Deploying Containers with Kubernetes

✅ Chapter 2: Creating and Managing Kubernetes Deployments

🔍 Introduction

In Kubernetes, Deployments are one of the most essential workload objects used to manage the lifecycle of Pods. Whether you're deploying a single web service or an entire microservices architecture, Deployments give you tools for declarative, automated, and fault-tolerant application management.

In this chapter, you’ll learn:

• What a Deployment is and how it works
• How to create and update Deployments
• Rolling updates and rollback
• Scaling replicas
• Common commands and YAML patterns

🚀 What is a Deployment in Kubernetes?

A Deployment defines the desired state for a set of Pods and the ReplicaSet managing them. It ensures:

• A specified number of Pods are running at any given time
• New versions are rolled out smoothly
• Old versions are rolled back if necessary

Think of it as the controller that manages ReplicaSets, which in turn manage Pods.

🧱 Anatomy of a Deployment

Let’s examine a basic Deployment definition.

yaml

apiVersion: apps/v1

kind: Deployment

metadata:

  name: my-app

spec:

  replicas: 3

  selector:

    matchLabels:

      app: my-app

  template:

    metadata:

      labels:

        app: my-app

    spec:

      containers:

      - name: web

        image: nginx:1.17

        ports:

        - containerPort: 80

🧾 Key Sections Explained

🛠️ Creating a Deployment

Save your configuration to a file, e.g., nginx-deployment.yaml, then run:

bash

kubectl apply -f nginx-deployment.yaml

This will create:

• A Deployment named my-app
• A ReplicaSet to manage 3 Pods
• 3 Pods running the Nginx container

📋 Verify the Deployment

bash

kubectl get deployments

kubectl get pods

kubectl describe deployment my-app

🔄 Updating a Deployment

Let’s say you want to upgrade from nginx:1.17 to nginx:1.21.

Option 1: Modify and apply YAML again
Option 2: Use the CLI:

bash

kubectl set image deployment/my-app web=nginx:1.21

Kubernetes will:

• Create new Pods with the new image
• Gradually replace old Pods
• Monitor health throughout the process

✅ Rolling Update Strategy

By default, Deployments use a rolling update strategy.

You can control rollout behavior:

yaml

strategy:

  type: RollingUpdate

  rollingUpdate:

    maxSurge: 1

    maxUnavailable: 1

⏪ Rolling Back a Deployment

If something goes wrong:

bash

kubectl rollout undo deployment my-app

To see revision history:

bash

kubectl rollout history deployment my-app

📈 Scaling a Deployment

Scaling increases or decreases the number of Pods.

bash

kubectl scale deployment my-app --replicas=5

In YAML:

yaml

spec:

  replicas: 5

Apply the updated YAML to scale.

🔍 Inspecting and Monitoring Deployments

Check status:

bash

kubectl rollout status deployment my-app

View logs of Pods:

bash

kubectl logs <pod-name>

Execute into container:

bash

kubectl exec -it <pod-name> -- /bin/sh

🧰 Useful Deployment Commands Cheat Sheet

🧪 Advanced Deployment Features

🔹 Probes: Readiness & Liveness

Ensure Pods are only used when healthy:

yaml

livenessProbe:

  httpGet:

    path: /

    port: 80

  initialDelaySeconds: 15

  periodSeconds: 20

• Liveness probe: Checks if app is alive
• Readiness probe: Checks if app is ready to receive traffic

🔹 Resource Requests & Limits

yaml

resources:

  requests:

    memory: "64Mi"

    cpu: "250m"

  limits:

    memory: "128Mi"

    cpu: "500m"

This helps with resource scheduling and prevents container overuse.

🔹 Environment Variables

yaml

env:

  - name: ENVIRONMENT

    value: "production"

🔹 Labels & Selectors

Labels help organize and query resources:

yaml

metadata:

  labels:

    tier: backend

Query with:

bash

kubectl get pods -l tier=backend

🧩 Comparing Deployments to Other Workloads

🧱 YAML Template for Reusability

yaml

apiVersion: apps/v1

kind: Deployment

metadata:

  name: {{ .Values.appName }}

spec:

  replicas: {{ .Values.replicaCount }}

  template:

    spec:

      containers:

        - name: {{ .Values.containerName }}

          image: {{ .Values.image }}

This can be used with Helm for dynamic deployments.

🚀 Summary: What You Learned in Chapter 2

• Kubernetes Deployments manage Pods through ReplicaSets
• You can define desired state via YAML or CLI
• Supports rolling updates, rollback, scaling, and health checks
• Critical for stateless applications
• Highly configurable and production-ready