Ansible Kubernetes Integration
Introduction
In today's cloud-native landscape, Kubernetes has become the de facto standard for container orchestration. However, managing Kubernetes clusters and deployments can be complex and time-consuming. This is where Ansible comes in - a powerful automation tool that can simplify Kubernetes management.
Ansible's Kubernetes integration allows you to automate the deployment, scaling, and management of containerized applications. By combining Ansible's simplicity with Kubernetes' powerful orchestration capabilities, you can create reproducible, version-controlled infrastructure as code.
In this tutorial, we'll explore how to use Ansible to automate various Kubernetes tasks, from cluster setup to application deployment and management.
Prerequisites
Before we begin, make sure you have:
- Basic knowledge of Ansible concepts (playbooks, modules, inventory)
- Basic understanding of Kubernetes concepts (pods, deployments, services)
- Ansible (version 2.9+) installed on your control node
- Access to a Kubernetes cluster or minikube for local testing
kubectlconfigured to communicate with your cluster
Ansible Kubernetes Modules
Ansible offers several modules specifically designed for Kubernetes integration:
k8s: The main module for managing Kubernetes objectsk8s_info: For gathering information about Kubernetes resourcesk8s_service: For managing Kubernetes service objectsk8s_scale: For scaling Kubernetes deployments, replica sets, etc.
Let's start by installing the required Python dependencies:
pip install kubernetes openshift
Basic Kubernetes Resource Management
Creating a Deployment with Ansible
Here's a simple playbook to create a Kubernetes deployment:
---
- name: Create Kubernetes Deployment
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Create a Deployment
k8s:
state: present
definition:
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
namespace: default
spec:
replicas: 3
selector:
matchLabels:
app: nginx
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.21
ports:
- containerPort: 80
Save this as create_deployment.yml and run it with:
ansible-playbook create_deployment.yml
Output:
PLAY [Create Kubernetes Deployment] ***************************
TASK [Create a Deployment] ***********************************
changed: [localhost]
PLAY RECAP **************************************************
localhost : ok=1 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
After running the playbook, verify the deployment with:
kubectl get deployments
You should see your nginx-deployment with 3 pods running.
Using Resource Files with Ansible
Instead of defining resources directly in the playbook, you can use separate YAML files:
---
- name: Apply Kubernetes resources from files
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Apply a deployment from a file
k8s:
state: present
src: deployment.yml
Where deployment.yml would contain:
apiVersion: apps/v1
kind: Deployment
metadata:
name: web-app
namespace: default
spec:
replicas: 2
selector:
matchLabels:
app: web-app
template:
metadata:
labels:
app: web-app
spec:
containers:
- name: web-app
image: nginx:latest
ports:
- containerPort: 80
Gathering Information About Kubernetes Resources
The k8s_info module allows you to gather information about existing resources:
---
- name: Get Kubernetes resource information
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Get info about all pods in a namespace
k8s_info:
kind: Pod
namespace: default
register: pod_list
- name: Display pod names
debug:
msg: "Pod: {{ item.metadata.name }}"
loop: "{{ pod_list.resources }}"
Output example:
TASK [Display pod names] *************************************
ok: [localhost] => (item={'metadata': {'name': 'nginx-deployment-7fb96c846b-2z9xj', ...}}) => {
"msg": "Pod: nginx-deployment-7fb96c846b-2z9xj"
}
ok: [localhost] => (item={'metadata': {'name': 'nginx-deployment-7fb96c846b-d8fjk', ...}}) => {
"msg": "Pod: nginx-deployment-7fb96c846b-d8fjk"
}
ok: [localhost] => (item={'metadata': {'name': 'nginx-deployment-7fb96c846b-q2xvm', ...}}) => {
"msg": "Pod: nginx-deployment-7fb96c846b-q2xvm"
}
Dynamic Inventory for Kubernetes
Ansible can use Kubernetes as a dynamic inventory source, allowing you to target pods and nodes directly in your playbooks.
Create a file named k8s_inventory.yml:
plugin: kubernetes.core.k8s
connections:
- name: default
kubeconfig: ~/.kube/config
Test the inventory with:
ansible-inventory -i k8s_inventory.yml --list
You'll see JSON output showing all your Kubernetes resources as potential Ansible targets.
