Examples and Automations¶

automation:
  - alias: "Stop multiple deployments at night"
    trigger:
      platform: time
      at: "22:00:00"
    action:
      - service: kubernetes.stop_workload
        data:
          workload_names:
            - switch.development_api
            - switch.staging_api
            - switch.monitoring
          namespace: "production"

automation:
  - alias: "Start multiple deployments in the morning"
    trigger:
      platform: time
      at: "07:00:00"
    action:
      - service: kubernetes.start_workload
        data:
          workload_names:
            - switch.web_app
            - switch.api_server
            - switch.cache_service
          replicas: 3
          namespace: "production"

automation:
  - alias: "Scale down for off-hours"
    trigger:
      platform: time
      at: "18:00:00"
    condition:
      condition: time
      weekday:
        - mon
        - tue
        - wed
        - thu
        - fri
    action:
      - service: kubernetes.scale_workload
        data:
          workload_names:
            - switch.web_frontend
            - switch.api_backend
          replicas: 1
          namespace: "production"

automation:
  - alias: "Scale up on high CPU usage"
    trigger:
      platform: numeric_state
      entity_id: sensor.server_cpu_usage
      above: 80
      for:
        minutes: 5
    action:
      - service: kubernetes.scale_workload
        data:
          workload_name: switch.web_app
          namespace: "production"
          replicas: 5

automation:
  - alias: "Emergency cluster shutdown"
    trigger:
      platform: state
      entity_id: binary_sensor.ups_power_failure
      to: "on"
    action:
      - service: kubernetes.stop_workload
        data:
          workload_names:
            - switch.non_critical_app
            - switch.development_services
          namespace: "default"
      - delay: "00:02:00"
      - service: kubernetes.scale_workload
        data:
          workload_names:
            - switch.critical_app
          replicas: 1
          namespace: "production"

automation:
  - alias: "Scale database StatefulSets for maintenance"
    trigger:
      platform: time
      at: "02:00:00"
    condition:
      condition: time
      weekday:
        - sun
    action:
      - service: kubernetes.scale_workload
        data:
          workload_names:
            - switch.database_primary
            - switch.database_replica
          replicas: 1
          namespace: "database"

automation:
  - alias: "Stop development databases over weekend"
    trigger:
      platform: time
      at: "19:00:00"
    condition:
      condition: time
      weekday:
        - fri
    action:
      - service: kubernetes.stop_workload
        data:
          workload_names:
            - switch.dev_postgres
            - switch.dev_redis
          namespace: "development"

automation:
  - alias: "Alert on node not ready"
    trigger:
      - platform: state
        entity_id:
          - sensor.kubernetes_node_master_1
          - sensor.kubernetes_node_worker_1
          - sensor.kubernetes_node_worker_2
        to: "NotReady"
        for:
          minutes: 2
    action:
      - service: notify.mobile_app_your_phone
        data:
          title: "Kubernetes Node Alert"
          message: >
            Node {{ trigger.entity_id.split('.')[-1].replace('kubernetes_node_', '').replace('_', '-') }}
            is not ready. Status: {{ trigger.to_state.state }}
          data:
            priority: high

automation:
  - alias: "High memory usage alert"
    trigger:
      - platform: template
        value_template: >
          {% set nodes = states.sensor | selectattr('entity_id', 'match', 'sensor.kubernetes_node_.*') | list %}
          {% for node in nodes %}
            {% set memory_used = state_attr(node.entity_id, 'memory_capacity_gb') | float - state_attr(node.entity_id, 'memory_allocatable_gb') | float %}
            {% set memory_total = state_attr(node.entity_id, 'memory_capacity_gb') | float %}
            {% if memory_used / memory_total > 0.9 %}
              true
            {% endif %}
          {% endfor %}
    action:
      - service: notify.persistent_notification
        data:
          title: "High Memory Usage"
          message: >
            One or more nodes are running low on memory.
            Check the node monitoring dashboard for details.

automation:
  - alias: "Log node status changes"
    trigger:
      - platform: state
        entity_id:
          - sensor.kubernetes_node_master_1
          - sensor.kubernetes_node_worker_1
          - sensor.kubernetes_node_worker_2
    condition:
      - condition: template
        value_template: "{{ trigger.from_state.state != trigger.to_state.state }}"
    action:
      - service: logbook.log
        data:
          name: "Kubernetes Node Status"
          message: >
            Node {{ trigger.entity_id.split('.')[-1].replace('kubernetes_node_', '').replace('_', '-') }}
            changed from {{ trigger.from_state.state }} to {{ trigger.to_state.state }}
          entity_id: "{{ trigger.entity_id }}"

views:
  - title: "Kubernetes Cluster"
    path: kubernetes
    cards:
      - type: entities
        title: "Cluster Status"
        entities:
          - entity: sensor.kubernetes_pods_count
            name: "Running Pods"
          - entity: sensor.kubernetes_nodes_count
            name: "Cluster Nodes"
          - entity: sensor.kubernetes_deployments_count
            name: "Active Deployments"
          - entity: sensor.kubernetes_statefulsets_count
            name: "StatefulSets"
          - entity: sensor.kubernetes_cronjobs_count
            name: "CronJobs"
          - entity: binary_sensor.kubernetes_cluster_health
            name: "Cluster Health"

