Docker & Kubernetes Python SDKs

Python's powerful SDKs allow you to automate infrastructure tasks that usually require CLI commands like docker run or kubectl apply.

1. Docker SDK for Python (`docker`)

The docker package allows you to communicate with the Docker daemon.

Basic Usage: List Running Containers

Action:

import docker
 
client = docker.from_env()
for container in client.containers.list():
    print(f"ID: {container.short_id}, Name: {container.name}, Status: {container.status}")

Result:

ID: a1b2c3d4, Name: redis_cache, Status: running
ID: f5e6d7c8, Name: nginx_proxy, Status: running

2. Kubernetes Python Client (`kubernetes`)

The official Kubernetes Python client is used for automating tasks like scaling deployments or auditing pods.

Listing All Pods in a Namespace

Action:

from kubernetes import client, config
 
# Load local kubeconfig
config.load_kube_config()
 
v1 = client.CoreV1Api()
print("Listing pods in 'default' namespace:")
pods = v1.list_namespaced_pod(namespace="default")
for pod in pods.items:
    print(f"Pod: {pod.metadata.name}, IP: {pod.status.pod_ip}")

Result:

Listing pods in 'default' namespace:
Pod: web-server-v1-6789fb8c-x2jkl, IP: 10.244.1.42
Pod: database-master-0, IP: 10.244.1.12

3. DevOps Automation: Health Checks

A common use case is writing a script that checks if a Kubernetes Deployment has the correct number of ready replicas and restarts it if necessary.

Snippet:

apps_v1 = client.AppsV1Api()
deploy = apps_v1.read_namespaced_deployment(name="my-app", namespace="prod")
 
if deploy.status.ready_replicas < deploy.spec.replicas:
    print("ALERT: Not all replicas are ready!")
    # Notification or auto-remediation logic...

4. Comparison: CLI vs. SDK

Feature	CLI (bash)	SDK (python)
Speed	Fast for single tasks	Better for complex logic
Parsing	Requires `jq` / `awk`	Native objects/JSON
Error Handling	Exit codes	Try/Except blocks
Integration	Difficult to test	Easy unit testing

Summary

Use docker-py for managing local or remote containers.
Use kubernetes-python-client for sophisticated orchestration.
Prefer SDKs when your logic involves multi-step decision-making.

1. Docker SDK for Python (docker)