Real-world Example: Complete Application Deployment
Let's put everything together by creating a playbook that deploys a complete application with multiple Kubernetes resources:
---
- name: Deploy Complete Application
hosts: localhost
connection: local
gather_facts: no
vars:
app_name: flask-demo
app_namespace: demo
app_replicas: 2
app_image: "flask-demo:latest"
tasks:
- name: Ensure namespace exists
k8s:
state: present
definition:
apiVersion: v1
kind: Namespace
metadata:
name: "{{ app_namespace }}"
- name: Deploy ConfigMap with application settings
k8s:
state: present
definition:
apiVersion: v1
kind: ConfigMap
metadata:
name: "{{ app_name }}-config"
namespace: "{{ app_namespace }}"
data:
APP_SETTINGS: "Production"
LOG_LEVEL: "INFO"
- name: Deploy application
k8s:
state: present
definition:
apiVersion: apps/v1
kind: Deployment
metadata:
name: "{{ app_name }}"
namespace: "{{ app_namespace }}"
spec:
replicas: "{{ app_replicas }}"
selector:
matchLabels:
app: "{{ app_name }}"
template:
metadata:
labels:
app: "{{ app_name }}"
spec:
containers:
- name: "{{ app_name }}"
image: "{{ app_image }}"
ports:
- containerPort: 5000
envFrom:
- configMapRef:
name: "{{ app_name }}-config"
resources:
limits:
memory: "256Mi"
cpu: "500m"
requests:
memory: "128Mi"
cpu: "100m"
- name: Expose the application with a service
k8s:
state: present
definition:
apiVersion: v1
kind: Service
metadata:
name: "{{ app_name }}"
namespace: "{{ app_namespace }}"
spec:
selector:
app: "{{ app_name }}"
ports:
- port: 80
targetPort: 5000
type: ClusterIP
Visualizing the Workflow
Here's a diagram showing the Ansible Kubernetes integration workflow:
Managing Kubernetes Clusters with Ansible
Beyond just managing resources within a cluster, Ansible can also help set up and manage the Kubernetes cluster itself:
Setting up a Kubernetes Cluster
For a production environment, consider using Ansible to set up Kubernetes clusters:
---
- name: Install Kubernetes dependencies
hosts: kube_nodes
become: true
tasks:
- name: Install required packages
apt:
name:
- apt-transport-https
- ca-certificates
- curl
- gnupg-agent
- software-properties-common
state: present
update_cache: yes
when: ansible_os_family == "Debian"
- name: Add Kubernetes apt key
apt_key:
url: https://packages.cloud.google.com/apt/doc/apt-key.gpg
state: present
when: ansible_os_family == "Debian"
- name: Add Kubernetes repository
apt_repository:
repo: deb https://apt.kubernetes.io/ kubernetes-xenial main
state: present
when: ansible_os_family == "Debian"
- name: Install Kubernetes packages
apt:
name:
- kubelet
- kubeadm
- kubectl
state: present
update_cache: yes
when: ansible_os_family == "Debian"
- name: Disable swap
command: swapoff -a
when: ansible_swaptotal_mb > 0
Advanced Features
Using Templates for Dynamic Configurations
Use Ansible templates to generate Kubernetes manifests dynamically:
---
- name: Deploy application with dynamic config
hosts: localhost
connection: local
gather_facts: no
vars:
app_name: web-app
app_image: nginx
app_version: 1.21
app_replicas: 3
app_port: 80
tasks:
- name: Create deployment from template
k8s:
state: present
template: 'templates/deployment-template.yml.j2'
And the template file templates/deployment-template.yml.j2:
apiVersion: apps/v1
kind: Deployment
metadata:
name: {{ app_name }}
namespace: default
spec:
replicas: {{ app_replicas }}
selector:
matchLabels:
app: {{ app_name }}
template:
metadata:
labels:
app: {{ app_name }}
spec:
containers:
- name: {{ app_name }}
image: {{ app_image }}:{{ app_version }}
ports:
- containerPort: {{ app_port }}
Scaling Applications
Scale deployments easily:
---
- name: Scale application
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Scale deployment
k8s_scale:
api_version: apps/v1
kind: Deployment
name: nginx-deployment
namespace: default
replicas: 5
Best Practices for Ansible Kubernetes Integration
-
Version Control: Keep all your Kubernetes manifests and Ansible playbooks in a version control system like Git.
-
Modularize Playbooks: Break down complex deployments into smaller, reusable roles.
-
Use Templates: Leverage Ansible's templating system for dynamic configuration.
-
Idempotency: Ensure your playbooks can be run multiple times without causing issues.
-
Environment Separation: Use variables to differentiate between development, staging, and production environments.
-
Secret Management: Use Ansible Vault or external secret management tools for sensitive data.
-
Testing: Test your playbooks in a development environment before applying them to production.
Troubleshooting Common Issues
Connection Issues
If you encounter connection problems:
---
- name: Debug Kubernetes connection
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Get cluster info
k8s_info:
kind: Node
register: node_info
ignore_errors: yes
- name: Display error
debug:
var: node_info
when: node_info is failed
Permission Issues
For permission problems, ensure your kubeconfig has the necessary permissions:
---
- name: Check permissions
hosts: localhost
connection: local
gather_facts: no
tasks:
- name: Try to list namespaces
k8s_info:
kind: Namespace
register: ns_info
ignore_errors: yes
- name: Display permissions error
debug:
var: ns_info
when: ns_info is failed
Summary
Ansible's Kubernetes integration provides a powerful way to automate the deployment and management of containerized applications. By combining Ansible's simplicity with Kubernetes' orchestration capabilities, you can create reproducible, version-controlled infrastructure as code.
Key benefits include:
- Simplified automation of Kubernetes deployments
- Consistent environment management
- Version-controlled infrastructure
- Reduced manual intervention and human error
- Seamless integration with existing Ansible workflows
Additional Resources
- Practice by creating a multi-tier application deployment using Ansible and Kubernetes
- Explore the Ansible Kubernetes Collection documentation
- Learn more about Kubernetes patterns
Exercises
- Create a playbook that deploys a stateful application with persistent storage in Kubernetes.
- Write a playbook that performs a blue-green deployment of a web application.
- Create an Ansible role that sets up monitoring for your Kubernetes cluster using Prometheus and Grafana.
- Develop a playbook that automates cluster maintenance tasks like upgrades and backups.
If you spot any mistakes on this website, please let me know at [email protected]. I’d greatly appreciate your feedback! :)