views:
  - title: "Deployment Control"
    path: kubernetes-deployments
    cards:
      - type: entities
        title: "Production Deployments"
        entities:
          - switch.kubernetes_deployment_web_app
          - switch.kubernetes_deployment_api_server
          - switch.kubernetes_deployment_cache_service
      - type: entities
        title: "Development Services"
        entities:
          - switch.kubernetes_deployment_dev_api
          - switch.kubernetes_deployment_test_runner

views:
  - title: "Resource Monitoring"
    path: kubernetes-resources
    cards:
      - type: gauge
        entity: sensor.kubernetes_pods_count
        min: 0
        max: 50
        name: "Pod Count"
        severity:
          green: 0
          yellow: 30
          red: 45
      - type: history-graph
        entities:
          - sensor.kubernetes_pods_count
          - sensor.kubernetes_deployments_count
        hours_to_show: 24

### Node Monitoring Dashboard

```yaml
views:
  - title: "Node Status"
    path: kubernetes-nodes
    cards:
      - type: entities
        title: "Node Overview"
        entities:
          - entity: sensor.kubernetes_node_master_1
            name: "Master Node"
          - entity: sensor.kubernetes_node_worker_1
            name: "Worker Node 1"
          - entity: sensor.kubernetes_node_worker_2
            name: "Worker Node 2"
        show_header_toggle: false
      - type: custom:auto-entities
        card:
          type: entities
          title: "All Cluster Nodes"
        filter:
          include:
            - entity_id: "sensor.kubernetes_node_*"
        sort:
          method: name
      - type: markdown
        content: >
          ## Node Resources

          **{{ states('sensor.kubernetes_node_worker_1') }}** Worker 1:
          - Memory: {{ state_attr('sensor.kubernetes_node_worker_1', 'memory_allocatable_gb') }}GB / {{ state_attr('sensor.kubernetes_node_worker_1', 'memory_capacity_gb') }}GB
          - CPU Cores: {{ state_attr('sensor.kubernetes_node_worker_1', 'cpu_cores') }}
          - Internal IP: {{ state_attr('sensor.kubernetes_node_worker_1', 'internal_ip') }}
          - OS: {{ state_attr('sensor.kubernetes_node_worker_1', 'os_image') }}

script:
  scale_production_environment:
    alias: "Scale Production Environment"
    sequence:
      - service: kubernetes.scale_workload
        data:
          workload_names:
            - switch.web_frontend
            - switch.api_backend
            - switch.worker_queue
          replicas: "{{ replicas | default(3) }}"
          namespace: "production"

script:
  rolling_restart_deployment:
    alias: "Rolling Restart Deployment"
    sequence:
      - service: kubernetes.scale_workload
        data:
          workload_name: "{{ workload_name }}"
          namespace: "{{ namespace }}"
          replicas: 0
      - delay: "00:00:30"
      - service: kubernetes.scale_workload
        data:
          workload_name: "{{ workload_name }}"
          namespace: "{{ namespace }}"
          replicas: "{{ original_replicas | default(1) }}"

automation:
  - alias: "Alert on deployment failure"
    trigger:
      platform: state
      entity_id: switch.kubernetes_deployment_critical_app
      to: "off"
      for:
        minutes: 2
    action:
      - service: notify.mobile_app
        data:
          title: "Deployment Alert"
          message: "Critical app deployment is down!"
          data:
            priority: high

automation:
  - alias: "Cluster health alert"
    trigger:
      platform: state
      entity_id: binary_sensor.kubernetes_cluster_health
      to: "off"
    action:
      - service: notify.slack
        data:
          message: "🚨 Kubernetes cluster is unhealthy!"
          target: "#devops"

automation:
  - alias: "Smart scaling based on day"
    trigger:
      platform: time
      at: "08:00:00"
    action:
      - choose:
          - conditions:
              - condition: time
                weekday:
                  - mon
                  - tue
                  - wed
                  - thu
                  - fri
            sequence:
              - service: kubernetes.scale_workload
                data:
                  workload_name: switch.business_app
                  namespace: "production"
                  replicas: 5
          - conditions:
              - condition: time
                weekday:
                  - sat
                  - sun
            sequence:
              - service: kubernetes.scale_workload
                data:
                  workload_name: switch.business_app
                  namespace: "production"
                  replicas: 2

script:
  promote_to_production:
    alias: "Promote to Production"
    sequence:
      - service: kubernetes.stop_workload
        data:
          workload_name: "{{ workload_name }}"
          namespace: "staging"
      - delay: "00:01:00"
      - service: kubernetes.start_workload
        data:
          workload_name: "{{ workload_name }}"
          namespace: "production"
          replicas: 3