

  systemd:

    name: k3s-init

    state: stopped

  failed_when: false

- name: Clean previous runs of k3s-init

  command: systemctl reset-failed k3s-init

  failed_when: false

  changed_when: false

  args:

    warn: false  # The ansible systemd module does not support reset-failed

- name: Init cluster inside the transient k3s-init service

  command:

    cmd: "systemd-run -p RestartSec=2 \

                      -p Restart=on-failure \

                      --unit=k3s-init \

                      k3s server {{ server_init_args }}"

    creates: "{{ systemd_dir }}/k3s.service"

  args:

    warn: false  # The ansible systemd module does not support transient units

- name: Verification

  block:

    - name: Verify that all nodes actually joined (check k3s-init.service if this fails)

      command:

        cmd: k3s kubectl get nodes -l "node-role.kubernetes.io/master=true" -o=jsonpath="{.items[*].metadata.name}"

      register: nodes

      until: nodes.rc == 0 and (nodes.stdout.split() | length) == (groups['master'] | length)

      retries: 20

      delay: 10

      changed_when: false

  always:

    - name: Kill the temporary service used for initialization

      systemd:

        name: k3s-init

        state: stopped

      failed_when: false

- name: Copy K3s service file

  register: k3s_service

    owner: "{{ ansible_user }}"

    mode: "u=rw,g=,o="

- name: Replace https://localhost:6443 by https://master-ip:6443

- name: Configure kubectl cluster to https://{{ apiserver_endpoint }}:6443

  command: >-

    k3s kubectl config set-cluster default

      --server=https://{{ master_ip }}:6443

      --server=https://{{ apiserver_endpoint }}:6443

      --kubeconfig ~{{ ansible_user }}/.kube/config

  changed_when: true

Type=notify

ExecStartPre=-/sbin/modprobe br_netfilter

ExecStartPre=-/sbin/modprobe overlay

ExecStart=/usr/local/bin/k3s agent --server https://{{ master_ip }}:6443 --token {{ hostvars[groups['master'][0]]['token'] }} {{ extra_agent_args | default("") }}

ExecStart=/usr/local/bin/k3s agent --server https://{{ apiserver_endpoint }}:6443 --token {{ hostvars[groups['master'][0]]['token'] | default(k3s_token) }} {{ extra_agent_args | default("") }}

KillMode=process

Delegate=yes

# Having non-zero Limit*s causes performance problems due to accounting overhead