Startup probe concept in Kubernetes - Mini Project: Build & Apply

Startup Probe Concept in Kubernetes

📖 Scenario: You are deploying a web application in Kubernetes. Sometimes the app takes time to start before it can respond to health checks. To avoid Kubernetes killing the pod too early, you will use a startupProbe.

🎯 Goal: You will create a Kubernetes pod manifest with a startupProbe that checks the app's readiness by sending an HTTP GET request to /healthz on port 8080. You will configure the probe to wait longer before failing, allowing the app to start properly.

📋 What You'll Learn

Create a pod manifest named web-app with a container named web-container running image nginx.

Add a startupProbe that sends an HTTP GET request to /healthz on port 8080.

Set failureThreshold to 10 and periodSeconds to 5 in the startupProbe.

Print the final pod manifest YAML.

💡 Why This Matters

🌍 Real World

Startup probes are used in Kubernetes to handle applications that take time to start. This prevents Kubernetes from killing pods too early during startup.

💼 Career

Understanding startup probes is important for DevOps roles managing Kubernetes clusters to ensure application stability and proper health monitoring.

1

Create the basic pod manifest

Create a Kubernetes pod manifest with apiVersion set to v1, kind set to Pod, and metadata.name set to web-app. Add a single container named web-container using the image nginx.

Kubernetes

# Write the pod manifest YAML with apiVersion, kind, metadata, and container name and image

Need a hint?

Start with the basic pod structure and add one container with the specified name and image.

2

Add the startupProbe configuration

Add a startupProbe section under the container web-container. Configure it to send an HTTP GET request to path /healthz on port 8080.

Kubernetes

apiVersion: v1
kind: Pod
metadata:
  name: web-app
spec:
  containers:
  - name: web-container
    image: nginx
    # Add startupProbe with httpGet path /healthz and port 8080

Need a hint?

Use startupProbe with httpGet specifying the path and port.

3

Configure failureThreshold and periodSeconds

Inside the startupProbe, add failureThreshold set to 10 and periodSeconds set to 5 to control how long Kubernetes waits before marking the pod as failed.

Kubernetes

apiVersion: v1
kind: Pod
metadata:
  name: web-app
spec:
  containers:
  - name: web-container
    image: nginx
    startupProbe:
      httpGet:
        path: /healthz
        port: 8080
      # Add failureThreshold: 10 and periodSeconds: 5 here

Need a hint?

These settings tell Kubernetes to try the probe 10 times every 5 seconds before failing.

4

Print the final pod manifest

Print the complete pod manifest YAML with the startupProbe configured as specified.

Kubernetes

# Print the full pod manifest YAML

Need a hint?

Make sure the entire YAML is printed exactly as configured.