fix master taint implementation - linting problems (#95)

* add virtual-ip to certificate SAN entries

Adds the kube-vip IP as a Subject Alternative Name in the TLS cert. It is needed otherwise you cannot access the cluster.

* fixes bug with master taints (#1)

- improves taint logic

* fixes typo

* fixes formatting

* fixes undefined group['node'] if missing from hosts.ini (#2)

* fixes undefined group['node'] if missing from hosts.ini

- improves application of master taint by centralizing code

* improves molecule testing, fixes linting

* hacking at linter problems, small tweaks

- increases the metallb timeout error due to intermittent testing errors in GitHub actions

* improves context by renaming taint variable

- makes variable boolean

* fix bug

* removes linting hacks

Co-authored-by: Ioannis Angelakopoulos <ioangel@gmail.com>

inventory/sample/group_vars/all.yml

@@ -22,16 +22,19 @@ k3s_token: "some-SUPER-DEDEUPER-secret-password"
 # it for each of your hosts, though.
 k3s_node_ip: '{{ ansible_facts[flannel_iface]["ipv4"]["address"] }}'
 
-k3s_single_node: "{{ 'true' if groups['k3s_cluster'] | length == 1 else 'false' }}"
+# Disable the taint manually by setting: k3s_master_taint = false
+k3s_master_taint: "{{ true if groups['node'] | default([]) | length >= 1 else false }}"
 
 # these arguments are recommended for servers as well as agents:
 extra_args: >-
   --flannel-iface={{ flannel_iface }}
   --node-ip={{ k3s_node_ip }}
 
-# change these to your liking, the only required one is --disable servicelb
+# change these to your liking, the only required are: --disable servicelb, --tls-san {{ apiserver_endpoint }}
 extra_server_args: >-
   {{ extra_args }}
+  {{ '--node-taint node-role.kubernetes.io/master=true:NoSchedule' if k3s_master_taint else '' }}
+  --tls-san {{ apiserver_endpoint }}
   --disable servicelb
   --disable traefik
 extra_agent_args: >-

molecule/ipv6/overrides.yml

@@ -36,6 +36,8 @@
         #    the default has IPv4 ranges only.
         extra_server_args: >-
           {{ extra_args }}
+          --tls-san {{ apiserver_endpoint }}
+          {{ '--node-taint node-role.kubernetes.io/master=true:NoSchedule' if k3s_master_taint else '' }}
           --disable servicelb
           --disable traefik
           --disable-network-policy

roles/k3s/master/tasks/main.yml

@@ -64,8 +64,7 @@
     cmd: "systemd-run -p RestartSec=2 \
                       -p Restart=on-failure \
                       --unit=k3s-init \
-                      k3s server {{ server_init_args }} \
-                      {{ '--node-taint CriticalAddonsOnly=true:NoExecute' if k3s_single_node|bool == false else ''}}"
+                      k3s server {{ server_init_args }}"
     creates: "{{ systemd_dir }}/k3s.service"
   args:
     warn: false  # The ansible systemd module does not support transient units

roles/k3s/master/templates/k3s.service.j2

@@ -7,7 +7,7 @@ After=network-online.target
 Type=notify
 ExecStartPre=-/sbin/modprobe br_netfilter
 ExecStartPre=-/sbin/modprobe overlay
-ExecStart=/usr/local/bin/k3s server {{ extra_server_args | default("") }} {{ '--node-taint CriticalAddonsOnly=true:NoExecute' if k3s_single_node|bool == false else ''}}
+ExecStart=/usr/local/bin/k3s server {{ extra_server_args | default("") }}
 KillMode=process
 Delegate=yes
 # Having non-zero Limit*s causes performance problems due to accounting overhead

roles/k3s/master/templates/metallb.crds.j2

@@ -1648,8 +1648,6 @@ spec:
       - effect: NoSchedule
         key: node-role.kubernetes.io/control-plane
         operator: Exists
-      - key: CriticalAddonsOnly
-        operator: Exists
 ---
 apiVersion: admissionregistration.k8s.io/v1
 kind: ValidatingWebhookConfiguration

roles/k3s/master/templates/vip.yaml.j2

@@ -69,8 +69,6 @@ spec:
         operator: Exists
       - effect: NoExecute
         operator: Exists
-      - key: CriticalAddonsOnly
-        operator: Exists
   updateStrategy: {}
 status:
   currentNumberScheduled: 0

roles/k3s/post/tasks/main.yml

@@ -28,9 +28,9 @@
   command: >-
     k3s kubectl wait {{ item.resource }}
     --namespace='metallb-system'
-    {% if item.name | default(False) -%} {{ item.name }} {%- endif %}
-    {% if item.selector | default(False) -%} --selector='{{ item.selector }}' {%- endif %}
-    {% if item.condition | default(False) -%} {{ item.condition }} {%- endif %}
+    {% if item.name | default(False) -%}{{ item.name }}{%- endif %}
+    {% if item.selector | default(False) -%}--selector='{{ item.selector }}'{%- endif %}
+    {% if item.condition | default(False) -%}{{ item.condition }}{%- endif %}
     --timeout='{{ metal_lb_available_timeout }}'
   changed_when: false
   run_once: true