Skip to content

🛠️ Controllers in Kubernetes

Kubernetes controllers manage the lifecycle and availability of Pods. They ensure that the desired state (as defined by YAML or commands) is continuously reconciled with the current state in the cluster.

✅ Why Use Controllers?

While a Pod can be created manually, it won't automatically recover if it fails. Controllers automate this, ensuring your workloads are:

  • Resilient (auto-restarted)
  • Scalable (adjust replicas)
  • Declarative (defined via YAML)

📦 ReplicaSet

A ReplicaSet ensures that a specified number of identical Pods are running at all times.

Example YAML

apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: myapp-replicaset
spec:
  replicas: 3
  selector:
    matchLabels:
      app: myapp
  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - name: myapp
        image: myimage

Commands

kubectl apply -f replicaset.yaml       # Create ReplicaSet
kubectl get rs                         # List ReplicaSets
kubectl describe rs <name>             # View details
kubectl edit rs <name>                 # Edit live
kubectl delete rs <name>               # Delete

💡 Important Notes

  • Ensure selector.matchLabels matches template.metadata.labels
  • ReplicaSets are rarely used directly; Deployments manage them

🚀 Deployment

A Deployment is the most commonly used controller. It manages ReplicaSets and provides features like:

  • Rolling updates
  • Rollbacks
  • Declarative scaling

Create via Command

kubectl create deployment nginx --image=nginx
kubectl create deployment nginx --image=nginx --replicas=3

Generate YAML

kubectl create deployment nginx --image=nginx --dry-run=client -o yaml > nginx-deploy.yaml

Apply YAML

kubectl apply -f nginx-deploy.yaml

Inspect & Manage

kubectl get deployments
kubectl describe deployment <name>
kubectl edit deployment <name>
kubectl delete deployment <name>

Handle Rollouts

kubectl rollout status deployment <name>
kubectl rollout undo deployment <name>

Common Issue Example

# Image error (e.g., typo)
kubectl describe pod <name>   # Check Events section for ImagePullBackOff

🧠 Other Controllers

StatefulSet

  • Manages stateful applications
  • Maintains unique identity per Pod (e.g., persistent volume)
  • Example: databases, Kafka, etc.

DaemonSet

  • Ensures a Pod runs on every node
  • Common use: log collectors, monitoring agents

Job & CronJob

  • Job: Run once and exit successfully (e.g., DB migration)
  • CronJob: Schedule recurring Jobs (e.g., backup every day)

✅ Summary

Controller Purpose
ReplicaSet Maintain a stable set of identical Pods
Deployment Manage ReplicaSets with updates/rollbacks
StatefulSet Ordered, identity-aware Pods (stateful apps)
DaemonSet One Pod per node
Job One-time task execution
CronJob Scheduled job execution

Controllers bring automation, scalability, and resilience to your Kubernetes workloads. Use Deployments as your default choice unless you have specific needs (e.g., persistent state).