Probe failure and container restart behavior in Kubernetes - Mini Project: Build & Apply

Probe failure and container restart behavior

📖 Scenario: You are managing a Kubernetes cluster running a simple web application. You want to understand how Kubernetes handles container health checks and restarts when probes fail.

🎯 Goal: Learn how to configure liveness and readiness probes in a Kubernetes Pod manifest and observe container restart behavior when probes fail.

📋 What You'll Learn

Create a Pod manifest with a container running the nginx image

Add a liveness probe that checks the HTTP endpoint /healthz on port 80

Add a readiness probe that checks the HTTP endpoint /ready on port 80

Simulate probe failure by using incorrect probe paths

Observe and print the Pod status to see container restart behavior

💡 Why This Matters

🌍 Real World

Kubernetes uses probes to check if containers are healthy and ready to serve traffic. Understanding probe failures helps keep applications stable.

💼 Career

DevOps engineers and site reliability engineers configure probes to automate container health checks and manage restarts, improving uptime.

1

Create a basic Pod manifest with nginx container

Create a Kubernetes Pod manifest named nginx-pod.yaml with a single container named nginx using the image nginx:latest. The container should expose port 80.

Kubernetes

# Write the Pod manifest YAML with apiVersion, kind, metadata, and spec including the nginx container

Need a hint?

Use apiVersion: v1 and kind: Pod. Define metadata.name as nginx-pod. Under spec.containers, add one container named nginx with image nginx:latest and expose port 80.

2

Add liveness and readiness probes with incorrect paths

Add a livenessProbe and a readinessProbe to the nginx container. Both probes should use httpGet on port 80. Set the livenessProbe path to /healthz and the readinessProbe path to /ready. Use default probe settings for initial delay and period.

Kubernetes

apiVersion: v1
kind: Pod
metadata:
  name: nginx-pod
spec:
  containers:
  - name: nginx
    image: nginx:latest
    ports:
    - containerPort: 80
    # Add livenessProbe and readinessProbe below

Need a hint?

Under the container spec, add livenessProbe and readinessProbe sections. Use httpGet with path and port fields. Use paths /healthz and /ready respectively.

3

Apply the Pod manifest and observe container restart behavior

Apply the Pod manifest using kubectl apply -f nginx-pod.yaml. Then use kubectl get pods nginx-pod -o jsonpath='{.status.containerStatuses[0].restartCount}' to check the container restart count. Write a command to get the restart count.

Kubernetes

# Apply the Pod manifest and write the command to get the restart count

Need a hint?

Use kubectl apply -f nginx-pod.yaml to create the Pod. Then use kubectl get pods nginx-pod -o jsonpath='{.status.containerStatuses[0].restartCount}' to see how many times the container restarted.

4

Fix probe paths and verify no restarts

Edit the Pod manifest to change the livenessProbe and readinessProbe paths to / (root path). Apply the updated manifest with kubectl apply -f nginx-pod.yaml. Then print the restart count again using kubectl get pods nginx-pod -o jsonpath='{.status.containerStatuses[0].restartCount}'. Write the command to print the restart count.

Kubernetes

# Change probe paths to '/' and apply the manifest
# Then print the restart count again

Need a hint?

Change both probe paths to / which is a valid path for nginx. Apply the manifest again and check the restart count. It should not increase if probes succeed.